Rob Carver on Trend Following, Skew, and the Total Portfolio Shift | Systematic Investor | Ep.375

Summary

• Trend Following: Robust discussion on edge vs. risk premium, accessibility of simple systems, and the long-run diversifier role of trend following.
• Managed Futures: CTAs framed as essential portfolio diversifiers, with 2024 dispersion driven by market universe and speed; equities challenging while precious metals, softs, and select currencies provided support.
• Managed Futures ETFs: Noted strong 2024 performance versus legacy funds, with lower cost and transparency; debate over marketing intensity and the primacy of exposure differences.
• Market Outlook: A neutral trend barometer (~48) and mixed November; precious metals and softs held up, while equities and short-term systems struggled amid higher volatility.
• Total Portfolio Approach: CalPERS’ move from SAA to TPA emphasizes portfolio-level contribution over asset-class silos; potentially constructive for CTAs to be judged on portfolio value-add.
• Risk & Execution: Preference for dynamic volatility targeting over static sizing due to superior Sharpe/KGR trade-offs; rebalancing guided by costs with buffering/smoothing to limit churn.
• AI & Sentiment: Nvidia (NVDA) earnings briefly tempered AI bubble fears, but the episode’s core focus remained on systematic strategies, diversification, and portfolio construction.

Transcript

Imagine spending an hour with the world's greatest traders. Imagine learning from their experiences, their successes, and their failures. Imagine no more. Welcome to Top Traders Unplugged, the place where you can learn from the best hedge fund managers in the world, so you can take your manager due diligence or investment career to the next level. Before we begin today's conversation, remember to keep two things in mind. All the discussion we will have about investment performance is about the past and past performance does not guarantee or even infer anything about future performance. Also understand that there's a significant risk of financial loss with all investment strategies and you need to request and understand the specific risks from the investment manager about their product before you make investment decisions. Here's your host, veteran hedge fund manager Neil's Krup Larson. Welcome or welcome back to this week's edition of the systematic investor series with Rob Carver and I Neils Castlassen where each week we take the pulse of the global market through the lens of a rules-based investor. Rob, it is great to be back with you this week. Hope you're doing well. How are things in the uh in the UK? Um it's cold and sunny. Um I think we're recording video for this episode. to people can probably see I'm wearing a lot of layers uh as my uh my garden office is not not particularly well insulated so I've got a heater running and I'm wearing lots of layers and hopefully I won't freeze to the death before the end of the podcast. >> Yeah, I'll make sure we'll keep you warm with some uh pointed topics for sure. >> If I get if I get nice and angry that will definitely keep me warm. >> Absolutely. And speaking of those, we've actually got quite a few questions this week. So that's going to keep you warm, I'm sure. and as well as a very very important topic that I personally think could be a gamecher not just for trend following and CTAs but generally speaking but we'll come to that uh a little bit later. Um of course before we get to any of that um I'm always curious what you've been thinking about the last few weeks since we since we spoke anything um that's stood out to you? Um, well, as always, I've I've been writing on my blog and people can can go and look there, but I'm not here to plug my blog. So, that's of course not. There are some very good articles on there, I have to say. Um, no, actually, the thing that that kind of came to my head this morning, I was trying to think of stuff to talk about was actually a discussion I had on um Twitter, now known as X, yesterday, and it was quite it's quite an interesting discussion. It comes up every now and then, and it's basically this discussion about whether trend following is an edge or a risk premium. >> Mhm. And of course to to you know I think the word edge is kind of one of these words that people use in different ways. Um so I think the most loose sense of it is basically if you can make money in the markets and you have an edge. I mean that you know my my personal belief is if anyone can do it then you haven't got an edge. um um if if you're making money out of something, but it's something that anyone can do. And of course, anyone can sort of follow a simple trend following system if they've got the right amount of capital, the right resources, um and and make money hopefully in the long run. Um then that money must be coming from what I would see as a risk premium. So basically, you're only making money because because other people are uncomfortable taking the kinds of risks that you're taking. So we can talk about psychological biases, but we can talk about um you know delayed reactions to information. Um those are the sort of main two kind of explanations for for why for why trend falling makes money. Um in fact there's a recent OddLotss podcast with Cliff Aseness where he he talked about that exact subject. So that that's worth looking up as well. So yeah, it was kind of interesting because because this debate comes along every now and then and it's amazing how heated people get about this stuff. So, so they're very insistent that, oh, no, I I I, you know, that there's definitely a tra an edge in using, you know, I don't know, a moving average crossover or something to to pick up trends. I'm like, well, how can it be an edge when literally anyone with a a spreadsheet can can do it or even just a calculator? Um, so that that was the the thing that got me heated up uh yesterday. Anyway, >> yeah, it's kind of interesting. I mean I I think sometimes we also try and over complicate stuff that we have to kind of be able to explain it in a very academic way. Um so um but yeah no I'll check it out. And by the way the the uh conversation with Cliff Fastness is on my uh it's queued up on my uh phone because I every time he speaks I I'm sure I learn something um and and have a laugh at the same time. So uh so he's one of my my favorites to listen to. So I'll definitely make sure I get that. a couple of things that stood out on my little um you know what's been on my radar. Don't know if you saw this um it kind of helps um make CTAs not look so bad this year and that is what happened uh at over at the the brilliant people at Renaissance Technologies uh in October. Uh I did notice that uh in an article that um a couple of their funds were behaving very uh differently. Uh the Renaissance institutional equity fund was down 14% in October um and down 8.3% so far this this year. And the diversified alpha strategy lost more than 15% last month according to the article putting it down more than 10 and a half% last month. Of course, what they're really known for is none of that. It's the medallion fund which they didn't quote the performance. I have a sneaky feeling it might have made money last month, but that's purely on my side. Um yeah, interesting. Um >> yeah, I mean there's always been a big disparity between um the Medallion Fund which has been closed forever, you know. Um only I think it it's only a very few small number of non-outsiders have money in it. Most of the money in there is is sort of um you know belongs to the employees or the the beneficiaries of the wills of the employees because of course Jim Simons is no longer with us. Um so so yeah there's often a big disparity between the returns of Medallion which are you know it's this highly secretive fund no one really knows how much money it makes either because it's not publicly you know there's no public information out there and uh the returns of the funds that you and I can buy the Renaissance sells um and I remember them them launching a CTA type strategy I think it was called the Renaissance institutional futures fund or something like that >> okay >> um I remember had a catchy acronym which that would have been riff of Um and that that did very badly um you know very much underperformed the the various other kind of benchmark CTA funds. So you know there's various explanations for this. Do they keep all the good stuff for themselves? Are they just good at one thing that has limited capacity and then the outside funds have to do other things? I don't know. But but yeah, I mean 14% in down in a month is isn't it's a big number, right? Um and yeah, >> I I think just to balance our conversation, um I I think to re I've seem to remember that actually these funds have probably done pretty well the last few years. So this is like the journalist picking up on a on an outlier number. Let's let's be fair. Um so um but I think you're right about the the the fact that they at some point also tried to get into our industry. Um that I haven't heard much about uh ever since. But, you know, we can all have a an out outlier month or an outlier year without a doubt. So, we'll we'll see where that come backs. The other thing that caught my attention, this is something I saw this morning. It was an article and I don't even know how I found it probably in one of those platforms that shows you all sorts of articles regarding or relating to to our industry. Um, it was an article in a on a web page called CT. Never heard about it. Uh the the um headline was uh volatility volatile commodity markets test quantitative trading funds performance claims and I oh that got my attention. Um one of the first sub headlines was something like performance falls short of marketing. Um they write about