1 00:00:01,040 --> 00:00:04,960 You're about to join Niels Kostrup Larson on a raw and honest 2 00:00:04,960 --> 00:00:08,280 journey into the world of systematic investing and learn about 3 00:00:08,280 --> 00:00:11,520 the most dependable and consistent, yet often overlooked 4 00:00:11,520 --> 00:00:12,720 investment strategy. 5 00:00:13,200 --> 00:00:16,000 Welcome to the Systematic Investor Series. 6 00:00:23,200 --> 00:00:25,760 Welcome or welcome back to this week's edition of the Systematic 7 00:00:25,760 --> 00:00:28,840 Investor series with Rob Carver and I, Nils Kasterblassen, 8 00:00:28,840 --> 00:00:31,920 where he each week we take the pulse of the global market through 9 00:00:31,920 --> 00:00:34,040 the lens of a rules based investor. 10 00:00:34,280 --> 00:00:37,520 Rob, it is wonderful to be back with you this week. 11 00:00:37,520 --> 00:00:38,880 First time in 2026. 12 00:00:38,880 --> 00:00:39,640 Hope you're doing well. 13 00:00:39,640 --> 00:00:41,560 How are things in the, in the UK? 14 00:00:42,280 --> 00:00:46,560 Yeah, it doesn't seem, I mean 2026 is only a few weeks old, but 15 00:00:46,560 --> 00:00:49,880 it seems like a hell of a lot's happened, so particularly in 16 00:00:49,880 --> 00:00:50,280 the market. 17 00:00:50,440 --> 00:00:52,760 So, yeah, it's all quite exciting. 18 00:00:52,920 --> 00:00:54,560 It is all very exciting. 19 00:00:54,560 --> 00:00:57,080 And actually speaking of exciting, we have a great lineup 20 00:00:57,080 --> 00:01:01,080 of topics today which I think people will really en. 21 00:01:01,780 --> 00:01:05,140 We're going to be tackling a couple of new papers, an article 22 00:01:05,460 --> 00:01:07,700 and a couple of questions that came in. 23 00:01:07,700 --> 00:01:10,660 So this is all super exciting. 24 00:01:11,300 --> 00:01:15,860 Rob, as you know, I'm always curious to hear what's kind of been 25 00:01:16,100 --> 00:01:19,580 on your radar since we last spoke, but not from the topics we're 26 00:01:19,580 --> 00:01:20,340 going to be talking about. 27 00:01:20,340 --> 00:01:24,260 But if there's something else that you found interesting, then 28 00:01:25,620 --> 00:01:26,340 let us know. 29 00:01:26,740 --> 00:01:30,100 Yeah, well, I was hoping by now I would have seen the new Melania 30 00:01:30,100 --> 00:01:33,730 film, but unfortunately I've not had that opportunity. 31 00:01:33,810 --> 00:01:36,890 And it looks actually looking at the viewing figures worldwide, 32 00:01:36,890 --> 00:01:38,690 not many other people have as well, so. 33 00:01:39,090 --> 00:01:42,050 So hopefully I'll get a chance to review that next time I'm on. 34 00:01:42,610 --> 00:01:46,770 So I saw an interesting paper published by Alliance Bernstein. 35 00:01:46,850 --> 00:01:51,210 I don't want to talk about it today, but what made me sort of impressed 36 00:01:51,210 --> 00:01:52,450 was the title of this paper. 37 00:01:53,330 --> 00:01:56,090 And I actually, for reasons that will become obvious, I actually 38 00:01:56,090 --> 00:01:59,760 had to ask my son who did German at school how to pronounce 39 00:01:59,840 --> 00:02:01,200 this word in this title. 40 00:02:01,200 --> 00:02:05,040 So the title of the paper is about tpa Total Portfolio Approach, 41 00:02:05,040 --> 00:02:06,320 which we've discussed before. 42 00:02:06,960 --> 00:02:12,080 Title of paper is Portfolio Design as Gesamtkunstwerk. 43 00:02:12,640 --> 00:02:14,240 The Total Portfolio Approach. 44 00:02:15,200 --> 00:02:22,520 And apparently Gesamtkuntwerk is something like a total piece of 45 00:02:22,520 --> 00:02:22,760 art. 46 00:02:22,760 --> 00:02:26,270 And it's apparently it's a term coined by Wagner in relation 47 00:02:26,270 --> 00:02:27,110 to his operas. 48 00:02:27,110 --> 00:02:29,710 So yeah, that intrigued me. 49 00:02:29,710 --> 00:02:33,030 But it goes to show if you're working as a sales side analyst, 50 00:02:33,030 --> 00:02:35,030 you desperately got to try and make your stuff interesting. 51 00:02:35,030 --> 00:02:35,230 Right. 52 00:02:35,230 --> 00:02:38,350 And one way of doing that is coming up with a title that makes 53 00:02:38,350 --> 00:02:40,070 people stop and Think and go. 54 00:02:40,550 --> 00:02:41,670 Well, that sounds weird. 55 00:02:41,750 --> 00:02:42,710 At the very least. 56 00:02:43,030 --> 00:02:46,110 And it worked because to be honest, I probably wouldn't even 57 00:02:46,110 --> 00:02:48,790 have glanced at that paper were it not for the funky titles. 58 00:02:49,430 --> 00:02:53,010 So, yeah, if you're interested in TPA and Wagner, then find that 59 00:02:53,010 --> 00:02:54,890 Alliance Bernstein paper and enjoy it. 60 00:02:55,530 --> 00:02:56,450 That is quite funny. 61 00:02:56,450 --> 00:02:59,850 I mean, since I have to come up with a lot of titles and headlines 62 00:02:59,850 --> 00:03:03,530 every week for the podcast episodes and the emails we send out, 63 00:03:03,930 --> 00:03:09,450 I'm always interested in a catchy method of getting people to 64 00:03:10,730 --> 00:03:13,450 listen or open what we share. 65 00:03:14,330 --> 00:03:18,250 I have to say, I didn't expect a German word to do the trick, but 66 00:03:19,430 --> 00:03:20,550 I might have to try it out. 67 00:03:20,550 --> 00:03:21,550 Well, there we go, Niels. 68 00:03:21,550 --> 00:03:22,070 There we go. 69 00:03:22,470 --> 00:03:25,870 All those long compound German words just waiting to be used in 70 00:03:25,870 --> 00:03:27,510 your email subject lines. 71 00:03:27,670 --> 00:03:28,310 Clearly. 72 00:03:28,390 --> 00:03:29,030 Clearly. 73 00:03:29,270 --> 00:03:32,870 Okay, well, let me tell you what caught my interest. 74 00:03:32,870 --> 00:03:35,790 And I know this first topic definitely caught your interest as 75 00:03:35,790 --> 00:03:36,070 well. 76 00:03:36,310 --> 00:03:39,030 And instead of putting it in the trend following section, I think 77 00:03:39,110 --> 00:03:43,110 we'll discuss it now, maybe spend a little bit more time on it. 78 00:03:43,590 --> 00:03:47,830 And it is what happened in the last week or so? 79 00:03:47,830 --> 00:03:50,620 Well, actually the last few months. 80 00:03:50,620 --> 00:03:56,060 But it kind of really came to a head on the last day of January. 81 00:03:56,620 --> 00:03:57,340 Freaky Friday. 82 00:03:58,220 --> 00:04:00,380 I've decided I'm going to christen it Freaky Friday. 83 00:04:00,780 --> 00:04:01,580 Freaky Friday. 84 00:04:01,660 --> 00:04:04,540 Especially if you're trading precious metals and especially metals 85 00:04:04,540 --> 00:04:06,780 and especially if you're trading silver. 86 00:04:07,260 --> 00:04:12,180 And actually, I would say this week has been pretty interesting 87 00:04:12,180 --> 00:04:12,780 as well. 88 00:04:13,500 --> 00:04:18,570 So why don't you tell us a little bit about your silver experience 89 00:04:19,130 --> 00:04:23,530 and maybe we can then talk a little bit about kind of the good 90 00:04:23,530 --> 00:04:28,090 old days and why silver is not gold in some way. 91 00:04:28,970 --> 00:04:32,730 Yeah, I mean, so I obviously trade silver. 92 00:04:33,050 --> 00:04:35,610 It's a futures contract, it's liquid. 93 00:04:36,090 --> 00:04:37,850 So it would be on the table for me to trade. 94 00:04:37,850 --> 00:04:41,050 Now, I've talked about this before, but the way I trade futures 95 00:04:41,050 --> 00:04:41,850 is a little bit complicated. 96 00:04:41,850 --> 00:04:43,850 And it's mainly to do with the fact that I have a relatively small 97 00:04:43,850 --> 00:04:48,430 portfolio, so I have a notional exposure to 350 futures 98 00:04:48,430 --> 00:04:49,030 contracts. 99 00:04:50,950 --> 00:04:53,230 Many of some of those I couldn't trade anyway because of 100 00:04:53,230 --> 00:04:54,150 regulatory restrictions. 101 00:04:54,150 --> 00:04:55,910 Of course they're not liquid or they're too expensive. 102 00:04:55,990 --> 00:04:57,670 Silver does not fall into that category. 103 00:04:59,190 --> 00:05:02,950 And then I basically dynamically optimize my portfolio 104 00:05:02,950 --> 00:05:06,750 every day to match to get the best possible match I could do to 105 00:05:06,750 --> 00:05:10,150 a sort of abstract portfolio, which I could have if I had a notional 106 00:05:10,150 --> 00:05:12,150 value of running into the hundreds of millions of dollars, 107 00:05:12,150 --> 00:05:13,350 which sadly I don't have. 108 00:05:13,940 --> 00:05:18,140 So that will mean that often that I won't necessarily have a position 109 00:05:18,140 --> 00:05:21,100 in a given instrument, even if I have quite a strong forecast on 110 00:05:21,100 --> 00:05:21,380 it. 111 00:05:22,580 --> 00:05:25,500 Now, Silver, though, I do have a position, and I did have a position 112 00:05:25,500 --> 00:05:30,300 in, I should say, and if I look at my history of exposure to 113 00:05:30,300 --> 00:05:32,100 that position, it's quite interesting. 114 00:05:33,620 --> 00:05:37,180 Now, again, it's slightly complicated by the fact that I have 115 00:05:37,180 --> 00:05:41,480 this dynamic optimization, but broadly speaking, you can sort of 116 00:05:41,480 --> 00:05:46,280 say that CTAs generally will have positions that are a function 117 00:05:46,280 --> 00:05:50,760 of two things, the strength of the trend and the volatility of the 118 00:05:50,760 --> 00:05:51,640 instrument you're trading. 119 00:05:53,640 --> 00:05:58,280 Now, that's obviously a broad brush thing because of course, some 120 00:05:58,280 --> 00:06:02,640 CTAs will take basically binary positions, so they'll, you 121 00:06:02,640 --> 00:06:05,560 know, they'll kind of go from fully short to fully long. 122 00:06:06,650 --> 00:06:10,330 Others will sort of do what I do, which is sort of continuous adjusting 123 00:06:10,330 --> 00:06:10,810 forecasts. 124 00:06:10,810 --> 00:06:13,090 So getting, you know, the moment the trend starts to look positive, 125 00:06:13,090 --> 00:06:16,010 they'll kind of increase their forecast and that would increase 126 00:06:16,010 --> 00:06:18,130 their position, all other things being equal. 127 00:06:18,130 --> 00:06:19,890 And of course they're not, which is what we'll get into in a 128 00:06:19,890 --> 00:06:20,170 second. 129 00:06:21,850 --> 00:06:26,210 And also, of course, some CTAs take volatility into account when 130 00:06:26,210 --> 00:06:29,490 they initially buy a position, but then don't make any further adjustments 131 00:06:29,490 --> 00:06:31,370 to that position as volatility changes. 132 00:06:33,020 --> 00:06:35,500 So we can characterize the silver story over the last few months 133 00:06:35,500 --> 00:06:39,500 as a positive, bullish signal getting stronger. 134 00:06:40,460 --> 00:06:43,820 This is all up till last Thursday, obviously, because obviously 135 00:06:43,820 --> 00:06:48,020 on Freaky Friday, things change very dramatically, but at 136 00:06:48,020 --> 00:06:50,300 the same time the volatility also increasing. 137 00:06:51,900 --> 00:06:54,660 So my position, broadly speaking, and the position of anyone 138 00:06:54,660 --> 00:06:58,710 trading like me would be a function of those kind of two effects. 139 00:06:59,030 --> 00:07:02,110 And if I could actually look at my position of my exposure to 140 00:07:02,110 --> 00:07:06,430 silver, I can see that although I was trading it on and 141 00:07:06,430 --> 00:07:12,390 off last year, the kind of current position, I bought some on 142 00:07:12,390 --> 00:07:17,630 27 November, and I bought some more on 3 December, and that would 143 00:07:17,630 --> 00:07:20,070 have been down to the forecast increasing. 144 00:07:21,510 --> 00:07:25,750 And then actually on the 31st of December, I sold a contract, and 145 00:07:25,750 --> 00:07:29,030 that would have been inevitably because the volatility 146 00:07:29,030 --> 00:07:29,870 had increased. 147 00:07:30,030 --> 00:07:32,830 So the combination of those two effects often leads to these 148 00:07:32,830 --> 00:07:33,790 interesting effects. 149 00:07:35,150 --> 00:07:39,750 And then subsequently on, well, actually not on Freaky Friday 150 00:07:39,750 --> 00:07:41,429 itself because I have a daily rebalance. 151 00:07:41,429 --> 00:07:45,710 I didn't do any trading in silver during Friday itself because 152 00:07:45,710 --> 00:07:48,630 my position was based on Thursday's close, which was kind 153 00:07:48,630 --> 00:07:49,630 of normal, if you like. 154 00:07:50,750 --> 00:07:54,360 But so on Monday, so that's a couple of days ago now, I actually 155 00:07:54,360 --> 00:07:58,080 then sold that contract, and that would have been, again, because 156 00:07:58,080 --> 00:08:04,680 volatility had increased my trend signal is still long because, 157 00:08:06,040 --> 00:08:09,800 you know, unless you're trading an insanely quick trend following 158 00:08:09,800 --> 00:08:15,000 system, one day of negative returns, even a day like Friday where 159 00:08:15,000 --> 00:08:18,360 silver dropped by the most in one day since. 160 00:08:19,370 --> 00:08:23,610 And you know, you'll remember this, Niels, of course, since about 161 00:08:23,930 --> 00:08:26,890 it was decades, at least decades ago, many decades ago, I 162 00:08:26,890 --> 00:08:30,490 think 40, 45, 46 years, 47 years ago to be exact. 163 00:08:31,610 --> 00:08:34,810 So that's why I then sold, sold out of that silver position 164 00:08:34,810 --> 00:08:35,529 on that day. 165 00:08:35,610 --> 00:08:37,010 So that's kind of interesting. 