Simplicity in Systematic Trading | Robert Carver

So there are actually a very very small number of circumstances at which I believe it's its right to meddle with the trading system is where override it the danger is if you're ever in a situation where you are spending too much time looking at what your system is doing and and following the financial news and all this kind of stuff all these things feed into a an environment in which you're you're more likely to Try and second-guess the system and override it that's why I I don't sit at my computer all day watching a train

I spend a lot of time setting it up so that it's fully automated and just reports to me when think things look like they might be going wrong in every case the accuracy of experts was matched or exceeded by a simple algorithm why are experts inferior to algorithms one reason is that experts try to be clever think outside the box and consider Complex combinations of features and making the predictions complexity may work in the odd case but more often than not it reduces validity this is a quote from Daniel Kahneman's book think fast and slow

and it's very relevant to the conversation you're about to listen to on today's show 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 trainers 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 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 products before you make investment decisions here's your host veteran hedge fund manager niels castro Larson [Music] welcome to top traders unplugged where my goal is to give you the clarity Confidence and courage you need to invest like or invest with one of the top traders in the world it is the stories that you never get to hear set

out as the most honest and transparent account that I can make of what goes on inside the minds of some of the best investors in the world today you're listening to episode 89 if this is the first episode you've heard you might want to go back and listen to all the earlier conversations before we find out Who's on today's show I want to mention that today's podcast is brought to you by Eurex and in today's conversation I actually learn about a brand new contract from my guest that Urich's have just launched which my guest finds

to be very useful for many investors and which you can find out much more about by visiting the Eurex website today I'm talking to Robert kava who's both an author and a trader and who spent a good chunk of his career at one of the Biggest European systemic trading firms namely AHL before leaving in 2013 to write his newly released book and by the way if you want to read the full transcript of today's episode just visit the top traders on plot common website and sign up to receive access to all of them now let's

get started with part 1 of my conversation I hope you will enjoy it [Music] Rob thank you so much for being with us Today I really appreciate your time thank you for 90 minutes my pleasure now Rob you are a little bit different from my usual guest which have been hedge fund managers or CTAs running their own firms and strategies today you're an author and you trade your own money having left the bossing world of the City of London behind but in your new book systemic trading a unique new method of designing trading and investment

systems I think you bring Such an important perspective on the systemic trading world and you break down the components in such an easy-to-read way that I wanted to make an exception to my usual lineup and bring you on for an in-depth conversation on a number of topics relating to this and for me it's a little bit funny because when I look back at all the podcasts interviews that I've done so far it's precisely one year ago I made the same exception for Katy Kaminski who had also just published a great book on the world a

trend following so for full disclosure I have tried to read all of the 300 pages but I may have skipped a few in order to make today's deadline but from what I have read I really recommend anyone interested in the subject to grab a copy of it or perhaps to since christmas is coming up and for those who think I'm getting paid to say this you're wrong this is purely a recommendation based on My assessment of the quality of the book now many people in the alternative investment industry are very familiar with some of your

previous employers in particular AHL which has been an institution in the CTA space for many decades but Before we jump into your background I just have a simple question that I try to ask all of my guests in order to appreciate the different answers that there is to this question it basically goes something like when You meet people that don't know you and they ask what you do how do you respond how do you explain what you do that's a very difficult question to answer I know you say that either that I'm a writer or

that I'm an independent trader depending on who I'm talking to and which answer I think will get the best response but I basically do both of those things I mean I the last thing about trading systematically is if done properly once you've actually designed Your system it takes it very little time sure and that leaves you more time for for thinking and writing which is what I actually enjoy doing more so that's how I would describe myself as fantastic now at this stage I would normally move on to my usual questions but I want to

do something a little bit different today since you have just published your book and I can't think of a better way to kick things off then by setting the scene and you reading a little bit of You know the book in order for people to get a sense of your writing style which I very much enjoyed so I I would kindly ask you maybe to read a little bit of the preface and then a little bit of the introduction to the book if you don't mind sure okay I am very bad at making financial decisions

like most people I find it difficult to manage my investments without becoming emotional and behaving irrationally this is deeply irritating as I consider Myself to be very knowledgeable about finance either a sh ously read the academic literature done my own detailed research spent 20 years investing my own money and nearly a decade managing funds for large institutions so in theory I know what I'm doing in practice when faced with a decision to buy or sell a stock things go wrong there in greed washed through my mind clouding my judgment even if I spent weeks researching

a company it's still hard to Click the trade button on my boat brokers website I have to stop myself buying or selling on a whim based on nothing more than a random newspaper articles or an anonymous bloggers opinion but then like you I'm only human fortunately there is a solution the answer is to fully or partly systematize your financial dis making creating a trading system removes the emotion and makes it easier to commit to a consistent strategy I spent Many years managing a large portfolio of trading strategies for a systematic hedge fund unfortunately I didn't