how um presentations to institutional investors um you know that highlights the outperformance during turbulent times like in 2022 is there uh and it positions um you know managed futures as an essential portfolio diversifier. Um, but it also talks about a Midwestern public pension fund being frustrated. It's uh quoting a female uh probably CIO being frustrated with CTA allocation and they obviously added it in 2023 after the exceptional year in 2022. So probably not the best timing and probably not enough time horizon um as well. Um but then it went on to something that kind of really caught my attention. Um, and I I I'm this kind of warms up to next week's conversation which will be be with Andrew because it also mentions the ETF uh offerings that of course have uh done very well this year. Um, and it talks about and I'm just quoting here a newer category of managed futures exchange traded funds has outperformed better uh has performed better sorry than many established ETA funds. these ETF packages, similar strategies in cheaper, more transparent structures accessible to retail investors and they have seen a lot of inflows blah blah blah. And then it goes on to say, "The irony is notable. Products marketed less aggressively and with few institutional trappings have delivered much better results than, you know, prestigious legacy funds, blah blah blah." And I'm just going to say I don't think it's a fair comment to say that the ETF space is not being aggressive in their marketing. After all, we see them a lot on all sorts of publications, including of course when Andrew is here. So, um, I thought that was kind of funny. Uh, if that's the impression they've left with this organization. >> Yeah. I mean, I'm I'm always getting content asking me to invest in CTA ETFs, but I suspect that's the various algorithms identifying me as somebody who' potentially be interested in that. So yeah, just to just to tie up on the the other point, um I believe it was the Renaissance International Equity Fund, >> which I think is a sort of equity market neutral type fund. Um probably not dissimilar from the sort of thing that the likes of AQR offer to be honest. Um and um it's worst year so far, I believe in the past in 2022, they were down 19%. Um so you know to be down what was it 7% for the year so far um is not like a you know it's not sort of an outlier um but their their performance since inception is only 3.7% so >> it's not really I think in fairness by the way again just to balance the conversation here I think I I read something in the in the art in the full article at least one of these funds actually had a mandate to be not very correlated with uh equities. So again, we can't translate just because so far this year the S&P 500 is is doing well that this fund has to do well. We we don't know. Anyways, just a little bit of good news before we go on and that is of course there is no AI bubble after all >> and that is that is purely based on the fact that yesterday the world can stop worrying about an AI bubble at least for now because AI chipm um lynchpin Nvidia dropped a stellar earnings report yesterday that has at least temporarily eased concerns that the economy is on the verge of collapsing like the New York Mets in September according to this article. I probably I got it from Bloomberg. Those uh those those nice words. Anyways, I have no idea um about AI or >> I have no idea about the new mess either. So, >> right. Yeah, exactly. But um >> I'm sure it' be fine. >> It the trend may continue a little bit longer. Anyway, speaking of trend, >> um my trend barometer finished at 48 last night. That's uh neutral, I would say. Um but it is a bit weaker than it was 10 days ago. And I think performance uh when we get to that in a second uh reflects that. Um so far in October I was kind of optimistic after last week I thought pretty good start after all. The first couple of days not great in November and then last week was actually decent. Um but this week not so much. It's been definitely a bit uh challenging a bit of headway for uh for for managers. At least that's what I uh see on my side. There are a couple of things that are standing out I guess so far this month. It's still, you know, okay to be uh long in in precious metals. It's okay in terms of the softs are doing okay. Uh maybe even some of the products, oil products. Um but equities is probably where the real challenge is. So depending on and this is where also, you know, funds like ETFs and replicators and all of that stuff, it's all about exposure this year. It's all about how much equity have you got or how much fixed income have you got and so on and so forth. So to say conclusively that one approach is better than the other I think is very dangerous this year because it's really depending on more I think the your market universe than anything else uh right now maybe speed of course uh short-term has been challenged for sure but market universe this year is really having a huge impact on uh on returns. Uh some of the smaller currencies New Zealand dollar Mexican peso still doing okay as far as I can tell. Um other than that and that's kind of been our experience. Um anything from you in terms of >> Yeah. >> Yeah. similar picture. Similar picture for me. So um I'm pretty much flat for November like I guess. Yeah. Like if most people had a few good days and given some of that back recently. >> Um for the year I'm up about sort of 5%. So it's okay. >> It's not too bad. So down down a couple of down about 3% from sort of mid-occtober when uh things were looking good. Um and I'm still in a sort of draw down of about just or just over 10% I'd say. >> Um in terms of risk kind of looking forward so uh interestingly my biggest short is in Bitcoin which is very satisfying because you know my thoughts on that subject but obviously that's the system hating it not me. Um I'm sort of net long equities although um for example I'm short DAX but I'm actually long the Italian equities so there's a fair bit of dispersion there. Um so but yeah like like I guess my my risk sizing is probably in line with your trend barometer. It's sort of definitely a bit below average. It's not the the sort of thought like we're going to get some fairly decent trends and performance is really picking up um kind of coming into uh sort of September October but yeah given things have got a little bit volatile since then. So >> you mentioned Bitcoin so I can't help asking you have you ever tried to model this um um cycle it has a h havinging cycle uh to see if there's any um truth to it. Well, as a quant as a quant I mean um I'm not sure off the top of my head how many h havingings there have been but I don't >> it's been like four or something. >> Okay. So that's not really enough data points for a statistically significant model is it? >> And you should know better than to even ask that question. Neil >> anything to do with crypto I have to kind of just poke a little bit. >> I know you're trying to make me hot. I know you're trying to get me. >> Exactly. I'm angry. Yeah. >> Good. All right. Good. All right. As of Tuesday, uh not so hot. I have to say the performance numbers uh down 62 basis points for the B top 50. down 60 basis points for the year. CTA uh Sockchen CT index down 69 basis points for the month, down 2.2% for the year. The trend index so gen is down 30 basis points, down 1.18% for the year. And the short-term traders index uh down 88 basis points, down 5.16% for the year. Yesterday probably a mixed day, some up, some down. Um so it could be a little bit worse uh as of today. Msei World also down 2.79%. so far this month and that is as of yesterday, Wednesday. Um, and it's up 16.86% for the year. The S&P uh US aggregate bond index down 10 basis points in November, up 6 and a half for the year. And the S&P 500 down 2.82% uh as of last night, up 14.2% so far this year. Now, we've got this wonderful topic that we're going to keep a little bit longer in terms of suspense, but we've got lots of great questions. So, we appreciate that. Um, thank you for sending them in. Now, I've not filtered them. I'm just going to take them as they came in in the order they came in and read them as best as I can. All right. First one is from George. My question is what does a solid reliable robustness testing framework look like? I'm struggling in the try different things uh out stage eg too robust every strategy fails even though they look good uh on unseen OOS data and if too loose we uh yield a fair few goodlooking strategies. Eg my current idea build strategy on three to five years of data depending on time frame and therefore risk of uh risk of regime change. Do a quick high precision test on the build data. Those that do significant worse bin them. Then a slippage test. Uh do they still make money if slippage is increased? I find this particular harsh assuming you get uh slipped five pips on every trade. uh Monte Carlo testing randomize trade order uh permutate strategy parameters not optimize but see if changing parameters by 10% up or down break the strategy I believe this helps to avoid curve fitting in the mining process then we test uh then we test on tick data uh by this point most strategies still in the running seem to survive this performance look good and ranking of the best strategies change slightly then the ultimate test is on unseen OS tick data one three one to three years depending on the data sample in the build stage. Then we have hopefully uh a good few strategies to incubate which hope um which hope leads to profitable live trading. Please can you guys comment discuss the process? I find my robustness test maybe too strict. Uh all my strategies are just beep. I'm just going to say that wordless beep. Um, but I always have in mind