166 00:08:37,010 --> 00:08:40,290 Now a big question is, did I, you know, useful question to ask 167 00:08:40,290 --> 00:08:41,850 is, well, did I actually make money doing this? 168 00:08:41,850 --> 00:08:43,130 And the answer is yes, I did. 169 00:08:43,130 --> 00:08:46,410 Because although silver dropped precipitously on Friday, 170 00:08:47,030 --> 00:08:49,270 it only dropped back to the level where it had been a few weeks 171 00:08:49,270 --> 00:08:51,590 beforehand, a couple of weeks beforehand. 172 00:08:52,470 --> 00:08:54,150 So I can actually give you hard numbers. 173 00:08:54,150 --> 00:09:00,710 So I bought into my silver position at levels of 53 and $59 174 00:09:01,030 --> 00:09:02,310 per ounce respectively. 175 00:09:03,590 --> 00:09:09,390 And then I sold at the end of December at $71 and I sold again 176 00:09:09,390 --> 00:09:10,150 at $81. 177 00:09:10,710 --> 00:09:13,810 So just looking at those trades, you know, that was nice. 178 00:09:13,810 --> 00:09:15,050 That's a profitable trade. 179 00:09:15,610 --> 00:09:19,170 Of course, what that misses out is that in between the end of 180 00:09:19,170 --> 00:09:21,890 December and the Freaky Friday, the price briefly went up 181 00:09:21,890 --> 00:09:25,690 to, you know, 110, $120 an ounce, I think it was. 182 00:09:27,050 --> 00:09:30,450 So that's kind of my kind of story with silver. 183 00:09:30,450 --> 00:09:34,890 Now obviously that would be different for someone who doesn't 184 00:09:34,890 --> 00:09:37,650 do the continuous volatility adjustment because what they would 185 00:09:37,650 --> 00:09:40,810 have done is seen the value of the silver in their portfolio kind 186 00:09:40,810 --> 00:09:44,640 of get bigger and bigger and bigger and then not done anything 187 00:09:44,640 --> 00:09:48,360 about it and just hung onto that and as a result would have basically 188 00:09:48,360 --> 00:09:49,960 would have seen a massive run up in value. 189 00:09:50,440 --> 00:09:54,040 And then Freaky Friday would have resulted in a quite substantial 190 00:09:54,040 --> 00:09:54,680 drop in value. 191 00:09:55,560 --> 00:09:57,920 And we've got the sort of monthly returns for January. 192 00:09:57,920 --> 00:10:02,720 Now for many CTAs and certain CTAs and many viewers of this podcast 193 00:10:02,720 --> 00:10:06,520 will know who we're talking about, who take very large leveraged 194 00:10:06,520 --> 00:10:09,600 positions and have very high risk and don't do this volatility 195 00:10:09,600 --> 00:10:10,120 adjustment. 196 00:10:10,120 --> 00:10:14,300 I think at one point probably went from being about sort of 50% 197 00:10:14,300 --> 00:10:16,780 up for the month to being 25% up for the month. 198 00:10:16,780 --> 00:10:20,860 Now, 25% up for the month is still amazing, but the actual one 199 00:10:20,860 --> 00:10:24,340 day loss on, on Freaky Friday would have been, you know, very substantial 200 00:10:24,340 --> 00:10:24,780 indeed. 201 00:10:25,020 --> 00:10:28,340 I think I can help you with the precision, actually. 202 00:10:28,340 --> 00:10:33,300 Think about being up 72% on the Thursday and ending the month 203 00:10:33,300 --> 00:10:34,620 up 27%. 204 00:10:34,700 --> 00:10:35,860 Yeah, no names here. 205 00:10:35,860 --> 00:10:39,370 Of course it is, it was, it was a large move. 206 00:10:39,370 --> 00:10:42,890 I will say, by the way, there is I guess one other way that you 207 00:10:42,890 --> 00:10:43,170 could. 208 00:10:43,170 --> 00:10:47,250 Even though the, you could say largely the trend is still up for 209 00:10:47,250 --> 00:10:48,490 Silver in many respect. 210 00:10:49,050 --> 00:10:52,570 But of course, if you were using a hard stop as one of your 211 00:10:52,570 --> 00:10:56,210 roles, and that stop may be calculated from what has been the 212 00:10:56,210 --> 00:11:01,610 most recent high of some sort, maybe with a vol stop linked to it, 213 00:11:01,770 --> 00:11:04,810 I guess you could have also been stopped out on Freaky Friday 214 00:11:04,970 --> 00:11:06,650 because the move was so big. 215 00:11:06,650 --> 00:11:09,520 But it really depends on your specific rule. 216 00:11:09,600 --> 00:11:10,360 It does, yeah. 217 00:11:10,360 --> 00:11:12,840 And because I don't use stop loss rules. 218 00:11:12,840 --> 00:11:14,080 Yeah, yeah, you're absolutely right. 219 00:11:14,080 --> 00:11:15,200 Any, any stop loss. 220 00:11:15,360 --> 00:11:20,040 I mean, I say that because it depends again on how what you do 221 00:11:20,040 --> 00:11:22,120 with your stop loss rules as your position runs up. 222 00:11:22,120 --> 00:11:26,680 So when I, if I was to run a system using stop losses, which I 223 00:11:26,680 --> 00:11:30,280 don't, as I said, but I design it so that the, the initial entry, 224 00:11:30,280 --> 00:11:35,040 the stock would be set as a, A, a ratio, a multiple of the current 225 00:11:35,040 --> 00:11:35,760 volatility. 226 00:11:37,220 --> 00:11:39,980 Now if you, as the price runs up and the volatility increases, 227 00:11:39,980 --> 00:11:44,900 if you then also increase your stop at the same time, then the position, 228 00:11:45,060 --> 00:11:48,780 you know, having said that, you know, a 25% fall probably still 229 00:11:48,780 --> 00:11:50,180 would have breached any reasonable stop. 230 00:11:50,660 --> 00:11:53,300 But importantly, if you do do that, if you do increase your stop 231 00:11:53,300 --> 00:11:55,980 width, you also should reduce your position at the same time. 232 00:11:55,980 --> 00:11:59,140 Otherwise you're basically taking more and more risk effectively 233 00:11:59,140 --> 00:11:59,700 on the position. 234 00:12:00,660 --> 00:12:03,740 If you don't increase your stop width level as volatility rises, 235 00:12:03,740 --> 00:12:06,820 then you, you're, you risk basically being shaken out the position 236 00:12:06,820 --> 00:12:07,380 too early. 237 00:12:07,460 --> 00:12:10,060 Now on Friday that would have been fine because would have been 238 00:12:10,060 --> 00:12:12,620 very nice if you'd, I don't know, you know, how realistic would 239 00:12:12,620 --> 00:12:14,740 have been to have executed a type stop on Friday. 240 00:12:14,740 --> 00:12:19,140 I don't know how, how rapid the move was, but it would have been, 241 00:12:19,140 --> 00:12:20,980 you know, you probably would have done better than somebody with 242 00:12:20,980 --> 00:12:21,940 a much looser stop. 243 00:12:22,820 --> 00:12:25,820 But you know, that also meant you would have been shaken out on 244 00:12:25,820 --> 00:12:28,700 a much smaller move in the middle of a bigger trend, which been 245 00:12:28,700 --> 00:12:29,500 the wrong thing to do. 246 00:12:29,500 --> 00:12:33,490 So you've got to be careful if you are using stocks how you adjust 247 00:12:33,490 --> 00:12:37,090 them as volatility changes, just as I adjust positions as volatility 248 00:12:37,090 --> 00:12:37,610 changes. 249 00:12:38,250 --> 00:12:44,170 You mentioned kind of if you had the ability to trade on Friday, 250 00:12:45,690 --> 00:12:47,930 clearly it was a huge volatile day. 251 00:12:47,930 --> 00:12:53,410 There's not that many people I've heard from in terms of a logical 252 00:12:53,410 --> 00:12:54,930 explanation as to what happened. 253 00:12:54,930 --> 00:12:58,400 I've seen some speculations about a large US Investment bank 254 00:12:58,400 --> 00:13:01,040 causing all of this, but I don't want to speculate on air on 255 00:13:01,040 --> 00:13:01,240 that. 256 00:13:01,240 --> 00:13:06,120 But do you have a sense or something you, you monitor in some 257 00:13:06,120 --> 00:13:06,440 way? 258 00:13:06,920 --> 00:13:08,840 How, how liquid? 259 00:13:08,920 --> 00:13:14,160 Because I think maybe because silver and gold and platinum, they've 260 00:13:14,160 --> 00:13:18,960 all been going higher in the last many months and of course it's 261 00:13:18,960 --> 00:13:20,600 been great for trend followers. 262 00:13:21,320 --> 00:13:24,320 Maybe people have kind of maybe become a little bit complacent 263 00:13:24,320 --> 00:13:26,880 thinking that these are all the same to trade and they think, 264 00:13:26,880 --> 00:13:29,720 oh, they are all going to be as liquid as gold, which is super 265 00:13:29,720 --> 00:13:30,400 liquid market. 266 00:13:31,200 --> 00:13:35,120 But, but, but they're not the same in, in my view. 267 00:13:35,120 --> 00:13:37,120 And, and I. 268 00:13:37,520 --> 00:13:40,560 So I don't know from your perspective in terms of liquidity 269 00:13:40,640 --> 00:13:44,800 and those kind of risks, if you have any thoughts on what happened. 270 00:13:45,280 --> 00:13:47,760 Yeah, I mean, obviously because I'm trading a relatively 271 00:13:47,760 --> 00:13:51,180 small amount of money, but I kind of have a fairly broad liquidity 272 00:13:51,180 --> 00:13:57,020 filter, so I would probably take positions in instruments where 273 00:13:57,020 --> 00:14:01,460 a lot of big CTAs would either not bother or have a very small position 274 00:14:01,460 --> 00:14:02,020 indeed. 275 00:14:03,140 --> 00:14:07,020 So silver, there's no question of silver not falling into the liquid 276 00:14:07,020 --> 00:14:07,700 bucket for me. 277 00:14:07,700 --> 00:14:08,980 So that wasn't an issue. 278 00:14:09,620 --> 00:14:13,820 But I think that one kind of thing that does happen in markets 279 00:14:13,820 --> 00:14:17,780 that are not as liquid as, you know, the most liquid markets is 280 00:14:17,780 --> 00:14:22,140 that obviously if something runs up in price and you get a slight 281 00:14:22,140 --> 00:14:24,700 pullback, well, all the buyers are gone, right? 282 00:14:24,700 --> 00:14:26,300 Everyone who wants to buy is bought. 283 00:14:26,300 --> 00:14:29,300 So there's no one sitting around waiting to buy off you. 284 00:14:30,100 --> 00:14:34,380 So, you know, people treat liquidity as a constant number, as 285 00:14:34,380 --> 00:14:37,260 a constant number of volume of contracts they can expect to trade. 286 00:14:37,260 --> 00:14:39,340 But of course it isn't right. 287 00:14:39,340 --> 00:14:42,460 And we know that in market crises, liquidity disappears and 288 00:14:42,460 --> 00:14:43,440 that happens everywhere. 289 00:14:44,390 --> 00:14:46,670 And obviously the less liquid the market is to begin with, the 290 00:14:46,670 --> 00:14:47,750 more likely that is to happen. 291 00:14:48,630 --> 00:14:52,950 And I think there is a risk if you, let's say something gets hot, 292 00:14:53,030 --> 00:14:53,350 right? 293 00:14:53,350 --> 00:14:55,710 And it runs up in price and lots of people start trading it, 294 00:14:55,710 --> 00:14:56,950 maybe you hadn't traded it before. 295 00:14:57,270 --> 00:14:59,629 And you look at the volume and the open interest numbers and you 296 00:14:59,629 --> 00:15:01,710 think, well, I can build up a pretty substantial position here 297 00:15:01,710 --> 00:15:03,990 because there's a lot of volume going through, there's a lot 298 00:15:03,990 --> 00:15:05,030 of open interest going through. 299 00:15:06,310 --> 00:15:10,230 You build up that big position and then if the price starts to move 300 00:15:10,230 --> 00:15:12,670 adversely against you, then everyone jams into. 301 00:15:12,670 --> 00:15:15,500 It's like the, the cinema with only one fire door, right? 302 00:15:15,500 --> 00:15:17,580 Everyone gets jammed trying to get through the exit at the same 303 00:15:17,580 --> 00:15:17,900 time. 304 00:15:18,220 --> 00:15:20,380 And the liquidity you thought was there wasn't there. 305 00:15:20,620 --> 00:15:23,540 And part of that's just that that could be driven by a few things, 306 00:15:23,540 --> 00:15:23,700 right? 307 00:15:23,700 --> 00:15:26,020 Part of it could be driven by the fact that, yeah, a lot of these 308 00:15:26,020 --> 00:15:28,380 people are silver tourists who don't normally trade silver. 309 00:15:29,580 --> 00:15:32,580 And the, you know, what you really need to do is go back a year 310 00:15:32,580 --> 00:15:35,460 or two years and say, well, over a longer period of time, what 311 00:15:35,460 --> 00:15:36,620 does liquidity look like? 312 00:15:37,660 --> 00:15:40,700 The other thing is the classic thing where if everyone's trading 313 00:15:40,700 --> 00:15:43,620 similar kind of strategy, I mean, if everyone is trend following, 314 00:15:44,820 --> 00:15:48,340 if everyone is doing volume adjustment and that does, you know, 315 00:15:48,820 --> 00:15:52,020 don't forget that CTAs do volume adjustment, but so do risk 316 00:15:52,020 --> 00:15:55,220 parity funds and so do other kind of quant strategies. 317 00:15:56,340 --> 00:16:01,340 So anyone who sells on the hint of volatility spiking, the more 318 00:16:01,340 --> 00:16:02,980 people there are in a particular market like that, the 319 00:16:02,980 --> 00:16:04,100 more it'll magnify the move. 320 00:16:04,740 --> 00:16:07,340 I mean, the proximate cause of all of this supposedly was the fact 321 00:16:07,340 --> 00:16:11,970 that a relatively sane person got elected to the Fed chair next. 322 00:16:11,970 --> 00:16:13,530 And we'll talk about that in a second, I guess. 323 00:16:13,530 --> 00:16:17,130 But as to the mechanics of why it affected silver more than gold, 324 00:16:17,130 --> 00:16:21,530 and silver went down more in price than gold, I mean, we can speculate 325 00:16:21,530 --> 00:16:23,290 all day, but I think the main lesson. 326 00:16:23,290 --> 00:16:23,570 Here. 327 00:16:25,810 --> 00:16:27,330 For me is risk management. 