how do I know I filtered out the good strategies that I will that will last? My process is strict and yields a few excellent strategies. Thanks in advance. Okay, that was a very long thing. >> Just to be clear, beep beep is a bad word. Like he doesn't beep is a bad word. >> He doesn't think these are good strategies. So yeah. >> Um, there's actually a big piece of information missing from this, which is what your expected holding period is. Um, >> because that's going to make quite a big difference to whether this process makes sense. Um so I would say for example if if you were trading with very short time frames so intraday um then this probably makes this bit this the process the time frames and the things in the process kind of make sense um although I would say if you are doing that then you would should probably start with higher frequency data because he talks about build data and then tick data. Um, so you know, if you're trading kind of fast time frames, you really ought to be starting with tick data with higher with, you know, with with data that's much faster than your holding period is. Um, on the other hand, if your holding period is more analogous to mine, which is about say an average of a month, um, then you don't need to go near dictator. There's absolutely no need to go near it at all. um you can just do simple things to sort of check the you know how what the effect of your execution is on your um performance and if you're holding periods a month then even delaying your trades by a day which is what I test in my back tester shouldn't change your results that much and if it does you've got a problem somewhere >> um the the so again if you're but on the other hand if if you are holding for say a month um then I would say that 3 to five years probably for me is an insufficient amount of time um to to do sort of the calibration testing of strategies and I know he then has another one to three years without out of sample. Um but but so I I you know I think um this feels to me like you might not be getting sort of robust results because of this sort of brief period of time. Just to say in advance by the way overall this there's no kind of massive red flags in here that all the steps individually kind of make sense. just I think it's more whether together they make sense and whe and again whether the sort of time periods being used and the data being used matches the strategy that you're trading. Anyway, so putting that aside then um I think the this is interesting stuff actually because not the book I'm currently writing but the book I'm going to write after this one. I'm going to write a book on back testing because I think there's a there's a real um not partly partly a misunderstanding but partly awful. there's there's a sort of deep conflict between what the point is of a back test um and how you should use the results of it. Um because a back test fulfill two two functions. Um they fulfill a function of finding you a strategy that's going to work well in the future. Um and they also tell you how well you would have performed in the past. And ideally they would do both. Um, and this particular framework is not going to do that for you. It's not going to tell you how well you would have done in the past because you're basically selecting the strategies that that did the best with the data that you've got. And you've got this in sample and out of sample um process. Um but but that's you know only going sort of part of the way to to doing that because a re the best you know the real the real kind of gold standard way of doing this is you start with say a thousand strategies and you basically do a rolling process where you you subselect from those strategies depending on the performance you've only seen in the past and you'd be doing that over the 8year period and at the start of that 8year period you've got no information so you just include all the strategies you back test including ones that are beep as well as ones that are good. Um, and your performance initially would be, you know, pretty I mean, let's say on average half your strategy is half a good half a beep. Then your your performance to begin with would be flat because you'd have good performance and bad performance. They cancel each other out. Over time, you would hopefully discover what the good strategies are and you would include more of those in your portfolio blend. um and the and you'd end up with the set of strategies that you'd want to trade going forwards. Um and then if you look at that historic back test of performance, that's telling you how well you would actually have done had you had no future information at the start of your back test. Um and it's also going to give you at the end of the back test is going to give you the mix of strategies that you should trade going forwards. Um, but importantly, doing it this way means that it's much more likely that if you're doing your sort of strategy selection in a statistically robust manner, you're much less likely to do something like, and it sounds like this guy's doing it, to pick say 1% out of those thousand strategies to pick just 10 that do the very best because it's extremely unlikely unless you, as I said, you are trading at a very short time frame. It's extremely like unlikely that after three, four, five, six, seven, eight years of of time that enough time will have passed you to be able to say with statistical confidence that these 10 strategies are the best and you shouldn't trade anything else. Much more likely that you'd end up with um so you know, you're kind of fitting weights to a thousand different strategies. Maybe half of them you'll you'll you'll kind of have enough evidence to get rid of entirely, but most of the rest would still be in there. Yes, your 10 superstars would would have a higher weight than the others, but they by no means would they be filling the entire portfolio up. Um, and the the the issue here is um there's this thing called the multiple testing um problem. Um, and you know, essentially if you set a very even if you set a very high bar, let's say you say, well, I only want to see the top 1%. If you've got a thousand things, then just by luck, even if on average they're kind of flat performance, just by luck, 10 of them are going to be amazing. And then you pick those 10 out. And yes, you he's doing these other robustness tests that are good and will hopefully kind of weed out anything too crazy. But but you know, I still I still think that potentially the main issue with this um is that you're selecting a a very small subset of strategies based on insufficient data. That's kind of the big red flag for me with with this process as described. Um but you know thing doing things like Monte Carlo testing um doing things like holding out out a sample that's all good. Um um but you know um so the the one other thing to say is as well you know he tests stuff and then tests to see what the effect is of adding trading costs. Trading cost should be in there from the very start. There's there's no point even considering things that that you can't make money with before slippage. So, so you know that unless unless there's something about your back testing process that that makes it very um you know computationally expensive to include costs and that's why you're doing this. Um I I don't understand otherwise why you do that. So that that's kind of my general thoughts and as I said I could go on and you know as I said if I'm going to write a book about this which will probably be at least 300 pages then I could clearly go on for a long time to discuss this topic but but that's sort of my thinking about how you should back test and the purposes of back test and hopefully that gives you an idea of some of the changes you can make to this process to improve it. It's all well and good uh to follow a process and you know you end up deciding on something you feel comfortable with but um things change um data change how often would you go back and do this and would you simply change your say models parameters whatever um on an ongoing basis in the future based on future tests that you make? Well, basically in theory you run the process forward. So if your back testing process consists of so if you've got a monthly holding period then I think it's probably appropriate to do this reoptimization of weights to strategies about every year say okay >> there's no point in doing it more frequently than that >> right >> um and then and that will then result in your weight slowly evolving and depending on your time frame um you know you probably you will get to see a slow evolution some things will become better some things will become worse so that in theory should just roll that forward so at AHL We used to have an annual refit of our trading strategies where basically we do exactly that and we we'd include another year of data. >> But if you think about it, if you've got a long enough history, so um you know my data goes back to 1970. So that's now um 55 year 55 years of data in at the beginning of next year I'll have an extra year. I'm adding less than 2% to my total data set. Now you could argue that you should weight more recent data more more highly than that. Um but but even so, you know, if you've got long back tests and relatively slow trading, then actually your weights probably aren't going to change that much. And the truth is, for me personally, um I re I've refitted my my weights on average about every three years. >> Um so, you know, cuz every year seems like a lot of work. So, normally only if I make other changes to my system do I then go and also change the weights. Because, as I said, with such long holding periods, there's not really evidence to say you should do anything differently. No, I think that's a perfect uh follow-up uh answer. Thanks for that. All right, we move on to a question from Frederick. He writes, "Hello, thanks for the show. I've learned a lot from it. My question for Rob is if he has any thoughts or statistics around intraday versus end of day trading for a medium to long-term trend system, would the faster entries and tighter risk control actually lead to better results?" Uh I'm a bit confused by this question to be honest because the the two sentences to me don't seem related to each other. So um so let's say you're trading medium to long-term trends. Okay. So you can trade those that system in different ways. So one way you could trade it is basically um so a lot of people back test their their stuff on end of day data which is completely fine if you're trading slowly enough as we discussed the first question. So, one thing you can do is is basically um then when you're trading actually try and trade as close to the end of day as possible. So, you're sort of matching your back test. Um now, um that that might not be ideal. Um and it's funny actually because I've done consulting and a lot of people have got this this sort of fixation with oh I must trade at the end of the day because that's my back test data from >> and I said to them well just just try this experiment which I already mentioned just delay your fills by a day in your back test. Does it make any difference? Does it not make any difference? Okay. Well, then you're fine. Or you you know, as long as you do something like what I do, which is essentially the pre, you know, take the previous day's closing prices and then at some point during the following day, do your trading, you'll know you're going to achieve performance somewhere between if you'd managed to do it exactly at the close of the previous day, which you can't, and you know, and the the following day is closed, which is what I back test, you're going to be somewhere between those two. And if those two things are close enough then obviously on a day-to-day basis you might see some differences if prices is a bit crazy but on average you'll be fine. >> Um so I would suggest and the other advantage of that of course is if the if the close is not liquid or is crazy or there's some weird auction process or whatever um you can you can focus your trading at other parts of the day that are more liquid. And of course if you're trading institutional size money the last thing you want to do is do one massive trade every day. you want to spread your trades throughout the day and try and reduce your market impact as much as possible. Um, so so if you're trading medium to long-term trends where basically you could you want to trade roughly once a day, then spreading that out is is no is no bad thing. The other thing you can do if you have a system like mine which essentially um dynamically looks at its positions sort of optimal versus what it actually has, I could actually run that thing throughout the day. >> Um and that would not increase the total amount of trading I did, >> but it would again result in a sort of smoothing process and it would allow me to react faster to very large intraday price movements, which may or may not be a good thing because obviously if prices bounce back then that's that's actually going to lose you money. Um so that that's kind of my answer to what I is quite for me quite a confusing question to be honest. Uh and I don't really understand what faster entries and tighter risk control actually mean unless he's talking about trading much more quickly. So he's actually talking about sort of intraday trend trading. >> Um so um you know in which case you know you've got two big I've got two big issues with people who trade trends intraday. Um one is obviously costs. costs are going to be higher >> um and because you're doing more trades, but also because um it's harder to to do your trades with sort of passive limit orders because you want to catch a trend that's going to only going to last a couple of hours. You don't want to sort of wait there for it to be passively executed. You need to jump in straight away. Um and the other issue is if you actually look at the um efficacy of trend systems, they they tend to start to decline for holding periods shorter than about a month or a couple of weeks depending on the asset class. And by the time you get down to to intraday, um it's less obvious that there is there are strong trend patterns there. And actually, if you get down to sort of sub one hour, you tend to start to see mean reversion, which obviously is the opposite of trend. So, I'd be very careful about looking at uh intraday and trading. Um faster entries and tighter risk control, whatever that means, um aren't going to be a lot of help uh when you're paying those those big trading costs. >> I'm thinking that, you know, if you're newer to the space and you get maybe information from, you know, online platforms, people promoting stuff, etc., etc. I can I can understand why my people might think that actually um oh trading faster as long as it's trend it works. And I even had a conversation on email uh from a nice guy in the US uh regarding um you know trading trend using trend following just on a single asset class. they were they're doing it on on equities and I'm thinking okay I'm it's great if it works for them but it's just not my experience that actually trend following is designed to work either on fast speeds or on a single uh um asset class. It's really the diversification and and the longer holding period and all of that stuff that makes it a little bit difficult, makes it a little bit uncomfortable that makes it work. So, so but I I do take that people might have a different impression coming to the space um and and maybe getting a lot of the information initially from um from um say more commercial outlets, let's call it that. Um anyways, we appreciate the question, Frederick. All right, question from Abram. I'm starting to read Mr. for Robert Carver's book on systematic trading and found his trading journal in elite trading spanning from February of 2015 until now. After reading the first pages, I skipped to the most recent pages and saw Mr. Carver inviting readers to ask questions in your upcoming interview. Can you please help to ask how Mr. Carver will start his automated systematic trading differently given over 10 years of experience. What kind of advice can Mr. Carver give to a new systematic trader just starting out in this area? Thank you and best regards from Abram. >> Uh I mean it's quite depressing that that you can only manage 10 pages of it's not clear whether it's my my trading journal or my book only 10 pages before he just grew so bored and just skipped. >> I think he did the right thing. He went straight to the source and asked his questions. I think he did fine. >> Yeah. Well, if if this is a ploy to try and avoid reading the rest of the the book or the or the trading journal, a tough luck, you know, you're still going to have to do that. So, that's my first piece of advice. Read read till the end. It's all good stuff. Um although to be fair, the the trading journal is many thousands of posts long now. So, you know, it'll take a while. The book's only a few hundred pages though. Um yeah, it's an interesting question actually because you you you know what's changed over the last 10 years. It's actually uh where are we? It's about 11 and a half years since I started um running my own automated training platform and it's nearly 20 years since I began working in the in the CTA industry. So you know you for quite a few things probably have changed in that time period. Um but I'm actually struggling. I mean Neils you you've been around even longer than I have. Um, although people might might not guess it because you still have a reasonable amount of hair, so you probably look about 10 years younger than me. Um, I mean, I don't know what you think, but I'm actually struggling to think how my advice would have changed from 10 or even 20 years ago. Um, I think I think I'd probably say the same things. I mean, if if I was to go book and revisit that that book I wrote, yeah, over 10 years ago now, um, I'm not sure I very much in it would I would change. I mean, I probably put some stuff in for crypto just for that specific asset class that's kind of come out of nowhere. Uh, and my my new book that I'm writing now has has got some stuff on in crypto on crypto just for that reason. But apart from from that, um, you know, I don't I'm not really feeling that there's been, you know, it's it's a things things I think the same principles apply. I mean, things we talked about about back testing, I think those are principles that have have applied and it's not like they're sort of new scientific discoveries. Um the advice I normally give to people is things like you know keep things simple. Um as we've discussed you know think about trading costs um think about leverage optimal leverage don't take too much risk because you know the two things that that the three things that kill systematic traders are spending too much on costs you know taking too much leverage um and overfitting their trading systems. Um, and those are the same three things I was warning about 10, 12 years ago, and those are the same things that I think were were um a problem 20 years ago. Um, so yeah, I'm going to hand this one back to you, Neil. What do you even more experienced than me? What do you think? >> So, so first of all, I would say um and I think this was your first book, if I'm not mistaken, um, systematic trading strategies. Am I correct in saying that? I think that was actually the first time uh you were on the podcast was we've reviewed that book um which is a great book and everybody should really read