328 00:16:29,010 --> 00:16:32,770 And the two key things about risk management are position sizing 329 00:16:34,090 --> 00:16:35,450 and diversification. 330 00:16:36,490 --> 00:16:40,530 So on Friday, I lost 3% of my total account value because I have 331 00:16:40,530 --> 00:16:42,250 a reasonably diversified portfolio. 332 00:16:43,450 --> 00:16:46,850 If I was very highly concentrated in silver, that would 333 00:16:46,850 --> 00:16:47,850 have been a much bigger number. 334 00:16:49,290 --> 00:16:52,010 And because I adjust my positions according to volatility 335 00:16:52,010 --> 00:16:54,610 as it changes again, I was relatively protected from a large 336 00:16:54,610 --> 00:16:54,970 move. 337 00:16:55,290 --> 00:16:58,770 Whereas as we've mentioned, funds that don't do that and have 338 00:16:58,770 --> 00:17:01,770 much more leverage potentially are exposed to even bigger moves. 339 00:17:02,160 --> 00:17:05,680 And yeah, month over the month, still profitable. 340 00:17:05,920 --> 00:17:06,960 Great work, guys. 341 00:17:06,960 --> 00:17:10,520 But you know, it scares the hell out of me the possibility of 342 00:17:10,520 --> 00:17:11,600 that happening, I have to say. 343 00:17:13,151 --> 00:17:15,471 Yeah, well, first of all, I mean, I think just to comment on 344 00:17:15,471 --> 00:17:19,551 that number, I think that the number you mentioned was pretty among 345 00:17:19,551 --> 00:17:23,671 people with that kind of annual volatility that you run it 346 00:17:23,671 --> 00:17:25,751 at, I think 3% is very decent. 347 00:17:25,751 --> 00:17:28,911 And I think that's kind of what I saw on the day from the people 348 00:17:28,911 --> 00:17:30,511 I can follow. 349 00:17:31,151 --> 00:17:33,951 So I think the industry as a whole did really, really well. 350 00:17:34,671 --> 00:17:38,911 And as you say, of course, the largest ETA's will be doing something 351 00:17:38,911 --> 00:17:42,431 similar to what you do, meaning adjust the position along 352 00:17:42,431 --> 00:17:42,791 the way. 353 00:17:42,791 --> 00:17:47,631 Which means, again, just maybe to reiterate for people, because 354 00:17:47,631 --> 00:17:51,471 I remember during the cocoa debacle where prices also had a kind 355 00:17:51,471 --> 00:17:55,351 of a similar move, as we saw in silver, where it becomes kind 356 00:17:55,351 --> 00:17:58,271 of a parabolic, and gold for that matter, kind of a parabolic 357 00:17:58,271 --> 00:17:58,591 move. 358 00:17:59,061 --> 00:18:05,301 It's very easy to blame the quant funds to be behind the parabolic 359 00:18:05,301 --> 00:18:05,581 move. 360 00:18:05,581 --> 00:18:11,501 But I would hesit to, to, to say that most CTAs probably got into 361 00:18:11,501 --> 00:18:14,021 silver more than a year ago for sure. 362 00:18:14,581 --> 00:18:19,061 And the majority of the larger CTAs, they, they're the ones who 363 00:18:19,061 --> 00:18:22,981 have been selling silver all the way through January, as you rightly 364 00:18:22,981 --> 00:18:27,861 said, maybe even through December as volatility started to 365 00:18:28,101 --> 00:18:28,741 expand. 366 00:18:29,251 --> 00:18:32,051 We're just going to have to reduce our notional exposure. 367 00:18:32,851 --> 00:18:37,571 So we are actually kind of trying to be the buffer here and 368 00:18:37,571 --> 00:18:39,571 limit these parabolic moves. 369 00:18:39,571 --> 00:18:40,371 If anything. 370 00:18:40,371 --> 00:18:43,771 I think that is important to say because most people will think 371 00:18:43,771 --> 00:18:44,611 the opposite. 372 00:18:45,331 --> 00:18:47,331 The other thing you touched on is risk management. 373 00:18:47,411 --> 00:18:51,691 And I just want to maybe share and I won't do this story justice 374 00:18:51,691 --> 00:18:52,771 in any way, shape or form. 375 00:18:52,771 --> 00:18:56,681 But of course, a lot of people remember silver specifically from 376 00:18:56,681 --> 00:19:02,241 the story of the 1970s where the, I think Texas oil billionaire 377 00:19:02,481 --> 00:19:08,521 Nelson Bunker Hunt, he became convinced that the US dollar would, 378 00:19:08,521 --> 00:19:09,201 would fail. 379 00:19:09,201 --> 00:19:12,761 I mean, it's kind of almost like deja vu at the moment. 380 00:19:12,761 --> 00:19:16,321 Anyways, in his mind, silver was not real money. 381 00:19:16,401 --> 00:19:20,521 That was kind of, or I should say silver was not just an investment 382 00:19:20,521 --> 00:19:22,161 for him, it was real money. 383 00:19:22,671 --> 00:19:25,471 So he started to buy silver and he did it physically. 384 00:19:25,791 --> 00:19:30,951 And he chartered several of these Boeing 707s and he flew that 385 00:19:30,951 --> 00:19:34,991 silver, as far as I understand, millions of ounces of 386 00:19:34,991 --> 00:19:38,351 silver to Switzerland to be stored in vaults. 387 00:19:40,671 --> 00:19:44,951 And so again, this for him was not about making a lot of money necessarily. 388 00:19:44,951 --> 00:19:48,071 It was about kind of controlling and making sure that 389 00:19:48,071 --> 00:19:50,431 the wealth he had built wouldn't disappear. 390 00:19:50,921 --> 00:19:52,361 Well, here comes the problem. 391 00:19:52,601 --> 00:19:56,681 I mean, he kept buying, but he kept buying now for borrowed money. 392 00:19:56,681 --> 00:20:00,841 So the leverage of his position became, you know, quite 393 00:20:00,841 --> 00:20:01,161 high. 394 00:20:02,201 --> 00:20:08,281 And very similar story as to what happened last week as the volatility 395 00:20:09,081 --> 00:20:09,561 rose. 396 00:20:10,281 --> 00:20:15,481 What happened in 1979, price wise was that silver went from $6 397 00:20:15,561 --> 00:20:17,321 to about $50 an ounce. 398 00:20:18,301 --> 00:20:20,141 And that causes a problem for the exchange. 399 00:20:20,301 --> 00:20:23,261 So Comex started to change the rules on margins. 400 00:20:23,261 --> 00:20:26,821 We saw the same thing happening, I think in January once 401 00:20:26,821 --> 00:20:28,861 or twice, where margin got increased. 402 00:20:28,861 --> 00:20:33,660 And at some point the people who have made money, but they've 403 00:20:33,660 --> 00:20:38,261 made money from borrowed or speculation from borrowed money to 404 00:20:38,261 --> 00:20:43,421 keep up with the margin calls, at some point they run out of money 405 00:20:43,581 --> 00:20:44,541 and so did he. 406 00:20:44,781 --> 00:20:50,231 So I think don't know the details, but he ended up having, 407 00:20:50,231 --> 00:20:52,951 you know, a catastrophic experience. 408 00:20:54,631 --> 00:20:57,671 Maybe he still ended up making a little bit of money, but, but not 409 00:20:57,671 --> 00:20:59,071 nearly as much as he had. 410 00:20:59,071 --> 00:21:03,871 And I think there was some agreement about paying back debt. 411 00:21:03,871 --> 00:21:09,191 So risk management and not letting positions get completely 412 00:21:09,191 --> 00:21:13,071 out of control by pyramiding and so on and so forth are all very 413 00:21:13,071 --> 00:21:13,831 important lessons. 414 00:21:13,831 --> 00:21:16,951 And even though this happened back in the 70s, as we can clearly 415 00:21:16,951 --> 00:21:21,471 see they get repeated over and over again, just in different markets 416 00:21:21,471 --> 00:21:22,231 at different times. 417 00:21:23,271 --> 00:21:27,751 But I think the beauty of all of this is the fact that many of 418 00:21:27,751 --> 00:21:32,151 the managers that we talk about, they've all been around for 419 00:21:32,151 --> 00:21:32,711 decades. 420 00:21:32,951 --> 00:21:35,951 We kind of know this, it's part of our DNA. 421 00:21:35,951 --> 00:21:37,751 We don't get emotionally attached to it. 422 00:21:37,751 --> 00:21:40,951 We manage our risk in a systematic, dispassionate way. 423 00:21:41,461 --> 00:21:42,501 And that's super important. 424 00:21:43,141 --> 00:21:46,901 And regarding this idea that people should also remember that 425 00:21:46,901 --> 00:21:49,301 gold is not, sorry, silver is not the same as gold. 426 00:21:50,341 --> 00:21:53,901 I've heard stories or information about the fact that the 427 00:21:53,901 --> 00:21:57,701 reason why this is important is that silver, you don't mine very 428 00:21:57,701 --> 00:21:58,981 much silver directly. 429 00:21:58,981 --> 00:22:02,021 It's only like 30% of the production actually every year is 430 00:22:02,341 --> 00:22:04,101 specifically mined as silver. 431 00:22:04,101 --> 00:22:07,541 The rest comes from a byproduct of gold and copper. 432 00:22:08,181 --> 00:22:12,101 So it's maybe less controllable or the supply side maybe 433 00:22:12,101 --> 00:22:19,261 is a little bit harder to predict and therefore it can maybe 434 00:22:19,261 --> 00:22:23,061 also be the reason why silver can be a much more difficult, much 435 00:22:23,061 --> 00:22:25,541 more volatile market to trade compared to gold. 436 00:22:25,861 --> 00:22:29,461 But these are all very important points in my opinion. 437 00:22:29,541 --> 00:22:30,341 Absolutely. 438 00:22:30,581 --> 00:22:33,831 Thing is, may also briefly mentioning, I think that Friday and 439 00:22:33,831 --> 00:22:36,191 actually continuing to this week, we're seeing a lot of dislocations 440 00:22:36,591 --> 00:22:37,911 across markets generally. 441 00:22:37,911 --> 00:22:41,471 So you know, long short quant equity factors have gone a balmy 442 00:22:41,871 --> 00:22:47,311 apparently AQR's long short fund had both one of its worst days 443 00:22:47,311 --> 00:22:49,351 and also one of its best days in the last few days. 444 00:22:49,351 --> 00:22:50,351 So that's interesting. 445 00:22:51,231 --> 00:22:53,151 Bitcoin of course is selling off massively. 446 00:22:53,791 --> 00:22:57,191 There's been kind of disruptions in the particularly in 447 00:22:57,191 --> 00:22:59,271 the sort of stocks related to AI. 448 00:22:59,271 --> 00:23:02,321 So it's all kind of interesting and at least at the moment, 449 00:23:02,481 --> 00:23:03,841 Touchwood, fingers crossed. 450 00:23:04,401 --> 00:23:06,801 My trend following portfolio seems to be coping pretty well with 451 00:23:06,801 --> 00:23:07,281 all of this. 452 00:23:08,641 --> 00:23:10,161 So yeah, it's an interesting time. 453 00:23:10,321 --> 00:23:13,561 Can I just mention, I mean we talk about precious metals as if 454 00:23:13,561 --> 00:23:19,241 it was the only thing that moved March natural gas, just to 455 00:23:19,241 --> 00:23:22,641 mention, from the 9th of January to the 30th of January it 456 00:23:23,761 --> 00:23:30,781 was up something like 60% and then on the 2nd of February, in one 457 00:23:30,781 --> 00:23:32,981 day it lost 26%. 458 00:23:33,221 --> 00:23:35,461 Similar move to more, more or less to Silver. 459 00:23:35,541 --> 00:23:37,781 And this is, this is again, this is weather related. 460 00:23:37,781 --> 00:23:40,181 This is due to the, to the massive storm. 461 00:23:40,421 --> 00:23:43,541 So we can't blame that, we can't blame that one on the Fed chair. 462 00:23:44,181 --> 00:23:45,501 Well, we can try, but. 463 00:23:45,501 --> 00:23:48,461 Okay, we'll speak about him in a second. 464 00:23:48,461 --> 00:23:51,941 But yeah, what's interesting about it is, and this is why I think 465 00:23:51,941 --> 00:23:56,191 we've been advocating for so many years now on the podcast and 466 00:23:56,191 --> 00:24:00,431 that is people need to realize how important it is to have the commodities 467 00:24:00,431 --> 00:24:04,191 as part of their trend following universe because this is 468 00:24:04,191 --> 00:24:10,551 where we sometimes see lots and lots of opportunities and often 469 00:24:11,431 --> 00:24:13,711 at times where there might be a risk off event. 470 00:24:13,711 --> 00:24:16,151 We didn't see any risk off event here in equities. 471 00:24:16,151 --> 00:24:18,791 They didn't really bother too much about what happened. 472 00:24:20,341 --> 00:24:23,781 But they're just wonderful in terms of opportunities. 473 00:24:23,941 --> 00:24:26,621 Not always for sure, but at times. 474 00:24:26,621 --> 00:24:28,101 And this is why we like them. 475 00:24:28,181 --> 00:24:31,261 But speaking of what you said could be the reason. 476 00:24:31,261 --> 00:24:33,181 I don't know that it really was, frankly. 477 00:24:33,181 --> 00:24:37,301 Again, I hear this story in my back of my mind about a certain investment 478 00:24:37,541 --> 00:24:41,781 bank that was behind a lot of this. 479 00:24:42,421 --> 00:24:47,661 But yes, at the same time we were told that we will have a new 480 00:24:48,141 --> 00:24:52,781 if, if confirmed, we will have a new Fed chair with the name of 481 00:24:52,781 --> 00:24:55,141 Kevin Walsh, which was a surprise. 482 00:24:55,141 --> 00:24:57,781 To me because anyone who listened to the Christmas specials 483 00:24:57,781 --> 00:25:00,461 will know that my outrageous prediction for this year was that 484 00:25:00,621 --> 00:25:02,460 the new Fed chair would be Baron Trump. 485 00:25:02,541 --> 00:25:07,021 So fortunately you already out of the. 486 00:25:07,181 --> 00:25:08,701 I'm already out of the predictions. 487 00:25:08,701 --> 00:25:09,461 Oh my God. 488 00:25:09,461 --> 00:25:12,301 Yeah, tough, tough start to the year actually. 489 00:25:12,621 --> 00:25:18,021 Anyways, tell me what you think about this. 490 00:25:18,021 --> 00:25:22,901 Not that our opinion is particular insightful, I'm sure, 491 00:25:22,901 --> 00:25:24,301 but what are your thoughts? 492 00:25:25,501 --> 00:25:29,701 I must say so, you know, I think I kind of want to sort of fess 493 00:25:29,701 --> 00:25:32,381 up that I'm not a fan of the current President of the United States. 