it. >> Um and it is timeless and I I tend to agree with you that >> not a lot have changed where I think things may have changed is that I don't necessarily feel that the way we try to identify entries would have changed. Um, of course, we still adhere to the various golden rules of trend following that have lasted for decades. I do think maybe the industry has evolved when it comes to how we manage risk. It's difficult for me to say specifically how each firm would have involved. But I if I look at, you know, what we do at Don, I would say we do things differently. We do it in a more smarter way. um in a sense um I think also we are more open to differentiating speeds a bit and I don't mean just deploying short-term uh strategies that's not what I mean but you can design your systems to react differently from entries to exits I think that that is something we do and I think that has helped us um and as I said overall risk management I feel has evolved as well Maybe people uh like you wrote another blog post which we're not going to talk about today but that Katie plucked for you last week was about uh predicting volatility. Of course volatility may maybe we are using volatility differently today than we did 20 years ago. Um and and maybe we can use it better now in the way we manage the risk. So I I generally agree with you that it's not massive changes but I do think that may you know we have evolved. I mean there's a reason why we have these um research teams uh who are very good at what they do. Um it doesn't mean that they can overcome bad market environments and we've had some of that in the last 20 years for sure and that's why people sometimes write that oh this strategy doesn't work as well anymore. I'm not so sure that's true. At the end of the day if you're starting out in this journey I think it's really about committing to that journey committing to these rules because it won't be an easy journey. um at all for for you Abraham. Um as we discuss every single week, uh I think that commitment and really wanting to do this and feeling completely comfortable with a a strategy that um is unpredictable in some ways, uh yet it's very predictable in the long run. It's unpredictable from a day-to-day basis. I think that's probably some of the things I would work on. All right, let's move on to Carlo's question. Carlo writes, "I would like to submit a question for Robert Carver. How should a trend following investor think about position sizing when volatility is time varying?" Okay, so now we getting into some of the stuff. In particular, I would be very interested in Robert's view uh on the following points. Static versus dynamic exposure. Some trend followers argue that maximize the economic benefits of outlier hunting. Um I think he's been listening to Rich too much actually. It is preferable to keep the numbers of contract unchanged throughout the trade so that only the volatility at entry matters. Others preferred a continuous volatility targeting framework based on Robert's back testing experience. Is there empirical evidence that dynamic volatility targeting produces better results than a static exposure? Or is the choice mainly driven by investors objective function? For example, prioritizing sharp ratio optimization versus maximizing long-term KGA as if that is something you can guarantee just by using static position sizing. I just want to caveat that. Rebalancing frequency. How often does it make sense to adjust exposure as as volatility changes? Nature of volatility. Should a trend trader treat volatility differently depending on whether it supports the trend positive volatility or threatens it negative volatility? Thank you so much from Carlo. >> This is a great a great question. I could almost have written it myself. Um so actually on the first question about whether the choice of dynamic or static volatility position sizing um actually let let me very quickly because I'm I'm sure 90% of the people listening to this know exactly what we're talking about but there's probably maybe 10% who have just just started listening maybe Abraham's just started listening. >> Y >> so um static exposure is where essentially um you you look you you start you put a trade on and you size your position according to the volatility that's current at the time. Um and then you you basically maintain that position to at the same level. Okay. So if the position if the market gets riskier, you know, your position stays the same size. If the market gets less risky, your position stays the same size. You may adjust your position for other reasons like if you're using some kind of pyramiding where you add as things um become have stronger trends or whatever. Um now the alternative is dynamic position sizing and that's where um you essentially measure the volatility throughout the life of the trades. If the market suddenly gets riskier, you reduce your position. If the market gets safer, you increase your position. Um and this is sort of relates to Oh, >> can I add one thing? >> You can >> uh the dynamic position sizing may not only be influenced by volatility. >> I was about to say that, Neil. I was about to say that. Will you not butt in, please? I'm I'm on a I'm on a I'm on a roll here. Come on. Give me some space. Um, yeah. So, um, the Yeah. So, what I was going to say was the the continuous what I call continuous trading is where basically where you you constantly evaluate your positions according to things like volatility, strength of your trend, but also things like potentially fact it could be even things like um the size of your account as well. So, as you lose money, you'll be taking money off the table uh and so on and so forth. Um now I've test and >> correlations Rob uh yeah potentially depending on how your your risk risk system works. So in in my particular system correlations wouldn't affect things to a first order but but if if things get very correlated then there is a sort of exogenous risk factor that kicks in. Um so and you know but if if yes if you're running your system at a fixed volatility target we always target the same V every day then one thing that would affect things every day is correlations as well. So absolutely. Okay. Now I have I have tested the these two things a few times on my blog and and one one sort of important question is yes. What what what do you care about as an investor? Um do you care about maximizing your sharp ratio? Do you care about maximizing your geometric returns? Um and the reason why that's important is um that things that have positive skew um may not have as good a sharp ratio but may produce you a better geometric return. things that have negative skew produce the opposite effect. Now, if you're using if you're not using volatility targeting, then your skew profile is much more likely to be positive skew. And the reason for that is quite straightforward. If you're holding a a position and the the thing suddenly leaps in price and let's think about Coco a couple years ago because that that's the example that I think still sticks in people's heads. Something leaps in price. um if you're not changing your position according to volatility, that leap in price, which of course also increases volatility, you keep your you keep your position the same size and that means you get this massive outsized positive outlier return. Um and then that that obviously will help improve your geometric return even though from a sharp ratio perspective it may actually look worse because your your risk is going to be much more variable and and sharp ratio calculations don't like risk that moves around a lot. If you've got a nice steady consistent set of returns, you're much more likely to have a higher SH ratio. Um, now it's basically possible um to look at the tradeoff between um I won't go into too much detail because you will have to go and look on my blog to find find the article, but especially you can look at the trade-off between um uh skew and sharp ratio for maximizing your kgar. So essentially there should be a certain amount of of of positive skew that you're willing to give up in return for higher sharp to get the same kar. So there sort of a almost like a riskreward trade-off except that it's skew versus sharp ratio trade-off. Um and what I found is that that basically and this is not an empirical result by the way. This this is actually theory. So you you know you can't argue with it essentially. It's not just a quirk of my data. um the the amount of um sharp ratio you should be prepared to give up for the sort of improvements in skew that you see from um these um you know using fixed vault scaling and not not change your position size. Um it to some to kind of the bottom line is it's not worth doing fixed V position sizing because the the additional skew you get does not pay you for the loss in sharp which is substantial. So you end up with a lower KR basically. Um and and um that you know so some just to clarify. >> Yeah, please go. >> Yeah. When you say fixed V, what you just mean is you keep the the position >> static position static position sizing. Yeah. Exactly. Yeah. >> Yeah. Um so actually now if we think about funds that have the you know some CTAs that have extremely good um returns very good and very good KARS and positive SKs and have a sort of outlier behavior um and and you know their volatility is often very big. Um you may they sort of will tend to outperform a you know some a system that's more like mine um which which has a higher sharp ratio, less positive skew, less outliers because it's dynamic position sizing. But but but what would what basically if I apply just a little bit of extra leverage to my system, I could very easily beat their geometric returns um because because I have the because my superior