494 00:25:32,381 --> 00:25:36,541 And this will come as a surprise to many people and I was 495 00:25:36,541 --> 00:25:39,141 of the opinion when he came into office a year ago that it would 496 00:25:39,141 --> 00:25:41,791 be very bad for markets. 497 00:25:42,511 --> 00:25:45,871 And mostly I admit I've been wrong. 498 00:25:46,591 --> 00:25:51,471 There was obviously the tariff tantrum and which did something very 499 00:25:51,471 --> 00:25:55,870 silly and non systematic and sold all of my equities that cost 500 00:25:55,870 --> 00:25:58,751 me a fair bit of relative performance, sadly. 501 00:26:00,111 --> 00:26:03,471 But generally speaking, markets have kind of shrugged off 502 00:26:04,751 --> 00:26:05,551 to an extent. 503 00:26:05,551 --> 00:26:08,631 And maybe that's because of Trump always chickens out the taco 504 00:26:08,631 --> 00:26:08,911 trade. 505 00:26:10,131 --> 00:26:12,771 Maybe, maybe no one really takes him seriously, who knows. 506 00:26:13,571 --> 00:26:17,051 But I think it was an expectation that he was going to 507 00:26:17,051 --> 00:26:20,651 appoint someone to the Fed who would be basically in line with his 508 00:26:20,651 --> 00:26:21,091 wishes. 509 00:26:21,091 --> 00:26:26,451 So now, and there's sort of a view that if you look at the spectrum 510 00:26:26,451 --> 00:26:31,091 of people who he could have appointed, then this guy is on the 511 00:26:31,091 --> 00:26:35,171 kind of more sensible, insane end of that spectrum. 512 00:26:35,631 --> 00:26:40,831 I mean, he does have some opinions that are less mainstream 513 00:26:40,831 --> 00:26:44,471 around things like QE and the Fed balance sheet and so on and so 514 00:26:44,471 --> 00:26:44,831 forth. 515 00:26:45,311 --> 00:26:48,911 But in terms of just interest rate setting, actually in, you know, 516 00:26:48,911 --> 00:26:50,871 some, some people have said, well, actually this guy is quite 517 00:26:50,871 --> 00:26:54,991 hawkish if you look at his, you know, look at some of his decision 518 00:26:55,071 --> 00:26:56,111 making over the years. 519 00:26:56,751 --> 00:27:01,921 Now a key question is, you know, given that Trump has installed 520 00:27:01,921 --> 00:27:04,041 or, you know, let's assume he gets confirmed. 521 00:27:04,201 --> 00:27:08,241 And I think he will do, to be honest, because I think the, you 522 00:27:08,241 --> 00:27:10,841 know, the Senate and the House will be just so relieved. 523 00:27:11,081 --> 00:27:14,001 They'll be like, yeah, okay, this, you know, this, this guy's 524 00:27:14,001 --> 00:27:15,961 probably better than any other option we could get. 525 00:27:15,961 --> 00:27:20,041 So I think he will be confirmed and this will all happen 526 00:27:20,041 --> 00:27:21,121 before the midterms anyway. 527 00:27:21,121 --> 00:27:22,841 So I think numerically the numbers are there. 528 00:27:23,801 --> 00:27:26,961 The question is, once he's actually there, given that he's been 529 00:27:26,961 --> 00:27:31,201 appointed, you know, by someone who demands loyalty, whether 530 00:27:31,761 --> 00:27:35,801 if he then, will he then basically do as he's expected to 531 00:27:35,801 --> 00:27:39,281 do, which is just to kind of cut rates and keep the former real 532 00:27:39,281 --> 00:27:44,441 estate developer happy, or will he not do that and then have 533 00:27:44,441 --> 00:27:48,281 pressure put upon him like the current Fed chair has pressure put 534 00:27:48,281 --> 00:27:53,481 upon him and if, and then will he then crumble and say, okay, fine, 535 00:27:53,481 --> 00:27:56,161 I'll do what you say, or will he, like the current Fed chairman 536 00:27:56,161 --> 00:27:59,481 has fight back and say, no, you've appointed me now I'm basically 537 00:28:00,521 --> 00:28:01,241 independent. 538 00:28:01,481 --> 00:28:02,841 I'm going to do what I think is best. 539 00:28:04,601 --> 00:28:10,321 So, yeah, it's interesting and I think the true answer is no one 540 00:28:10,321 --> 00:28:10,921 really knows. 541 00:28:11,001 --> 00:28:13,961 And this is true of most things related to the current President. 542 00:28:14,601 --> 00:28:17,081 No one really knows what's going to happen in May. 543 00:28:17,081 --> 00:28:20,681 I think it's May when Powell's terms up, no one really knows what's 544 00:28:20,681 --> 00:28:21,081 going to happen. 545 00:28:21,081 --> 00:28:27,351 So I think the reaction was if that, you know, I think that was 546 00:28:27,351 --> 00:28:29,231 quite a big reaction, to be honest. 547 00:28:29,631 --> 00:28:31,391 If that really is approximate cause. 548 00:28:32,271 --> 00:28:34,631 And you know, if you actually look at say the bond market, where 549 00:28:34,631 --> 00:28:37,870 you could, which you could argue is the most direct measure 550 00:28:37,870 --> 00:28:41,551 of this, it was a roughly muted mood in the bond market. 551 00:28:42,511 --> 00:28:45,151 So I haven't got the figures to hand, but I don't remember anyone's, 552 00:28:45,151 --> 00:28:48,091 you know, I don't remember the, you know, the two year bond 553 00:28:48,171 --> 00:28:51,491 kind of moving by sort of 50 basis points on the day, which is 554 00:28:51,491 --> 00:28:54,731 what you would have expected if this genuinely was like a novel 555 00:28:54,731 --> 00:28:55,931 and interesting piece of information. 556 00:28:57,611 --> 00:29:00,171 So I think everyone's relieved. 557 00:29:00,651 --> 00:29:02,731 Got a bit carried away with their relief perhaps. 558 00:29:03,051 --> 00:29:06,291 But as to whether this is good news or bad news for the bond market, 559 00:29:06,291 --> 00:29:09,931 for the economy generally, we'll have to wait and see. 560 00:29:10,331 --> 00:29:13,131 I think you kind of touched on, hit it on the nail here because 561 00:29:13,131 --> 00:29:14,541 I hit the nail on the head here because. 562 00:29:14,611 --> 00:29:19,091 Because the reaction in the financial markets were very muted. 563 00:29:19,491 --> 00:29:23,491 So this thing about him being the cause of a silver sell off, I 564 00:29:23,491 --> 00:29:26,251 think it had much more to do with positioning and how people had 565 00:29:26,251 --> 00:29:29,451 been piling into this and they talk about the Chinese retail investor 566 00:29:29,451 --> 00:29:30,171 and all that stuff. 567 00:29:30,171 --> 00:29:33,971 I really don't think it had anything to do with him really. 568 00:29:34,131 --> 00:29:37,851 But of course we always hear the, in the news that there is some 569 00:29:37,851 --> 00:29:40,211 kind of cause you can point to. 570 00:29:40,741 --> 00:29:43,621 I think you are, I mean for people who don't know. 571 00:29:43,621 --> 00:29:48,581 But as far as I can tell, his father in law is a massive financial 572 00:29:48,661 --> 00:29:49,621 backer of Trump. 573 00:29:49,781 --> 00:29:53,461 So obviously it may not be that surprising that he got the job. 574 00:29:53,941 --> 00:29:56,420 Having said that, let's not forget it was Trump who appointed 575 00:29:56,420 --> 00:29:56,901 Powell. 576 00:29:57,061 --> 00:29:58,181 So you never know. 577 00:29:58,181 --> 00:30:01,541 As you say, once they are in and they can't be in something which. 578 00:30:01,541 --> 00:30:03,301 He appears to have forgotten as well. 579 00:30:03,461 --> 00:30:03,861 Right. 580 00:30:04,021 --> 00:30:07,871 And once they're in they really should concern themselves 581 00:30:07,871 --> 00:30:10,911 about their own legacy and not so much the legacy of others and 582 00:30:10,911 --> 00:30:12,391 then do the best possible job. 583 00:30:12,391 --> 00:30:13,271 So we don't know. 584 00:30:13,271 --> 00:30:19,471 And then also let's not forget he's one of is it 12 votes in total? 585 00:30:19,471 --> 00:30:21,991 I mean it's not like he can decide what interest rates are going 586 00:30:21,991 --> 00:30:22,351 to do. 587 00:30:22,511 --> 00:30:29,111 He has to compel or convince or force others to go in his direction. 588 00:30:29,111 --> 00:30:32,511 Now there might be one or two that is very easy to, to convince 589 00:30:33,231 --> 00:30:36,151 steering Steve Moran and maybe another one. 590 00:30:36,151 --> 00:30:39,231 But we're still waiting for this Supreme Court decision as to 591 00:30:39,931 --> 00:30:42,611 Trump can effectively then pack the rest of the board with whoever 592 00:30:42,611 --> 00:30:43,051 he wants. 593 00:30:43,531 --> 00:30:44,691 Fire and higher at will. 594 00:30:44,691 --> 00:30:45,691 Yeah, yeah, yeah. 595 00:30:46,011 --> 00:30:47,611 All things up in the air. 596 00:30:47,771 --> 00:30:52,131 But what you could say in general is of course that because 597 00:30:52,131 --> 00:30:56,491 you mentioned this thing that yeah a year ago you weren't too bullish 598 00:30:56,491 --> 00:30:57,171 about equities. 599 00:30:57,171 --> 00:31:00,571 But I mean at the end of the day, however high this market goes, 600 00:31:00,891 --> 00:31:04,851 it could still all end, you know, pretty pear shaped because 601 00:31:04,851 --> 00:31:07,451 of all these things that are building up to it. 602 00:31:07,451 --> 00:31:10,491 It's always the timing of these things that is impossible to. 603 00:31:11,271 --> 00:31:12,151 But we'll see. 604 00:31:12,151 --> 00:31:17,391 And I think as again as we mentioned many times in what's going 605 00:31:17,391 --> 00:31:18,551 on in the world right now. 606 00:31:19,031 --> 00:31:24,991 Being a systematic diversified investor I think is so much better 607 00:31:24,991 --> 00:31:28,031 than trying to make any sense of this from a discretionary point 608 00:31:28,031 --> 00:31:28,391 of view. 609 00:31:28,871 --> 00:31:31,351 Even though those who get it right will probably get it really 610 00:31:31,351 --> 00:31:33,831 right, but you can really get it wrong as well. 611 00:31:34,071 --> 00:31:38,151 Yeah, well Discretionary Rob sold all of his equities in April 612 00:31:38,471 --> 00:31:41,111 last year and Systematic Rob is up since then. 613 00:31:41,111 --> 00:31:44,031 Is up, well, you know, 20, 25%. 614 00:31:44,031 --> 00:31:46,391 So I think you should listen more to the Systematic Rob. 615 00:31:46,391 --> 00:31:46,951 Clearly. 616 00:31:47,271 --> 00:31:47,831 Clearly. 617 00:31:48,631 --> 00:31:53,031 Anyways, talking about quote unquote, the Systematic Rob before 618 00:31:53,031 --> 00:31:55,711 we move often to the trend following part and I know maybe we'll 619 00:31:55,711 --> 00:31:59,631 have time, I'm not so sure to go back to this point about AI but 620 00:31:59,631 --> 00:32:03,871 what did hit my radar just this morning related to AI was just 621 00:32:03,871 --> 00:32:07,741 this headline I think in Financial Times that Google said 622 00:32:07,741 --> 00:32:12,541 that it's planning to double its capex this year to as much as 623 00:32:12,541 --> 00:32:17,501 $185 billion to really bet huge on AI. 624 00:32:17,741 --> 00:32:21,541 Now I know they just announced results and they made profits of 625 00:32:21,541 --> 00:32:24,061 something like 130, 32 billion. 626 00:32:24,061 --> 00:32:27,741 So I mean they have a lot of money to spend but still doubling 627 00:32:27,741 --> 00:32:30,501 the AI spend to almost $200 billion. 628 00:32:30,501 --> 00:32:34,871 I mean this does sound like a bubble to me, but who knows. 629 00:32:34,871 --> 00:32:38,191 Yeah, we're both old enough to remember 99, 2000, right? 630 00:32:38,191 --> 00:32:40,911 So yes we are, unfortunately. 631 00:32:40,911 --> 00:32:44,391 Anyways, let's talk about some trend following updates before we 632 00:32:44,391 --> 00:32:47,031 dive into a really, really interesting article. 633 00:32:47,031 --> 00:32:48,631 People should really look forward to this. 634 00:32:48,711 --> 00:32:53,071 Anyways, not surprising, my trend barometer is having a great 635 00:32:53,071 --> 00:32:53,351 time. 636 00:32:53,351 --> 00:32:57,871 It finished yesterday which was Wednesday at 64. 637 00:32:57,871 --> 00:32:59,351 That's a super strong reading. 638 00:32:59,351 --> 00:33:02,991 And of course what it really tells you is that there is a lot 639 00:33:02,991 --> 00:33:06,311 of breadth because the percentage or 64 means that there 640 00:33:06,311 --> 00:33:10,151 are 64 of the market that in the portfolio it tracks that are 641 00:33:10,151 --> 00:33:11,391 quote unquote trending. 642 00:33:11,391 --> 00:33:14,991 So it's, it's about breadth in the environment right now. 643 00:33:15,711 --> 00:33:17,591 So that's the important part. 644 00:33:17,591 --> 00:33:21,351 And if we look at, I mean we already alluded to the fact that 645 00:33:21,351 --> 00:33:25,231 January was a strong month for, for, for managers, February 646 00:33:25,231 --> 00:33:26,911 is off to a really good start as well. 647 00:33:27,551 --> 00:33:32,271 Mainly this time driven by, as far as I can tell, the financial 648 00:33:32,431 --> 00:33:35,831 sectors like equities, a bit of currency, some of the fixed income 649 00:33:35,831 --> 00:33:40,471 markets even are supporting the P and L this month. 650 00:33:40,471 --> 00:33:45,351 And also things like natural gas, a little bit of meats, some 651 00:33:45,351 --> 00:33:46,031 of the grains. 652 00:33:46,511 --> 00:33:50,111 So the breadth of the trend environment as far as I can tell 653 00:33:51,311 --> 00:33:52,271 is pretty good. 654 00:33:53,471 --> 00:33:57,591 So before I dive into the numbers, Rob, anything you want to 655 00:33:57,591 --> 00:34:02,011 Add to kind of where we stand and, and, and your. 656 00:34:02,011 --> 00:34:04,451 From your vantage point overall. 657 00:34:04,851 --> 00:34:06,011 Yeah, yeah. 658 00:34:06,011 --> 00:34:09,651 So I actually don't think I've got a position in natural gas, sadly. 659 00:34:10,451 --> 00:34:11,691 Well, the ball is huge. 660 00:34:11,691 --> 00:34:13,651 So yeah, the position size would be. 661 00:34:13,731 --> 00:34:15,651 Very too big for my tiny account. 662 00:34:15,731 --> 00:34:18,971 So yeah, the one thing I would add, and I know not everyone trades 663 00:34:18,971 --> 00:34:22,611 it but bitcoin, obviously, I think the market that's on everyone's 664 00:34:22,611 --> 00:34:25,491 lips this morning, at least on X Twitter is, is bitcoin. 665 00:34:25,820 --> 00:34:29,420 Oh and I think most CTAs would, if we're trading it would 666 00:34:29,580 --> 00:34:32,340 probably, I mean I trade it just for the futures. 667 00:34:32,340 --> 00:34:36,220 Just to say most CTAs who trade it would probably be short. 668 00:34:37,100 --> 00:34:40,780 And that, that's especially my biggest short position at the moment. 669 00:34:40,780 --> 00:34:46,060 So that, that will, that might be kind of doing some interesting 670 00:34:46,060 --> 00:34:48,140 things for some people as well, I would say. 671 00:34:48,780 --> 00:34:49,420 But definitely. 672 00:34:49,420 --> 00:34:54,831 Yeah, No, I, I'm also my, my risk is a little bit higher than 673 00:34:54,831 --> 00:34:55,711 it has been for a while. 674 00:34:55,711 --> 00:34:57,671 So that kind of matches up with your trend barometer. 675 00:34:57,671 --> 00:34:58,151 Definitely. 676 00:34:58,391 --> 00:34:58,871 Yeah. 677 00:34:59,031 --> 00:35:00,671 Okay, cool. 678 00:35:00,671 --> 00:35:01,671 All right, good stuff. 679 00:35:01,991 --> 00:35:04,711 Let's run through the numbers quickly and then we'll dive into 680 00:35:05,031 --> 00:35:05,831 the first article. 681 00:35:06,631 --> 00:35:10,391 So these are numbers are as of Tuesday because the Wednesday numbers 682 00:35:10,391 --> 00:35:12,791 have not been published at the time of we are recording today. 683 00:35:13,431 --> 00:35:17,671 But as I mentioned, A positive start beta 50 up 33 basis points, 684 00:35:17,671 --> 00:35:19,751 up 5.38% so far this year. 685 00:35:19,751 --> 00:35:20,471 So very strong. 686 00:35:22,101 --> 00:35:26,821 Soc Gen CTA index up 27 basis points, up 5% so far this year. 687 00:35:27,301 --> 00:35:30,701 Sucgen trend up another half a percent in February, up five and 688 00:35:30,701 --> 00:35:31,701 a quarter for the year. 689 00:35:32,101 --> 00:35:35,581 And the Short Term traders index also up 13 basis points and 690 00:35:35,581 --> 00:35:37,861 up 2.37% so far this year. 691 00:35:38,421 --> 00:35:41,621 The MSCI World Index on the other hand is down half a percent 692 00:35:41,621 --> 00:35:45,781 so far in February as of last night, up 1.76 for the month. 693 00:35:46,351 --> 00:35:49,991 And the US aggregate bond index so far pretty flat in February 694 00:35:49,991 --> 00:35:51,391 and was pretty flat in January. 695 00:35:51,391 --> 00:35:53,071 So that means we're pretty flat for the year. 696 00:35:53,231 --> 00:35:57,551 And the S&P 500 down 81 basis points so far in February and up 697 00:35:57,551 --> 00:36:01,151 63 basis points so far in 2026. 698 00:36:07,151 --> 00:36:11,431 Now before we get to the article, we actually got two questions 699 00:36:11,431 --> 00:36:15,911 in from Pedro who wrote in and let me just run through those with 700 00:36:15,911 --> 00:36:17,911 you before we dive into it. 701 00:36:18,311 --> 00:36:22,031 The first question, and this is a long one so I'll probably probably 702 00:36:22,031 --> 00:36:23,111 butcher it a little bit. 703 00:36:23,111 --> 00:36:26,511 I'll do my best estimating realized book correlation from P 704 00:36:26,511 --> 00:36:30,311 and L rather than estimating pairwise asset correlations. 705 00:36:30,311 --> 00:36:33,191 I've been thinking about measuring the daily correlation of 706 00:36:33,191 --> 00:36:39,831 my entire book's return using something like a exponentially weighed 707 00:36:39,831 --> 00:36:40,551 moving average. 708 00:36:40,921 --> 00:36:45,401 The logic is that this already factors in both asset correlations 709 00:36:45,401 --> 00:36:48,921 and signal correlations as expressed in realized P and L. I 710 00:36:48,921 --> 00:36:53,081 understand this adds noise, but the alternative, assuming a high 711 00:36:53,321 --> 00:36:57,081 fixed high correlation like 0.7 doesn't match reality. 712 00:36:57,081 --> 00:37:00,441 For my book, my average realized correlation is much lower. 713 00:37:00,681 --> 00:37:05,041 So assuming 0.7 means I consistently undershoot my volume 714 00:37:05,041 --> 00:37:05,321 target. 715 00:37:05,721 --> 00:37:09,371 Is there a sensible middle ground here or a bottom better way 716 00:37:09,371 --> 00:37:13,971 to calibrate this interesting point because also if you listen 717 00:37:13,971 --> 00:37:17,771 to which I'm sure you do, Rob, last week's conversation with Katie, 718 00:37:17,771 --> 00:37:21,891 we did actually talk about not so much correlations, but certainly 719 00:37:21,891 --> 00:37:23,051 also Vol. 720 00:37:23,291 --> 00:37:27,731 And how that really can have a huge impact as we just talked about 721 00:37:27,731 --> 00:37:32,611 with Silver, in terms of your overall position size and, and so 722 00:37:32,611 --> 00:37:33,451 on and so forth. 723 00:37:35,221 --> 00:37:39,741 Yeah, I'm a little bit unclear about this question, so I'm going 724 00:37:39,741 --> 00:37:42,781 to make some assumptions and apologies Pedro, if my assumptions 725 00:37:42,781 --> 00:37:44,261 are wrong and I'm answering the wrong question. 726 00:37:44,261 --> 00:37:47,381 But what I think you're talking about is imagine you're trading 727 00:37:47,381 --> 00:37:50,901 a bunch of instruments so you've got a bunch of trend following 728 00:37:51,221 --> 00:37:55,609 systems, one for silver, one for gold, one for, you know, S&P 729 00:37:55,713 --> 00:37:56,821 500 or whatever. 730 00:37:57,221 --> 00:38:01,371 And then what you need to do is essentially scale look, kind of 731 00:38:01,371 --> 00:38:04,531 work out what the sort of total leverage is required on that 732 00:38:04,771 --> 00:38:07,331 to get to a given risk target. 733 00:38:07,891 --> 00:38:11,451 Now if all assuming that you're doing volatility targeting 734 00:38:11,451 --> 00:38:14,291 and everything, I mean it's got the same volatility, that means 735 00:38:14,291 --> 00:38:17,531 the only thing that's going to change your your risk outcome is 736 00:38:17,531 --> 00:38:18,771 the correlation of those things. 737 00:38:19,331 --> 00:38:22,491 So if they're perfectly correlated, if you're targeting say 738 00:38:22,491 --> 00:38:26,291 a 15% risk target on each of them, you'll end up with a 15% risk 739 00:38:26,291 --> 00:38:28,691 target on the entire book because it's perfectly correlated. 740 00:38:29,601 --> 00:38:33,241 Now if your correlation is less than 1, which is hopefully the 741 00:38:33,241 --> 00:38:36,121 case because we love diversification in trend following 742 00:38:36,121 --> 00:38:36,401 space. 743 00:38:36,401 --> 00:38:37,441 We absolutely love it. 744 00:38:38,001 --> 00:38:39,601 We like nice low correlations. 745 00:38:39,921 --> 00:38:44,201 The lower that correlation is, the more likely it is that you'll 746 00:38:44,201 --> 00:38:45,201 undershoot your risk target. 747 00:38:45,201 --> 00:38:48,121 So instead of getting to 15% on your whole book, you'll only have 748 00:38:48,121 --> 00:38:49,201 10 or perhaps five. 749 00:38:49,521 --> 00:38:54,881 And actually the kind of when I actually do this exercise and work 750 00:38:54,881 --> 00:38:57,961 out how much I need to increase my position size to compensate 751 00:38:57,961 --> 00:39:01,461 for the diversification effect across instrument, I get numbers 752 00:39:01,461 --> 00:39:02,701 of between 4 and 5. 753 00:39:03,421 --> 00:39:07,941 So a 15% risk target would turn into just 3% risk if I didn't 754 00:39:07,941 --> 00:39:09,261 do this scaling up. 755 00:39:09,901 --> 00:39:13,821 So the question Pedro's asking I think is how do we actually kind 756 00:39:13,821 --> 00:39:15,501 of calculate this number? 757 00:39:16,461 --> 00:39:19,181 Now he's saying use a fixed correlation of 0.7. 758 00:39:19,901 --> 00:39:23,301 My gut feeling is that he's got that 0.7 from one of my books 759 00:39:23,301 --> 00:39:25,661 and it's a slight misunderstanding on his part because 760 00:39:26,061 --> 00:39:31,291 0.7 is actually roughly speaking the ratio between the correlation 761 00:39:31,291 --> 00:39:34,451 between the underlying instrument returns and the correlation 762 00:39:34,451 --> 00:39:37,051 between what happens if you trend follow those instruments. 763 00:39:37,371 --> 00:39:40,931 Because trend following them, because you're trend following them, 764 00:39:40,931 --> 00:39:42,731 you're not going to get, you're going to actually get a lower 765 00:39:42,731 --> 00:39:45,691 correlation than if you were just holding them long only positions 766 00:39:45,691 --> 00:39:45,971 on each. 767 00:39:45,971 --> 00:39:47,851 So that's where I think the 0.7 comes from. 768 00:39:48,331 --> 00:39:50,651 So in answer to the question, how should you actually estimate 769 00:39:50,651 --> 00:39:51,331 this correlation? 770 00:39:51,331 --> 00:39:53,811 Well, yes, it's a perfectly reasonable thing to do to look at 771 00:39:53,811 --> 00:39:57,231 the, the P and L stream that comes from Silver, the P and L stream 772 00:39:57,231 --> 00:40:00,191 comes from S&P 500 and then measure the correlation of those. 773 00:40:00,191 --> 00:40:01,391 And Pedro's exactly right. 774 00:40:01,391 --> 00:40:07,111 That will basically be factoring in both asset correlations 775 00:40:07,111 --> 00:40:08,271 and signal correlations. 776 00:40:08,271 --> 00:40:13,031 And yeah, you can get, can just use a rolling estimation window 777 00:40:13,031 --> 00:40:14,351 of about six months. 778 00:40:14,351 --> 00:40:15,151 Works quite well. 779 00:40:15,151 --> 00:40:17,871 Or you know, if you want a smoother correlation estimate, you 780 00:40:17,871 --> 00:40:19,711 can use an exponential weighting of correlation. 781 00:40:19,711 --> 00:40:20,991 That that's also fine. 782 00:40:22,561 --> 00:40:24,801 So yeah, I think the answer is yes. 783 00:40:25,521 --> 00:40:27,761 I'm not sure what he means by adds noise. 784 00:40:29,361 --> 00:40:33,041 Maybe he means referring to the fact that if you used a fixed 785 00:40:33,201 --> 00:40:36,321 because of the correlation estimates always changing that will 786 00:40:36,721 --> 00:40:38,401 result in some extra trading. 787 00:40:38,401 --> 00:40:40,841 But to be honest, that is as long as your correlation estimate's 788 00:40:40,841 --> 00:40:43,841 quite slow, like six months is pretty good, then that's going to 789 00:40:43,841 --> 00:40:47,921 be effectively irrelevant because your actual underlying trading 790 00:40:47,921 --> 00:40:50,641 process would be giving you a much bigger effect. 791 00:40:51,421 --> 00:40:53,421 Yeah, I think that's a beautiful answer. 792 00:40:53,821 --> 00:40:56,381 He had one more follow up question that is predictability. 793 00:40:56,781 --> 00:41:02,141 He says the research suggests correlation is weakly persistent 794 00:41:02,701 --> 00:41:07,661 but that the best forecast is often just the long run average with 795 00:41:07,661 --> 00:41:11,021 the main predictable component being volatility regime. 796 00:41:11,341 --> 00:41:14,301 Do you find it worth adjusting correlation assumptions based on 797 00:41:14,301 --> 00:41:18,761 volume regime or is a constant conservative assumption sufficient? 798 00:41:19,881 --> 00:41:22,921 So this is a little bit again, I think there's a bit of confusion 799 00:41:23,161 --> 00:41:25,801 maybe in the way the question's phrased from Pedro's understanding. 800 00:41:26,041 --> 00:41:29,561 Now most of the academic research on estimation is around 801 00:41:29,561 --> 00:41:33,000 the estimation of covariance matrices, not correlation matrices. 802 00:41:34,041 --> 00:41:37,521 Now a covariance matrix is basically what you get if you combine 803 00:41:37,521 --> 00:41:40,521 a volatility and a correlation matrix together. 804 00:41:41,481 --> 00:41:46,011 Because in most kind of financial Optimization, it's the 805 00:41:46,011 --> 00:41:47,611 covariance matrix that you use. 806 00:41:47,611 --> 00:41:50,371 So if you're doing, you know, your Markovits, you invert your covariance 807 00:41:50,371 --> 00:41:50,931 matrix. 808 00:41:52,211 --> 00:41:57,091 Now what I found and what the academic research finds as well is 809 00:41:57,091 --> 00:42:00,371 that the predictability of volatility and correlation is different 810 00:42:01,011 --> 00:42:03,571 and it's better to estimate them separately and then combine 811 00:42:03,571 --> 00:42:05,091 them together if that's what you want to do. 812 00:42:05,811 --> 00:42:09,891 So volatility, for example, is much more predictable than correlation. 813 00:42:10,131 --> 00:42:11,931 And I think that's kind of what Pedro is saying. 814 00:42:11,931 --> 00:42:13,091 And that's true, that's correct. 815 00:42:14,131 --> 00:42:18,691 But the other thing is that the correlation is still pretty predictable. 816 00:42:20,211 --> 00:42:22,251 So if, if you want some hard numbers. 