sharp ratio um you know more than pays for the the loss in positive skew that I have. So that kind of answers that question. It's quite a quite a long-winded answer but but it is quite a complex subject but yeah go on my blog and there's an article about it. Um then there's a question about rebalancing frequency. Well, essentially this is the standard question about rebalancing is when you're doing rebalancing, you're trying to trade off two things. The benefits you'd get from having your position close to what your optimal position should be and the cost of trading to get there. Uh, and the worst thing in the world is is if you rebalance the wrong way. So, let's suppose volatility is sort of jiggling around a bit but basically falling. So all the things being equal, I would increase my position size. If I was to to rebalance, say every day without fail, then some days I'd be buying, some days I'd be selling. I'm assuming everything else is fixed, of course. Um so I've been incurring a lot of extra trading costs. Um so um you know, that would imply that my rebalancing frequency should be a bit slower because those extra trading costs are going to more than kill um you know, the extra benefit I might be getting. Um but there is an a solution to this conundrum and that's to use um either buffering or smoothing. So with buffering essentially you you only trade if your positions further away from the optimal position and with smoothing you you basically take something like a moving average of what your pos want your position to be. And a lot of a lot of trading rules that we use obviously incorporate moving averages in them already. Um so with volatility you know if you were using something like an exp weight weighted moving average of volatility estimates that would be relatively smooth and you'd do fewer trades. So the answer to the the question about rebalancing frequencies to be honest I can't tell you because it's going to depend on costs um but as I said if you use these other strategies then you can do more rebalancing more frequently without without actually incurring any trading costs you're going to get get the win there. Uh now the last question actually is the most intriguing because I don't actually have an answer for it and that's about using um an asmmetric volatility measure. So if we think about Coco as an example um then that was positive volatility right the price went up we were long hooray um and you know maybe in those circumstances we shouldn't reduce our position um now I my concern with this is that um you know returns are a coin flip uh pretty much um so you know if you're a really good trader you might make money for 51% of days and lose money 49% of days. Um, so if you've got a massive risk on because the thing's just gone up in a straight line and that's wonderful. Um, and you just hold your position the same size, you're taking this huge amount of upside of risk. Your upside risk and your downside risk are the same. Okay. Um, you're just a coin flip away from those massive profits becoming massive losses. So overall, I'd say no. I'd say you should treat both kinds of volatility the same. That's not to say though that assets do have different behavior on the upside and the downside and there are interesting things we can do to model that. But but this specific question I I suggest that that you just treat risk as a symmetrical. >> Well, that's definitely an option for you to include that in one of your next books whether there is a difference in in that subject. Anyways, let's move on because we've got still got a couple of questions before we get to the big secret uh topic of today. Um question from Dario Mr. Hi Mr. Carver. I hope you're well. Thank you so much for your contributions to the investment community. I'm curious about sentiment indicators used in low-frequency algorithms. I wonder if you have ever tried trading off sentiment indicators. Are your skew signals a proxy for sentiment? Thank you from Dario. >> I must say I'm I'm I'm I'm a slightly uncomfortable with all these people calling me Mr. Carver. Um it's it's uh it's a bit formal for me. But anyway, um, okay. So, sentiment is interesting because so I don't actually use uh any any sort of um sentiment indicators myself. Um, so my skew signal, which he asked specifically about, that's that's that's something that's looking at historic levels of skew. Now, you could argue that is a sentiment indicator. You could you could you could argue that because because if an asset has recently seen very violent falls in price that suggests that most people had would have a negative sentiment towards that asset. >> And in the case of of my trading system I would actually buy that asset because because something that's got very heavy negative skew would tend to be underpriced because people don't like it. So actually I'm a buyer, you know, in that specific example I'm a buyer of negative sentiment. Um but but generally speaking um you know partly because of resources but I tend to stay away from let's say weirder data um things like sentiment scores um you know also some of them tend to be um effectively in sample fitted um so this you know the when I used to work for um AHL we had a classic thing when where sellside traders would come in and say look we've got this amazing model that predicts this do you want to buy it or trade it you or trade it through us and give us a commission or whatever. And you'd say, "Well, how this is, you know, how did you construct it?" And after a lot of plotting and proddding and pulling and asking you questions and digging into details, you find out essentially it's massively in sample fitted and therefore of course it looks very good. Um, and um, you know, we we generally speaking would prefer to do that job ourselves, you know, um, and do it properly. Um so so I'd be wary of kind of buying any or sort of random third party sentiment indicators but there are there are things that kind of indicate the sentiment of the market. So skew is one. Um there's there's sort of various options ratios as well you can use. You could just look at the level of the VIX and say well that's sort of a a proxy for how scared people are. That's that's a measure of sentiment. So all of these are good things that are potentially useful for for predicting predicting prices. I just tend not to use them uh myself, but that doesn't mean that they're not a valid research area. >> Sure. Fine. Final question today is from Andreas. Actually, someone I know who Andreas is. So, um he has two questions for you, Rob, and he does call you Rob, by the way. In Rob's research, the relationship between ATR and standard deviation, he concludes with an empirical and theoretical solution. Question A. Is there a mathematical logical relationship between daily and weekly ATR? And B, if a model is trading once a week only, is it advisable to use daily data or weekly data with respective to ATR? And then there's a follow-up question, but let's do one at a time. >> Yeah. So, uh, ATR for those who aren't familiar is average true range. So, the true range essentially is is the the the difference in the highest and low lowest prices you see in a time period like a day. So um the this is you know it's it's sort of a bit like standard deviation in the sense it's designed to measure how much market move it sense it looks at the rather than just looking at single price per time interval what you know technical analysis call a bar um it looks at the the size of the bar itself um as well um now essentially the the it's not possible to um to work out um so you could look at the empirical relations in daily and weekly ATRs, but it depends on essentially on auto correlation depends on how the price return in in in one period influences a return in the following period and this actually affects standard deviation estimates as well. So um for example um if if prices tend returns tend to cluster so good returns tend to be followed by other good returns um then then generally speaking uh if you compare say your daily standard deviation and your your annual standard deviation you'll find that there isn't the kind of theoretical square root of time relationship between the two. You'll find that the the the an you know that um so let's take an extreme example. Suppose prices go up by 1% and down by 1% every day of the year. If it's a leap year, there's an even number of days in the year. So the return for the year will be zero. So if you if that happened forever, then the standard deviation of annual returns would be zero. But the standard deviation of daily returns is 1%. So in that that simple example where you've got very strong negative auto correlation, the daily standard deviation is much higher than the annual standard deviation. If you had positive order correlation, then it would the other way around basically. Um and this is this quite interesting for people who do things like look at you so there's a paper by Andrew Low where he looks at hedge fun performance and sharp ratios and you know concludes that you really want to try and get sort of monthly weekly or daily data rather than just looking at annual because annual returns hide a lot of a lot of fun autocorrelation properties but that's another story. Um so yeah it's not really so I think off the top of my head the weekly to daily ATR relationship should behave the same the same as standard deviation. If you assume zero autocorrelation and probably some other distributional assumptions then it should follow the square root of time rule and that means that if if markets are trading um on weekdays only then there should be a sort of multiple of about square of five between those two values in theory. Um but yeah, it depends on the autocorrelation. Um >> and