817 00:42:22,251 --> 00:42:27,411 So off the top of my head, if you do a regression of next month's 818 00:42:27,411 --> 00:42:30,931 volatility on last month's volatility, you get an R squared 819 00:42:30,931 --> 00:42:32,131 of about 0.35. 820 00:42:32,371 --> 00:42:35,971 If you don't know what R squared is, don't worry, but 0.35 821 00:42:35,971 --> 00:42:36,691 is pretty good. 822 00:42:36,891 --> 00:42:37,931 Okay, that's pretty good. 823 00:42:38,971 --> 00:42:42,811 For correlation you get something more like 0.2 or 0.25, 824 00:42:43,691 --> 00:42:45,051 which isn't quite as good. 825 00:42:45,211 --> 00:42:48,251 But if you were to try and do, to try and predict, you know, means, 826 00:42:48,331 --> 00:42:51,451 which is the other moment of the distribution, you would, would, 827 00:42:51,451 --> 00:42:52,211 you'd get nothing. 828 00:42:52,211 --> 00:42:53,451 You guys get noise. 829 00:42:53,451 --> 00:42:56,411 So that it's, it's, these things are relatively predictable. 830 00:42:57,611 --> 00:43:01,851 Now the other thing about correlation is the, as I said, you 831 00:43:01,851 --> 00:43:04,531 probably only use something like about a six month rolling estimate 832 00:43:04,531 --> 00:43:06,741 versus about a one month rolling estimate for volume. 833 00:43:07,451 --> 00:43:11,091 So the kind of correlation structure changes more slowly than 834 00:43:11,091 --> 00:43:12,811 the volatility structure changes. 835 00:43:13,531 --> 00:43:17,211 So what I personally do and what I recommend is to, yes, is to 836 00:43:17,211 --> 00:43:18,731 actually estimate these things separately. 837 00:43:20,171 --> 00:43:24,331 If you are using the correlation, the volatilities for 838 00:43:24,891 --> 00:43:27,971 this scaling thing that we've talked about already, well then the 839 00:43:27,971 --> 00:43:31,931 things you're actually playing with should already be at the same 840 00:43:31,931 --> 00:43:32,851 target volume. 841 00:43:32,851 --> 00:43:35,611 So you don't actually need to estimate a volatility estimate. 842 00:43:36,561 --> 00:43:39,281 Obviously you will still need your volatility estimate for position 843 00:43:39,281 --> 00:43:43,481 sizing, as we've already discussed, but for actual sort of 844 00:43:43,481 --> 00:43:47,401 risk management kind of portfolio construction part, the 845 00:43:47,401 --> 00:43:51,201 nice thing about doing things in a volume targeted way is you don't 846 00:43:51,201 --> 00:43:52,201 need to estimate voles. 847 00:43:52,201 --> 00:43:55,321 Again, you just assume that your volume targeting works, which 848 00:43:55,321 --> 00:43:57,281 is a first order approximation is fine. 849 00:43:57,521 --> 00:44:00,001 Then you just focus on your correlations and then you can focus 850 00:44:00,001 --> 00:44:02,561 on estimating your correlations in the best possible 851 00:44:02,561 --> 00:44:02,961 way. 852 00:44:04,291 --> 00:44:06,891 So it's quite technical, quite a technical answer, but it's quite 853 00:44:06,891 --> 00:44:07,531 a technical question. 854 00:44:07,531 --> 00:44:10,091 And I think if you understand the Question, you'll understand the 855 00:44:10,091 --> 00:44:10,611 answer. 856 00:44:11,011 --> 00:44:13,331 If you don't, then don't worry too much about it. 857 00:44:13,411 --> 00:44:15,211 Yeah, no, perfect, Absolutely. 858 00:44:15,211 --> 00:44:16,851 Thank you for doing that. 859 00:44:16,931 --> 00:44:20,811 Okay, well, as we just talked about predictability, and there is 860 00:44:20,811 --> 00:44:24,091 another thing that's very predictable and that is when we do 861 00:44:24,091 --> 00:44:28,171 review papers, which we do very often on the podcast, it's very 862 00:44:28,171 --> 00:44:32,451 predictable to say that it's most likely from either MAN or aqr. 863 00:44:32,451 --> 00:44:37,211 And yet again, here we are now in the interest of time, I don't 864 00:44:37,211 --> 00:44:40,131 know if we will have time for both because I think the first one 865 00:44:40,131 --> 00:44:41,931 we're going to be doing is the man paper. 866 00:44:42,011 --> 00:44:43,531 It's a super good paper. 867 00:44:44,171 --> 00:44:47,211 I'm going to try and remember to put a link to it. 868 00:44:48,091 --> 00:44:54,171 But this is a paper about not really about better models or faster 869 00:44:54,171 --> 00:44:54,651 signals. 870 00:44:54,651 --> 00:44:58,981 It's really about what markets managers, you know, kind of choose 871 00:44:58,981 --> 00:45:03,341 to trade and therefore what kind of trend following they are 872 00:45:03,341 --> 00:45:03,821 running. 873 00:45:03,821 --> 00:45:09,221 But I'm gonna let you take us into the paper and I'm gonna follow 874 00:45:09,221 --> 00:45:12,461 along and see if I have any, any thoughts along the way. 875 00:45:12,461 --> 00:45:14,621 But yeah, I'm gonna turn it over to you. 876 00:45:14,621 --> 00:45:15,741 No, it's a very good paper. 877 00:45:15,741 --> 00:45:18,541 It's actually almost maybe three papers in one, to be honest. 878 00:45:18,541 --> 00:45:20,181 So there's an awful lot of content here. 879 00:45:21,061 --> 00:45:23,681 Now, the is thing, it's been written by three people and I have 880 00:45:23,681 --> 00:45:27,241 to say it's now been so long since I've been away from AHL that 881 00:45:27,241 --> 00:45:28,521 I don't know any of these people are. 882 00:45:29,401 --> 00:45:30,601 I've never met any of them. 883 00:45:30,601 --> 00:45:34,081 But so that, that also means I shouldn't be biased in loving this 884 00:45:34,081 --> 00:45:36,401 paper because it's not like it's been written by an old friend 885 00:45:36,401 --> 00:45:38,721 of mine who's going to buy me a beer in exchange for mentioning 886 00:45:38,721 --> 00:45:38,961 it. 887 00:45:38,961 --> 00:45:40,281 So be reassured. 888 00:45:41,561 --> 00:45:45,801 So the, I guess the main thing that they sort of do in this paper 889 00:45:45,801 --> 00:45:48,641 is say, well, actually there are kind of different kinds of markets, 890 00:45:48,641 --> 00:45:48,921 right. 891 00:45:49,611 --> 00:45:53,171 And we've already discussed the fact that different markets have 892 00:45:53,171 --> 00:45:54,971 different levels of liquidity, for example. 893 00:45:56,571 --> 00:46:00,091 And what they say is, well, we're going to divide our market 894 00:46:00,091 --> 00:46:02,571 universe into three categories, traditional markets, 895 00:46:02,731 --> 00:46:06,091 alternative markets, and alternative esoteric markets. 896 00:46:07,131 --> 00:46:10,891 And I guess, although a lot of that is liquidity. 897 00:46:12,011 --> 00:46:15,171 So then they've got this really nice graph. 898 00:46:15,171 --> 00:46:18,871 It's Figure 5, if you're following along at home, which shows 899 00:46:18,871 --> 00:46:24,271 liquidity on one axis and complexity on the other axis. 900 00:46:24,431 --> 00:46:28,071 So esoteric markets may be less liquid, but not necessarily 901 00:46:28,071 --> 00:46:28,271 so. 902 00:46:28,271 --> 00:46:31,711 One of their esoteric markets they've labeled as China. 903 00:46:32,751 --> 00:46:36,311 I think they might mean futures markets in China rather than 904 00:46:36,311 --> 00:46:37,071 the whole country. 905 00:46:37,631 --> 00:46:40,831 Another one is physical cryptocurrencies, many of which are 906 00:46:41,231 --> 00:46:44,961 relatively liquid, of course, but those are quite high on the complexity 907 00:46:44,961 --> 00:46:45,561 axis. 908 00:46:46,281 --> 00:46:49,401 Whereas onto traditional markets they put cotton futures, 909 00:46:50,201 --> 00:46:55,121 which they put low complexity, because here I guess complexity means 910 00:46:55,121 --> 00:46:55,961 more like a future. 911 00:46:57,241 --> 00:47:00,321 But they have actually on their graph those are less liquid 912 00:47:00,321 --> 00:47:03,961 than China and physical cryptocurrencies. 913 00:47:03,961 --> 00:47:09,241 So it's a bit richer than the normal categorization around liquidity 914 00:47:09,241 --> 00:47:12,401 or around, you know, are these futures not futures? 915 00:47:12,401 --> 00:47:13,121 They look at both. 916 00:47:14,561 --> 00:47:18,721 But then what they do is say, well, the interesting thing about 917 00:47:20,161 --> 00:47:24,401 these different markets is kind of where the money comes from. 918 00:47:25,601 --> 00:47:29,121 Now, finance people love doing something called a factor decomposition. 919 00:47:29,121 --> 00:47:33,201 And basically what a factor decomposition does is say, well, 920 00:47:33,441 --> 00:47:38,951 given, you know that I'm making some money from this portfolio, 921 00:47:39,271 --> 00:47:42,951 can I identify sort of where the key sources of those returns 922 00:47:42,951 --> 00:47:43,351 are from? 923 00:47:44,871 --> 00:47:49,271 And in trend following, one thing you can do is to say, well, 924 00:47:49,991 --> 00:47:54,071 it looks like if you do something like a PCA or principle 925 00:47:54,071 --> 00:47:56,911 component analysis, it looks like a lot of the return in my trend 926 00:47:56,911 --> 00:47:59,831 following portfolio is actually coming from a kind of risk 927 00:47:59,831 --> 00:48:02,881 on, risk off factor, which I'm trend following. 928 00:48:03,041 --> 00:48:09,161 So, you know, if you look at 2008, for example, CTAs made a lot 929 00:48:09,161 --> 00:48:09,561 of money. 930 00:48:09,561 --> 00:48:11,041 Bonds went up, equities went down. 931 00:48:11,041 --> 00:48:13,601 That was a perfect example of that factor in action. 932 00:48:15,041 --> 00:48:18,121 And then you basically, then what you can do is say, well, what's 933 00:48:18,121 --> 00:48:18,481 left over? 934 00:48:18,481 --> 00:48:21,201 What I'll get left over that isn't explained by that factor return. 935 00:48:21,521 --> 00:48:24,241 And they call that the idiosyncratic return. 936 00:48:25,601 --> 00:48:28,361 So you could argue that, you know, the returns we saw in January, 937 00:48:28,361 --> 00:48:31,761 where as we've said, bonds have gone nowhere, equities slightly 938 00:48:31,761 --> 00:48:35,721 up, you know, February is looking a bit different maybe, but 939 00:48:35,721 --> 00:48:38,601 financial markets generally didn't, you know, that kind of first 940 00:48:38,601 --> 00:48:40,241 principle component didn't contribute much. 941 00:48:40,241 --> 00:48:43,641 Well, so maybe it was more something idiosyncratic around gold 942 00:48:43,641 --> 00:48:47,561 and silver, you know, or something else, or in natural gas. 943 00:48:48,841 --> 00:48:53,601 So what they do is say, well, if we look at this idea of decomposing 944 00:48:53,601 --> 00:48:58,031 into, you know, the main things explaining a portfolio return 945 00:48:58,031 --> 00:49:01,111 and the idiosyncratic return, the key thing that's interesting 946 00:49:01,111 --> 00:49:04,751 about these alternative and esoteric markets is that a much bigger 947 00:49:04,911 --> 00:49:08,311 proportion of their returns are idiosyncratic. 948 00:49:08,311 --> 00:49:13,071 They're not coming from these big factors and in hard numbers. 949 00:49:14,031 --> 00:49:18,351 So roughly Speaking, in traditional CTAs, about half the 950 00:49:18,351 --> 00:49:21,871 returns are from factors and about half of it is syncretic. 951 00:49:22,361 --> 00:49:27,241 Actually, it's probably more like 45, 55, 45% idiosyncratic, 55% 952 00:49:27,241 --> 00:49:27,801 factors. 953 00:49:29,001 --> 00:49:32,601 But in these unusual markets, these alternative and esoteric markets, 954 00:49:32,761 --> 00:49:34,921 it's more like 1/3, 2/3. 955 00:49:35,321 --> 00:49:38,641 So only about one third is coming from the main factors and 956 00:49:38,641 --> 00:49:40,121 2/3 is idiosyncratic. 957 00:49:40,441 --> 00:49:43,921 So the main thing about alternative markets is they give 958 00:49:43,921 --> 00:49:47,801 you an exposure to weird stuff, to weird sources of return. 959 00:49:49,421 --> 00:49:53,741 Now, if you're doing any kind of portfolio construction, you like 960 00:49:53,741 --> 00:49:54,901 weird stuff, okay? 961 00:49:54,901 --> 00:49:58,421 You like diversifications we've discussed, you like idiosyncratic 962 00:49:58,421 --> 00:49:58,941 returns. 963 00:49:59,501 --> 00:50:03,621 If the only returns in your portfolio are coming from trend following, 964 00:50:03,621 --> 00:50:11,141 a kind of equity risk on, risk off return, that's going to be not 965 00:50:11,141 --> 00:50:12,381 very helpful, right? 966 00:50:12,941 --> 00:50:15,141 You want to be able to be trend following other things as well, 967 00:50:15,141 --> 00:50:17,701 getting credit returns from lots of different places. 968 00:50:18,181 --> 00:50:21,061 So what that means is in a completely unconstrained portfolio, 969 00:50:21,141 --> 00:50:24,301 if you do a standard portfolio optimization, you're going to want 970 00:50:24,301 --> 00:50:27,861 to have quite a lot of these alternative and esoteric markets. 971 00:50:27,941 --> 00:50:32,141 And of course this is one of the common themes of, I think ever 972 00:50:32,141 --> 00:50:35,261 since I've been talking to you Niels, alternative markets, which 973 00:50:35,261 --> 00:50:40,411 obviously has been a big thing for ahl, but also for there are people 974 00:50:40,411 --> 00:50:42,731 like Florincort, for example, there's a very successful CTA that 975 00:50:42,731 --> 00:50:44,291 focused just on alternative markets. 976 00:50:44,531 --> 00:50:48,411 And most big CTAs have gone into alternative markets to a greater 977 00:50:48,411 --> 00:50:49,251 or lesser degree. 