would you use daily data? Sorry, weekly data if you only trade once a week. >> I mean, yeah, there's no reason not to like what what you it was like the earlier discussion when we were saying, well, why use tick data if you're trading daily? Like what what's the extra what's it what extra information are you going to get? >> Now, having said that, sorry, the question was about estimating the ATR. I would say yes, you get you always get a more accurate estimation of volatility if you use um data that's more frequent definitely. So yeah, I would use daily daily price changes to calculate my ATRs. But in terms of actual general back testing of my system then yeah sure weekly for weekly daily to daily there's no reason to do it faster. >> Yeah. No, that that's important. All right. Final um second part of a question. Has Rob done any research on using intraday stops versus using market orders at the next day open if the hard stop has been reached? How much additional room eg expressed in ATRs is advisable to add to an intraday stop? >> I mean I don't use stop losses when I'm trading and I don't use ATRs either. So I'm not sure how how qualified I I am to answer this question. Um the one thing I will say is that um there is a a nice paper by I can't remember who it's by. It's inevitably going to be someone who works at AQR. I'll say that much. Um I don't think it's auntie. Um and I don't think it's Lars. I think it might be Frank. Anyway, um but the paper basically discusses how like a very neat way of combining slow momentum with um shortterm mean reversion is to do something really simple which is when you come to execute your trades the the day after the previous day's close as we've been discussing only only execute those that have moved in your favor. So if you're buying only execute if the price has dropped between the last closing price and vice versa. Um, so that's a very a very quite a cheap way of of getting potentially better execution costs by essentially combining your slow system with a fast fast system because normally the problem as we've discussed with faster systems they cost too much in trading costs. This way you you know you're going to do that trade anyway. It's just whether you do it at which level you do it at. >> So the key issue there is whether that that delay of one two maybe more days. We know that a delay of one day is not going to affect us too much. We discussed that already. But whether delaying by um sort of potentially more than one day because you're waiting for the price to reach a nice level whether that's that's an an issue or not. And that's something I want I need to test because that's that's in my backbook of things I need to research and implement. Definitely. I've not answered the question at all because I feel completely unqualified to. It's just that was a random a random idea that that came out in my head that I think is sort of related to what he's talking about. Let's move on to the um to the very important topic I think because yesterday I think it was there was a vote in California not about election districts or anything like that. It was from at the board of administration of US pension fund Kalpus and they voted in favor of adopting the total portfolio approach TPA which is going to replace their existing strategic asset allocation the SAA model um next year July 1st 2026. Um and um they also said that they are going to change their model reference portfolio um to 75% equities, 25% bonds. Um and they also said, and this is a quote because you and I were not entirely sure whether this was true or not, but it is a quote. They say they are the first pension fund in the United States to adopt TPA. Um and uh David Miller, the the uh uh investment committee chairman said um this will give Kalpa staff the edge they need to make sound investment decisions. Now in the article that I found which is not the FT article that you send Rob, it also goes on to say under the TPA the focus will be on which investments can best contribute to the performance of the entire Kalpus portfolio as opposed to achieving individual asset class targets that were period periodically reviewed. Now, I'm going to shut up now pretty much because I want you to take us through what you found to be the interesting part. But as I said last week when I spoke with Katie, because of that sentence that each component will be judged on what they add in terms of value to the portfolio. Personally, I think that makes it very very interesting from a trend following perspective. >> I think it does. Yeah. I mean, although cynically I wonder whether this is just going to be used an excuse to jam more private equity, [Laughter] >> you know, or private private debt or other private private anything private. Yeah. Because people people love the private stuff at the moment, don't they? >> But but no, let's be upbeat and positive and assume it's going to be a good thing for for trend following. Now, I have to say I I was very intrigued by this this subject um for a number of reasons. One is that um portfolio optimization is is sort of my hobby. I suppose it's my one of the big research area coming keep coming back to and and um one I've done a lot of spending time and thinking about. So you know think the sort of basic maths of the the Maravitz um sort of so-called modern portfolio theory modern even though it was invented in the 1950s um is is you know under has underlied pretty much you know everything ever since um and um is is what the the kind of asset allocation industry use as well. Um, so I've never worked as an asset allocator, but obviously I've had to deal with them as a as a potential supplier of of alpha >> as you one might put it. Um, and also um because Mang Group had a a number of um multi-manager um strands within it. I've also worked with the quants working those multi-manage organizations helping them with things like analyzing returns. So I'm kind of I have some understanding of of the way that asset allocators think. Um but bas but let's sort of briefly discuss this. So so SAA or strategic asset allocation is kind of the way that people have done asset allocation for almost I would say my entire lifetime pretty much with some some tweaks >> and it does grow out of the mark of its model and the basic idea is is this and it's simplest to think about it in a equity bond setup. So you think about the the cover, you know, your expectations for forward returns, risk and correlations between two asset classes that you're considering. Um, and then you you you set some target return. Um, and that's for example, if you're a pension fund, um, you know, that's going to be based on something like the average age of your members and, you know, how much money you need to sort of pay them out. And there's these, you know, a recent innovation has been these things called target date funds where you dynamically change your allocation as as people age effectively. And you know, simply in simple terms, that will put more into bonds and less into equities, right? Which kind of makes sense. Um, and then and and then you build this thing called the efficient frontier, which is the sort of portfolio of of, you know, the best portfolio, these two assets that has the, you know, you pick a point on that that has the highest shar ratio. Um, and then if you can use leverage, you basically construct a tangent. If you can't use leverage, you pick some point on the efficient frontier where you get essentially the lowest risk that still meets your hurdle return. So, what that means in practice is if you're um if you you're a retirement fund, but your members are quite young, you can you want to achieve a certain amount, you can you can sort of have a higher return to try and meet your obligations in the future and and that will mean a higher mix to equities. And obviously if you've got a lot of old retirees and this is particularly true of the UK. So in the UK we've got a lot of um so-called defined benefit pension funds where the investment risk is all on the manager and not the person the retiree. >> Um and that means that they have a lot of money in bonds. Um and that's been particularly good for longdated UK bonds actually. Um that does weird things to the yield curve. That's another story. Anyway, so the that's your sort of starting point and then you use then got your SAA your strategic asset allocation. Um and then of course you can add to that other things other sort of they use this expression sleeves which I've never really understood but you know you can have sleeves for for hedge funds you know equity market neutral hedge funds specifically perhaps you can have them for private credit and again for for private equity and of course you could have a sleeve for CTAs. Um and then the idea then is did that you then sort of go down to the next level and you say right we need we need to have exposure to this asset class we you know and then we did maybe make some determination about how much is passive how much is acquid active maybe we have some country waitings and then and then so on and so forth so the key the key points here are that it's a very kind of structured top- down process basically I like to think of it as a series of boxes so you start off by saying, "Well, I've got this big box and in it I need to put asset classes in different sizes." And then I open up those boxes and I put other stuff inside those those boxes and so on and so forth. And importantly, no one takes into view the fact that if for example um let's say um I happen to put into one of my boxes something that's highly correlated something in another box that does not enter into the decision process because no no do we're just we're just putting stuff in boxes now. We're not we're not considering