978 00:50:50,371 --> 00:50:54,411 So this is kind of something that's not a surprise that we already 979 00:50:54,411 --> 00:50:54,931 know about. 980 00:50:55,891 --> 00:51:00,051 And if they do this kind of sort of standard portfolio optimization, 981 00:51:00,611 --> 00:51:05,901 they find that only about 30% of their portfolio needs to should 982 00:51:05,901 --> 00:51:10,181 be going into these traditional markets, which isn't 983 00:51:10,181 --> 00:51:10,861 a lot, right? 984 00:51:11,661 --> 00:51:14,141 And just to emphasize again, this is unconstrained. 985 00:51:14,141 --> 00:51:17,221 So this, this assumes you've got no issues with liquidity and 986 00:51:17,221 --> 00:51:19,981 you can put as much into the weird stuff as you as you possibly 987 00:51:19,981 --> 00:51:25,501 want, which a big CTF course can't do, right, because they're 988 00:51:25,501 --> 00:51:26,061 just too big. 989 00:51:26,461 --> 00:51:31,701 And then they have the rest of that portfolio, the 70% is in a mixture 990 00:51:31,701 --> 00:51:34,851 of alternative and, you know, alternative esoteric. 991 00:51:34,851 --> 00:51:37,811 And there's a bit of a breakdown and I can see that they 992 00:51:37,811 --> 00:51:40,091 put 5% into crypto, for example. 993 00:51:40,091 --> 00:51:43,451 So the crypto bulls would love that, I'm sure. 994 00:51:44,491 --> 00:51:50,331 Now one question that again we keep returning to is why should we 995 00:51:50,331 --> 00:51:51,291 invest in CTOs? 996 00:51:51,291 --> 00:51:52,491 Why should we invest in trend following? 997 00:51:55,211 --> 00:51:58,171 And it depends, I think I've mentioned this before, it depends 998 00:51:58,171 --> 00:52:03,031 on whether you are looking at it as a standalone product or as 999 00:52:03,031 --> 00:52:07,071 something that goes alongside an existing 6040 or some other traditional 1000 00:52:07,071 --> 00:52:07,671 kind of Assets. 1001 00:52:07,671 --> 00:52:12,351 Right now, if you're doing as a standalone product, you want, you 1002 00:52:12,351 --> 00:52:15,511 know, to the first order approximation, you want a maximum 1003 00:52:15,511 --> 00:52:16,391 Sharpe ratio. 1004 00:52:16,551 --> 00:52:19,751 You'll do this kind of standard portfolio portfolio optimization. 1005 00:52:20,231 --> 00:52:23,711 And so as a standalone product, as a cto, without constraints 1006 00:52:23,711 --> 00:52:26,651 around liquidity, what you would do is offer something that 1007 00:52:26,651 --> 00:52:30,571 was 30% traditional, 70% alternatives, and that would be the 1008 00:52:30,571 --> 00:52:32,051 best product for the investor to buy. 1009 00:52:32,851 --> 00:52:35,571 But most people of course, don't aren't doing that. 1010 00:52:35,571 --> 00:52:39,091 Most people are putting a slice of trend following into their 1011 00:52:39,091 --> 00:52:40,451 6040 portfolio. 1012 00:52:40,851 --> 00:52:43,931 We'd like it to be a bigger slice, of course, but you know, that's 1013 00:52:43,931 --> 00:52:44,211 life. 1014 00:52:44,851 --> 00:52:47,491 And that means what they're looking for is an element of tail 1015 00:52:47,491 --> 00:52:48,051 protection. 1016 00:52:49,011 --> 00:52:52,051 And it's a pity we don't have Katie with us because of course she 1017 00:52:52,051 --> 00:52:52,611 wrote the book. 1018 00:52:54,521 --> 00:52:58,361 But they're looking for some kind of crisis alpha tail protection, 1019 00:52:58,361 --> 00:52:59,401 whatever you want to call it. 1020 00:53:00,121 --> 00:53:05,241 So what these guys did was say, well, what if we, instead of 1021 00:53:05,241 --> 00:53:08,321 looking at a portfolio optimization that basically just 1022 00:53:08,321 --> 00:53:12,601 treats all states of the world the same, let's focus in on periods 1023 00:53:12,601 --> 00:53:14,121 when equity markets fell. 1024 00:53:15,081 --> 00:53:19,421 Because what we want to do is find the best portfolio for that 1025 00:53:19,661 --> 00:53:20,381 scenario. 1026 00:53:20,781 --> 00:53:23,941 Because that's when people really want to benefit from what 1027 00:53:23,941 --> 00:53:25,421 CTA performance is like. 1028 00:53:26,141 --> 00:53:32,861 So they constructed their correlation matrix differently to 1029 00:53:32,861 --> 00:53:33,661 account for that. 1030 00:53:34,300 --> 00:53:40,301 And what they found was that the traditional portfolio, the traditional 1031 00:53:40,301 --> 00:53:43,021 trend for things you trend for like so the liquid. 1032 00:53:43,021 --> 00:53:44,141 Markets, let's call them. 1033 00:53:44,461 --> 00:53:47,641 Well remember Niels, it's not just about liquidity, it's about, 1034 00:53:47,721 --> 00:53:48,281 it's about. 1035 00:53:48,361 --> 00:53:52,721 So basically it's kind of, it's futures, traditional futures 1036 00:53:52,721 --> 00:53:55,961 markets that a CTA would have had in it 30 years ago before people 1037 00:53:55,961 --> 00:54:00,601 started doing all this weird stuff like trading German Power and 1038 00:54:00,761 --> 00:54:02,241 CBS and all this kind of stuff. 1039 00:54:02,241 --> 00:54:03,161 Right, exactly. 1040 00:54:03,321 --> 00:54:07,041 So it's things that are relatively liquid, but also basically 1041 00:54:07,041 --> 00:54:07,521 futures. 1042 00:54:07,521 --> 00:54:09,721 And therefore for a cta, relatively simple. 1043 00:54:10,041 --> 00:54:12,281 But more importantly, it's things that have a. 1044 00:54:13,011 --> 00:54:18,051 Whose returns are mostly being driven by this factor risk rather 1045 00:54:18,051 --> 00:54:19,331 than this idiosyncratic risk. 1046 00:54:19,411 --> 00:54:20,931 That's the key point about this article. 1047 00:54:20,931 --> 00:54:24,131 That's the key innovation in how they categorize markets. 1048 00:54:24,131 --> 00:54:29,531 I think now it turns out that actually what you want to do in the 1049 00:54:29,531 --> 00:54:33,011 crisis situation is not have 30% traditional markets. 1050 00:54:33,091 --> 00:54:35,251 You want to have 50% traditional markets. 1051 00:54:35,331 --> 00:54:38,251 And then obviously the non traditional stuff gets squeezed down 1052 00:54:38,251 --> 00:54:39,231 to a smaller fraction. 1053 00:54:40,101 --> 00:54:43,621 And the intuition behind that, and which they explain is because 1054 00:54:44,421 --> 00:54:50,501 it's that total factor return, that thing that explains most of 1055 00:54:50,501 --> 00:54:53,421 the risk of traditional markets, that makes traditional markets 1056 00:54:53,421 --> 00:54:53,981 look bad. 1057 00:54:53,981 --> 00:54:57,021 From a kind of generic optimization perspective, because 1058 00:54:57,021 --> 00:54:58,741 they haven't got much of idiosyncratic risk. 1059 00:54:58,981 --> 00:55:01,221 It turns out in a crisis, that's what you want. 1060 00:55:01,221 --> 00:55:03,141 Okay, that's what you want. 1061 00:55:03,541 --> 00:55:08,321 And this kind of makes intuitive sense because in a crisis 1062 00:55:09,201 --> 00:55:15,681 you basically want to just make fairly big bets on going 2008, 1063 00:55:15,761 --> 00:55:17,681 going short, equities, long bonds. 1064 00:55:17,841 --> 00:55:23,041 You're not really going to need your fancy other weird markets 1065 00:55:23,121 --> 00:55:25,361 there to help you. 1066 00:55:26,001 --> 00:55:30,481 And of course, there is a risk that many of those markets might 1067 00:55:30,481 --> 00:55:34,691 have issues around liquidity, which is another thing that they 1068 00:55:34,691 --> 00:55:38,651 talk about next or around, for example, whether you can go short 1069 00:55:39,851 --> 00:55:42,411 because a lot of these things in a crisis you do want to go short 1070 00:55:42,491 --> 00:55:47,691 because a lot of them trade light risk on assets because things 1071 00:55:47,691 --> 00:55:49,611 like, well, we talked about cryptocurrency. 1072 00:55:50,091 --> 00:55:51,931 Cryptocurrency for example, does that. 1073 00:55:53,691 --> 00:55:55,811 So that's kind of the interesting thing. 1074 00:55:55,811 --> 00:56:00,481 Now they then take this a step further and say, well, all this time 1075 00:56:03,121 --> 00:56:06,241 we've been kind of assuming that we're completely unconstrained. 1076 00:56:07,441 --> 00:56:12,921 But actually another fantastic thing about CTAs or about anything 1077 00:56:12,921 --> 00:56:17,761 that trades futures is the fact that they're very cash efficient. 1078 00:56:19,041 --> 00:56:24,601 So if I look at my own futures account, which has a kind of a relatively 1079 00:56:24,601 --> 00:56:30,441 high risk target for, by institutional standards, some people 1080 00:56:30,441 --> 00:56:34,201 are a bit crazy, of course, as we discussed, but I'm probably running 1081 00:56:34,201 --> 00:56:37,081 at about one and a half times the volume of say AHL is on most 1082 00:56:37,081 --> 00:56:37,761 of their funds. 1083 00:56:38,401 --> 00:56:41,201 I'm only using 30% of my cash as margin. 1084 00:56:41,921 --> 00:56:46,001 So a typical CTA, that number's 20% and that's super cash 1085 00:56:46,001 --> 00:56:46,561 efficient. 1086 00:56:47,041 --> 00:56:50,001 And that makes them very attractive from a structuring perspective. 1087 00:56:51,521 --> 00:56:59,251 So the next thing they did was say, well, let's say we want to go 1088 00:56:59,251 --> 00:57:02,291 a bit further and say, well, actually we want to kind of maintain 1089 00:57:02,291 --> 00:57:03,611 this cash efficiency. 1090 00:57:04,091 --> 00:57:07,291 Now the non futures markets obviously aren't as cash efficient. 1091 00:57:08,331 --> 00:57:10,771 Most of them have margin requirements that are higher than 1092 00:57:10,771 --> 00:57:11,011 future. 1093 00:57:11,011 --> 00:57:12,971 Well, all of them have margin requirements that are higher than 1094 00:57:12,971 --> 00:57:13,531 futures. 1095 00:57:13,771 --> 00:57:17,371 Some of them even have 100% margin in the sense that you just, 1096 00:57:17,451 --> 00:57:18,531 you can't margin them. 1097 00:57:18,531 --> 00:57:20,611 You have to just put, put up cash essentially. 1098 00:57:21,971 --> 00:57:23,971 And if they do that again, it's not a surprise. 1099 00:57:23,971 --> 00:57:28,011 But then what happens is we get even more into traditional markets. 1100 00:57:28,011 --> 00:57:34,051 So we end up with 70% in traditional markets and just 30% 1101 00:57:35,011 --> 00:57:37,411 in the alternative markets. 1102 00:57:38,131 --> 00:57:44,771 And one final point to make is that when is cash efficiency most 1103 00:57:44,771 --> 00:57:47,491 important is when we need cash. 1104 00:57:48,211 --> 00:57:49,411 When do we need cash? 1105 00:57:50,371 --> 00:57:51,571 When the world's ending. 1106 00:57:52,131 --> 00:58:00,531 So in 2008, the fact that CTAs had plenty of cash daily liquidity, 1107 00:58:00,851 --> 00:58:05,331 no gating meant that many people use them as cash machines. 1108 00:58:06,691 --> 00:58:10,331 So we have this weird situation where although the industry 1109 00:58:10,331 --> 00:58:13,071 made a lot of money at the same time, people were withdrawing 1110 00:58:13,071 --> 00:58:15,791 it because it was a source of ready cash that wasn't available 1111 00:58:15,791 --> 00:58:16,431 elsewhere. 1112 00:58:18,911 --> 00:58:20,271 So that's the article. 1113 00:58:20,271 --> 00:58:23,071 I mean so there's a lot of really interesting things there and 1114 00:58:24,511 --> 00:58:28,751 like my many AHL articles, it's very well written, not technical, 1115 00:58:29,071 --> 00:58:32,911 lots of nice graphs, very well presented and yeah for me at least 1116 00:58:32,911 --> 00:58:36,151 some really interesting innovations around a few different 1117 00:58:36,151 --> 00:58:36,671 topics. 1118 00:58:38,312 --> 00:58:39,352 Yeah, no, I couldn't agree more. 1119 00:58:39,352 --> 00:58:42,592 People should definitely go and download it and read it and as 1120 00:58:42,592 --> 00:58:46,392 I said, I'll try to remember put in the link in the show notes. 1121 00:58:46,392 --> 00:58:50,192 But I just want to kind of summarize all of this just to make 1122 00:58:50,192 --> 00:58:53,032 sure people really get the importance of this. 1123 00:58:53,832 --> 00:58:58,632 And that is if you were just kind of trying to maximize and looking 1124 00:58:58,632 --> 00:59:02,872 at your trend portfolio in isolation and you say oh, what could 1125 00:59:02,872 --> 00:59:03,552 be what? 1126 00:59:03,552 --> 00:59:07,182 You know, what kind of market portfolio should I go for when I'm 1127 00:59:07,182 --> 00:59:13,542 looking to maximize my, my long term sharp of, of that in all 1128 00:59:13,542 --> 00:59:14,102 periods? 1129 00:59:14,102 --> 00:59:16,982 Well, you would go for a portfolio that probably trades more 1130 00:59:16,982 --> 00:59:17,542 markets. 1131 00:59:17,542 --> 00:59:20,982 That's kind of what they, they conclude. 