the total portfolio the holistic portfolio. Um, and that might mean, for example, that uh, well, let let's take a what I think is a pretty madeup example, but let's just think of it anyway. Let's suppose you have a CTA that has amazing performance, but for some reason has a very high correlation with the Danish equity market. This is I'm not this completely made up example, right? Um, and let's also suppose that the that separately you're the guy that manages your equity portfolio really likes Danish equities. um well then you've got a problem because you have a big exposure to Danish equities in two parts of your portfolio. Um and maybe somewhere there's a risk management team that will point this out to you. But in terms of the portfolio construction process, this this sort of weird correlation hasn't been taken into account. So that my understanding a total portfolio um approach is is that rather than doing this kind of top down bits and pieces blah blah blah that we look at the thing as a whole and then we we we'd potentially make a decision along lines just saying well well we've got this amazing CTO manager but he's got this massive exposure to Danish equities and we've already got a big exposure to Danish equities or does it make sense to add them to the portfolio? Does it make sense to dial down the long only Danish equities and then add these guys in or does it make sense to take out the Danish equities long only and just put the CTA in? So you you make those decisions on a more holistic basis. Now I have to say that to an extent this whole can we call it an industry yet? I don't know >> which part. >> Well the TPA is TPA an industry yet you know. >> Well there are like I mean >> there are people there are people going to do this for a job right going forward. They quote like five or six sovereign wealth sovereign wealth funds or big pension funds worldwide. >> Yeah. >> It doesn't I'm not so sure that's a qualifies it for just yet. >> Let's call it a cult then. Uh instead >> because it's like a probably a few hundred people doing this globally. That's about the size of a cult. >> Or a group. We could just call them a group. >> A group. I don't know. Cult cult maybe is a bit has has connotations you think. Yeah. >> Yeah. Um anyway, one of the problems I found with this this group is that it does seem to be something that's been invented by some management consultants and it's very vague and there's a lot of very nice charts and pictures and expressions like I'm going to read this out. Um uh TPA shifts the focus from rigid associate allocations to unifying strategy where decisions are guided by the fund's overarching goals drawing on ideas from entire investment portfolio induces the flexibility to adapt to emerging priorities such as sustainability inter generational equity evolving economic conditions. The approach enhances governance fosters collaboration unlocks new opportunities and supports multi-dimensional 3D investing. Now that sentence was definitely written by a management consultant and not by a quant because a quant would say something like maximize utility function where you input the correlation the standard deviation and the expected vector of mean returns plus your constraints and that I've just described essentially the mark of its optimization. So, it's a bit hard for me to kind of make judgments on this generally speaking because there's no real precise mathematical definition of what it is. And I I you know, I really feel like I need I need that to hang on to. But as a general rule, clearly I think it makes a lot more sense to think about your portfolio in a holistic sense rather than just putting things in little boxes because I think another point I would make is that one of the issues the CTA industry has is which box do we go in? M >> are we in the hedge fund box? >> Because but but we're we're quite different from most hedge funds. We got quite a different correlation and we you know are we in in some kind of tail nebulous tail protection box? Well, that's tricky because unlike proper tail protection funds, we can't guarantee that we'll make you money in in a in a down market. We hope we will. Um but but there's no guarantee of it. You know, on the other hand, we've got a positive sharp ratio which those guys don't have. So, you know, we don't fit in that box either. So maybe, just maybe, this this move away from a reductionist approach of putting everything into little boxes will also be good because people will be able to say, well, I've got this my portfolio here, and if I add this this this weird thing here, it does improve the overall setup as long as there's not only too many Danish equities in there, of course. And that's that's great. So that's kind of where I am. I'm hopeful. Um I'm a little bit skeptical because it does sound a little bit management consultancy. Um I'm also as a quant struggling with the fact there's not really any formal definition of of how you do this. Uh and there does seem to be a lot of handwaving going on. But that's that's my take on it. But I think this you're right this could this could be a gamecher. I mean if Kalpas are doing it you know they're big players. Um and um you know if for example someone like Adia starts looking at it um you know the quant team at Adia are sort of off I mean they I think they employ something like 20% of all the quants of the world now they're off you know there's some very famous people there. Um so if they start doing it then then I think that really will be a gamecher and maybe they'll actually write some papers and explain to us what it is because I'm still struggling slightly. >> Yeah. No I think you make a lot of great points. Um, and um, I was I was thinking if it wasn't a management consultant that wrote that text you read, it could have been chat GBT. And the reason I say that is you also sent me an article and I have no idea what you were thinking. You sent me an article in French expecting me to read about the total portfolio approach in French. And to be very open about it, I'm not very good at French anyways. So I had to resort to chat GPT just to make some sense of it and what it did I just asked for a summary and when it got to the TPA it wrote the following and that's why I'm thinking could be JBT one portfolio one goal maximize total return within a defined risk budget asset classes don't matter only contribution to total outcome does CIOled dynamic collaboration performance equals fund objective, not asset class alpha. This is portfolio construction without walls. And then I asked, okay, so why does this work? And it says it empowers teams to act on the best opportunities, not just stay in their lane. Freeze capital for macro, trend, and uncorrelated strategies. Puts risk at the center where it belongs. favors adaptability over prediction. >> Wow, >> this is fantastic. I love chatbt >> that. Yeah, I I Yeah, I don't know. It's becoming increasingly difficult to tell the difference between AI and non AI related content, especially if it comes from a management consultant. I have to say >> it even went on to say, here's the TTU, the top traders on block take. Okay, it says SAA is a map. TPA is GPS and in this environment you need to navigate not just rebalance. For trend followers, TPA is the best chance we've ever had to be judged on what matters. Our ability to deliver value at the portfolio level. This is our moment if we don't if if we are ready to own it. I love Chat TV. It's so it's fantastic. I'm I'm I'm I'm sure in the not too distant future there's there's there's going to be uh just just a blank screen next to you as you press play on the Rob Carver >> Mr. Rob Mr. >> Mr. Mr. Rob Carver chat GPT trained uh LLM uh with with complete with voice uh and uh yeah off off it will go. There'll be no need for me me to keep keep signing in. >> The good news is that is not going to happen just now. you will be back soon. But >> yes, Christmas episode >> and and actually if there there is a Christmas episode coming up which will be fantastic with all the co-hosts. Um I think maybe one won't be there. But anyways, um so but before we get to even any of that, um I think that hopefully I'm hoping uh people will go to their terminals, find their favorite podcast platform, and say a big thank you to you by leaving a rating and review for this episode. Um because you put a lot of hard work in answering all those questions. So, we really appreciate that as well as reviewing uh the uh the uh articles and information about the TPA. So, that's uh great. Um and um if you have any questions, comments, otherwise, as usual, you can send them to uh info@toptradersunplot.com. And especially if you have something for Andrew Beer u because he's coming up next week. This will be fun. This will be interesting. It always is with Andrew. Um, I think we may have a little bit of back and forth on some of the new stuff that he's going to be talking about some some news that just came out actually in the last uh day or two. So, um, join us again next week uh, when Andrew is here and send us some questions like you did for Rob this week. Anyways, from Rob and me, thanks ever so much for listening. We look forward to being back with you next week. And in the meantime, as always, take care of yourself and take care of each other. Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today's episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you'll be sure to get all the new episodes as they're released. We have some amazing guests lined up for you. And to ensure our show continues to grow, please leave us an honest rating and review in iTunes. It only takes a minute and it's the best way to show us you love the podcast. We'll see you next time on Top Traders Unplugged. [Music]