1132 00:59:21,222 --> 00:59:25,622 However, as Rob rightly say, a lot of investors, and certainly I 1133 00:59:25,622 --> 00:59:28,982 would say most investors that I come across, they include trend 1134 00:59:28,982 --> 00:59:32,262 following and CTA specifically because they want to have something 1135 00:59:32,262 --> 00:59:34,022 that really does well for them. 1136 00:59:34,352 --> 00:59:37,232 When equity markets are not doing well or when bond markets are 1137 00:59:37,232 --> 00:59:40,952 not doing well and in their work they find that then you need 1138 00:59:40,952 --> 00:59:44,512 to find managers who are more focused on the traditional markets, 1139 00:59:44,512 --> 00:59:47,992 the more liquid markets, the classical futures markets we've been 1140 00:59:47,992 --> 00:59:52,272 trading for the last 30, 40 years and go, we want a manager that 1141 00:59:52,432 --> 00:59:53,072 does that. 1142 00:59:53,232 --> 00:59:56,352 And then on top of that, and I think I completely agree, this is 1143 00:59:56,352 --> 01:00:01,082 the interesting point they make and that is if you then also 1144 01:00:01,162 --> 01:00:06,882 are an investor where you might need to get some cash out in 1145 01:00:06,882 --> 01:00:10,322 terms of a crisis, if your private equity fund is not giving 1146 01:00:10,322 --> 01:00:14,522 you any money or drawing down or whatever comes during a crisis, 1147 01:00:14,762 --> 01:00:18,442 then you need to be even more focused on managers that trades the 1148 01:00:18,842 --> 01:00:24,842 classical liquid futures markets because the cash efficiency 1149 01:00:25,412 --> 01:00:30,812 is important and, and so I thought it was really well done as 1150 01:00:30,812 --> 01:00:35,892 well and, and to make that point the way they did so that everybody 1151 01:00:35,892 --> 01:00:38,972 kind of understand the difference between the choices we 1152 01:00:38,972 --> 01:00:40,372 all have to make as managers. 1153 01:00:40,692 --> 01:00:43,452 So it's not really all about kind of returns. 1154 01:00:43,452 --> 01:00:48,452 It's also about, you know, what's the best manager fit for me, 1155 01:00:49,172 --> 01:00:53,872 in terms of what my objective is with investing with the cta, that 1156 01:00:53,872 --> 01:00:57,472 should lead you down the path of identifying a smaller number of 1157 01:00:57,472 --> 01:01:00,472 managers that might be relevant for, for, for, for you. 1158 01:01:00,472 --> 01:01:02,592 So really good paper. 1159 01:01:03,312 --> 01:01:06,432 Kudos to the team for, for writing that. 1160 01:01:06,912 --> 01:01:09,792 The next paper, which we won't have time to do justice. 1161 01:01:09,792 --> 01:01:11,232 So we'll, we'll wait with that. 1162 01:01:11,552 --> 01:01:15,152 But it's also a super interesting paper. 1163 01:01:15,792 --> 01:01:19,992 We'll see if we wait until Rob comes back or maybe we'll do it beforehand. 1164 01:01:19,992 --> 01:01:24,662 But it's something that is also super relevant in the current 1165 01:01:24,662 --> 01:01:29,102 environment and what's going on in some of the, in some of the 1166 01:01:29,102 --> 01:01:29,622 markets. 1167 01:01:29,862 --> 01:01:34,422 I don't know if you want to spend five minutes, Rob, it's your 1168 01:01:34,422 --> 01:01:39,222 call on the Financial Times article you sent me this morning. 1169 01:01:39,222 --> 01:01:43,222 I have not read it myself, so I don't know what's in it, but please 1170 01:01:43,222 --> 01:01:45,702 do if you want to spend five, seven minutes on that. 1171 01:01:45,862 --> 01:01:50,772 Yeah, so the articles on the FT, AlphaVal, AlphaVille, part, the 1172 01:01:50,772 --> 01:01:54,132 FT, which is important because that's actually in front of the paywall, 1173 01:01:54,132 --> 01:01:55,132 so anyone can read it. 1174 01:01:56,412 --> 01:01:59,532 And it's called the Quant Shop, AI Lab Convergence. 1175 01:02:01,132 --> 01:02:06,332 And it's kind of interesting because basically what they do in 1176 01:02:06,332 --> 01:02:10,652 the article is they draw parallels between kind of finance 1177 01:02:11,052 --> 01:02:15,692 quant and AI quant, for want of a better word. 1178 01:02:17,272 --> 01:02:20,312 And their sort of thesis is that basically these people are becoming 1179 01:02:20,312 --> 01:02:21,512 more and more like each other. 1180 01:02:22,072 --> 01:02:27,352 Now, obviously there's always been a crossover between the sort 1181 01:02:27,352 --> 01:02:31,832 of worlds of finance, quant and, you know, software engineering 1182 01:02:32,712 --> 01:02:37,912 more generally and kind of, you know, so, and obviously a lot 1183 01:02:37,912 --> 01:02:43,282 of places are now using, well, let's be careful. 1184 01:02:43,922 --> 01:02:49,202 So in finance, in systematic finance, we've always used something 1185 01:02:49,202 --> 01:02:51,922 that I suppose you could wave your hands and call machine learning. 1186 01:02:53,042 --> 01:02:56,402 Now, machine learning to some people has to be very complicated. 1187 01:02:56,402 --> 01:03:01,842 But for example, if you're running something like a simple regression, 1188 01:03:01,842 --> 01:03:05,922 which is a very simple, a relatively simple way of working 1189 01:03:05,922 --> 01:03:10,082 out, for example, as we've just done, you know, how much money 1190 01:03:10,082 --> 01:03:11,602 you should give to a particular market. 1191 01:03:12,162 --> 01:03:15,522 And you could use a regression to do that, or you could use an optimization. 1192 01:03:15,922 --> 01:03:17,842 Those are forms of machine learning. 1193 01:03:19,042 --> 01:03:21,722 They're forms of what you might call supervised machine learning 1194 01:03:21,722 --> 01:03:24,762 because the, we're sort of telling the computer what to do and 1195 01:03:24,762 --> 01:03:26,642 then it's giving us the numbers that come out. 1196 01:03:28,162 --> 01:03:31,722 So that, that's, you know, all the systematic finance is machine 1197 01:03:31,722 --> 01:03:32,002 learning. 1198 01:03:32,002 --> 01:03:34,972 Although some, some, you know, some people say, oh, no, that can't 1199 01:03:34,972 --> 01:03:37,892 be machine learning because it's, it's too simple, but, but you 1200 01:03:37,892 --> 01:03:40,412 know, then they can never tell me where the boundary is. 1201 01:03:40,412 --> 01:03:42,252 So I say well no, it must be machine learning. 1202 01:03:43,772 --> 01:03:46,572 Now then, then you have something which you might call unsupervised 1203 01:03:46,572 --> 01:03:50,492 machine learning and that kind of then blends into our AI and LLMs 1204 01:03:50,492 --> 01:03:55,852 and Neural Networks and all this kind of stuff and the, the sort 1205 01:03:55,852 --> 01:03:56,012 of. 1206 01:03:56,012 --> 01:04:01,532 I think there's certainly a lot of people trying to use AI and 1207 01:04:01,532 --> 01:04:03,892 that goes for every industry of course, not ours. 1208 01:04:04,932 --> 01:04:10,932 I think pretty much every fund of a decent size will have some kind 1209 01:04:10,932 --> 01:04:13,012 of AI thing going on. 1210 01:04:13,892 --> 01:04:17,732 And how much of that there is is an open question. 1211 01:04:17,732 --> 01:04:19,732 How much of it is just a marketing thing? 1212 01:04:19,972 --> 01:04:21,092 Is an open question. 1213 01:04:21,412 --> 01:04:25,172 How much of it's kind of blue sky research, like a, you know, let's 1214 01:04:25,172 --> 01:04:27,612 buy a call option on this technology and you never know, we 1215 01:04:27,612 --> 01:04:31,592 might find something that no one else can do, is an open question. 1216 01:04:31,592 --> 01:04:34,472 I think the people who are doing it best are ones who are using 1217 01:04:34,472 --> 01:04:36,232 it in very tightly defined areas. 1218 01:04:37,912 --> 01:04:41,312 So things like for example, execution research, I think that's 1219 01:04:41,312 --> 01:04:44,552 where it's potentially more interesting because that's where 1220 01:04:44,552 --> 01:04:50,272 something like well first of all you got a lot of data, so we're 1221 01:04:50,272 --> 01:04:53,592 building our kind of normal models with 50 odd years of daily 1222 01:04:53,592 --> 01:04:53,912 data. 1223 01:04:55,352 --> 01:05:00,952 And in execution research you've got tick data which is obviously 1224 01:05:01,112 --> 01:05:03,352 much richer, much faster, there's much more of it. 1225 01:05:03,672 --> 01:05:06,432 So it's much more suited to something like building any kind 1226 01:05:06,432 --> 01:05:09,512 of nonlinear analysis which would include something like training 1227 01:05:09,512 --> 01:05:11,512 and LLM on that data. 1228 01:05:13,032 --> 01:05:16,552 All that aside though, the article is quite interesting because 1229 01:05:16,552 --> 01:05:21,192 rather than just making the obvious point that well, everything's 1230 01:05:21,192 --> 01:05:24,792 AI now, which is a slightly lazy argument. 1231 01:05:25,192 --> 01:05:28,722 And I do think they do, and this is probably because they're 1232 01:05:28,722 --> 01:05:32,042 talking their own book, they do slightly overstate, I think the 1233 01:05:32,042 --> 01:05:37,202 amount of AI that sort of is going on, but they do kind of go 1234 01:05:37,202 --> 01:05:39,842 further and say, well if you actually look at the kind of workflow 1235 01:05:40,402 --> 01:05:45,082 in quant finance and the workflow in AI that's similar, if 1236 01:05:45,082 --> 01:05:49,242 you look at the sorts of people that are being hired, if you 1237 01:05:49,242 --> 01:05:54,482 look at the technology stack, they even say, well if you look at 1238 01:05:54,562 --> 01:05:59,062 the way that kind of people are employed, so you know, we're 1239 01:05:59,062 --> 01:06:03,462 used in finance to sort of non competes and gardening leave and 1240 01:06:03,462 --> 01:06:07,142 things like that and signing on bonuses, well they've got that 1241 01:06:07,542 --> 01:06:08,382 in AI now. 1242 01:06:08,382 --> 01:06:11,702 And you know, you hear stories of people being offered kind of nine 1243 01:06:11,702 --> 01:06:17,222 figure sums to move from meta to move from OpenAI to meta. 1244 01:06:17,222 --> 01:06:21,982 So actually, you know, many of the top AI researchers are making 1245 01:06:21,982 --> 01:06:26,022 far more money than any person quant in finances sadly I have to 1246 01:06:26,022 --> 01:06:26,262 say. 1247 01:06:28,202 --> 01:06:31,962 So yeah, it's quite an interesting article. 1248 01:06:32,122 --> 01:06:33,602 I am a skeptic of this stuff. 1249 01:06:33,602 --> 01:06:36,482 I've said it before, I've said it again, but I still found it an 1250 01:06:36,482 --> 01:06:38,602 interesting article so it must have been good. 1251 01:06:40,442 --> 01:06:41,562 Fair enough, fair enough. 1252 01:06:41,802 --> 01:06:46,602 Anyways, we have come up to a little over the hour so we're going 1253 01:06:46,602 --> 01:06:48,802 to wrap up our conversation today. 1254 01:06:48,802 --> 01:06:50,482 Rob, this was fantastic. 1255 01:06:50,482 --> 01:06:53,802 Thanks so much for spending all the time looking into this and 1256 01:06:53,802 --> 01:06:55,572 for answering the questions for for Pedro. 1257 01:06:55,972 --> 01:06:59,852 I hope everybody found it useful because I think actually our 1258 01:06:59,852 --> 01:07:03,132 conversation today touches on on so many important points when 1259 01:07:03,132 --> 01:07:05,892 it comes to CTAs and trend following and investing in general 1260 01:07:05,892 --> 01:07:06,772 and all of that stuff. 1261 01:07:06,772 --> 01:07:09,892 So if you want to show your appreciation for Rob and all the 1262 01:07:09,892 --> 01:07:14,812 other co hosts, please go to your podcast platform of choice and 1263 01:07:14,812 --> 01:07:15,812 leave a rating and review. 1264 01:07:16,132 --> 01:07:20,132 It really does help and we really appreciate the support we 1265 01:07:20,772 --> 01:07:21,652 can get. 1266 01:07:22,932 --> 01:07:26,052 If you have a question for next week where I will have two guests, 1267 01:07:26,292 --> 01:07:29,872 Andre Beer will come back and he will be joined this time by Tom 1268 01:07:30,032 --> 01:07:31,392 Robel from SocGen. 1269 01:07:31,872 --> 01:07:36,432 So you can send your questions as usual infobtraders on plot.com 1270 01:07:36,432 --> 01:07:42,272 I'll do my best to get them in front of Andrew and Tom when we speak, 1271 01:07:43,232 --> 01:07:47,992 but for now for today from Rob and me, thanks ever so much for listening. 1272 01:07:47,992 --> 01:07:49,712 We look forward to being back with you next week. 1273 01:07:49,712 --> 01:07:53,792 And in the meantime, as usual, take care of yourself and take care 1274 01:07:53,792 --> 01:07:54,592 of each other. 1275 01:07:55,552 --> 01:07:58,592 Thanks for listening to the Systematic Investor podcast series. 1276 01:07:58,952 --> 01:08:02,032 If you enjoy this series, go on over to itunes and leave an honest 1277 01:08:02,032 --> 01:08:04,712 rating and review and be sure to listen to all the other episodes 1278 01:08:04,712 --> 01:08:06,392 from Top Traders Unplugged. 1279 01:08:06,632 --> 01:08:09,552 If you have questions about Systematic investing, send us an 1280 01:08:09,552 --> 01:08:14,272 email with the word question in the subject line to infooptoptradersunplugged.com 1281 01:08:14,272 --> 01:08:15,832 and we'll try to get it on the show. 1282 01:08:16,152 --> 01:08:18,872 And remember, all the discussion that we have about investment 1283 01:08:18,952 --> 01:08:22,592 performance is about the past and past performance does not guarantee 1284 01:08:22,592 --> 01:08:24,882 or even infer anything about future performance. 1285 01:08:25,202 --> 01:08:28,122 Also, understand that there's a significant risk of financial loss 1286 01:08:28,122 --> 01:08:31,442 with all investment strategies and you need to request and understand 1287 01:08:31,442 --> 01:08:34,642 the specific risks from the investment manager about their products 1288 01:08:34,722 --> 01:08:36,482 before you make investment decisions. 1289 01:08:36,562 --> 01:08:39,042 Thanks for spending some of your valuable time with us and we'll 1290 01:08:39,042 --> 01:08:42,002 see you on the next episode of the Systematic Investor.