have the opportunity to develop and trade systems to look after my personal portfolio but after leaving the industry I've been able to make my own trading process entirely systematic resulting in significantly better performance an hour read for an introduction sure it was the 23rd of January 2009 and I was in my London office although I had A desk overlooking the Thames I was usually too busy to appreciate that new my day job was managing a portfolio of systemic trading strategies for a large hedge fund but right now I was focusing on my own bank balance

data was about to be released indicating how the UK economy had performed in the last three months of 2008 it would be bad news the official confirmation that we were in recession but nobody knew her how bad this didn't mean extra work for me However since a bank of computers would adjust our clients portfolios automatically when the news arrived so I decided to devote some rare free trying to trade my own money with a stressful full-time job I was not a particularly active trader but very occasionally an opportunity came up that was too good to

miss this was one of them in my research I found that historically when people's fears were confirmed by terrible economic numbers was often the best time To buy and this was potentially the worst news I've seen in my lifetime careful analysis showed that the bank's hardest hit by the financial crisis should rebound the most if things improved I was particularly attracted to Barclays I traded for their investment bank appears before and their balance sheet was in relatively good condition but they also looked at investing in the other major UK banks in all I was prepared

to risk 10% of my Portfolio on for banking stops then her figures came out they were worse than expected with GDP falling by 1.5 percent Barclays dropped 15 percent almost immediately taking in to the lowest level I have ever seen I waited for the market to stabilize and prepared to trade then I hesitated everything had happened is expected I should go ahead and buy but what if this was wrong what if the financial industry really was imploding as everyone else seemed to Think panicking I quickly changed my orders knocking a zero off each one so

that only 1% of my portfolio was at risk he was one of the biggest mistakes my investing career thanks very much Rob it's funny when you read that you you you it really does take you back to that time of 2009 which was a scary time to be in in the investment world now this was of course just a little bit of a background and I Wanted to ask you perhaps to take us a little bit further back in your own career to sort of appreciate and understand how you came about ending up where you

are today so if you don't mind you know take us take us take us back even to you know when you're a kid growing up what were your interests were you know where you're always curious about things give us a little bit of color if you don't mind sure and I think it's fair to say that And from quite a young age I was fascinated with computers and I'm 41 years old so I've got a computer when I was 7 at home so I was probably one of the first generation to have access to a

personal computer at from quite a young age sure and this interest have developed to the point where I went to university to study computer science which in my knee that naivety I'd assume would mean that I would be taught how to program computer games which at the time I think was the career that all nerdy boys of my life my friends wanted to do sure unfortunately that the course was very dry and scientific in had too much theoretical mathematics so I didn't complete that course I spent a few years working in a completely unrelated industry

and then quite by accident came across a copy of the Michael Lewis's famous book lies at oka yes sure in the bookshop and I read the book and the Interesting thing is that Michael Lewis wrote this book to to sort of scare people off from working in finance but I think it had the opposite effect yeah and I know so many people who said yeah I read lies poker and it sounded amazing and you know I want to where I want to get into that world and do that yeah it was a great book so

then I became interested in economics and ended up studying economics since my university and going back and this time I managed To complete my course so obviously I'd found the right the right thing I then went to work for Barclays Capital and investment bank and I spent a couple of years of them trading exotic interest rates options which was an experience but not one I would want to repeat okay I think it's fair to say that I'm not really cut out for that the current trust of a of a bank trading floor what time period

are we talk about knowing this was in the early tooth okay Okay after the dot-com boom okay what things got crazy again I think also from reading that section of my book you'll appreciate that actually I'm not emotionally cut out for making investment decisions throughout and I actually believe that very few people are yeah which is why I think systemic trading is the way to go and should will discuss that social sure anyway after that rather traumatic experience I spent a couple of years Doing my master's degree part-time but then whilst also working for a

an economic think tank which was a more cerebral activity and a bit of a rest from the frantic trading no world feel and then again quite by chance see to get ahead in life you have to have a lot of luck very lucky quite by chance I saw an advert in the Financial Times um for AHL okay and I'm you have a child because I'd actually work for them as a summer intern a few years beforehand and They were looking for someone who had both financial markets background but also an economics background and also an

interest in computer programming am i certain they're thinking this time vote has been written for me exactly I don't do many of those people are L exactly so I applied for the job and got through the rather tough interview process and I got the job for testing so you joined a chill when was that before that I did yeah I joined in 2006 okay so when I've Got a very different world from the one we live in now yeah and I also feel not sure if Luck's the right word but I I think it's been

an excellent education having spent the the period from 2006 2013 when I left the HL yeah probably you know the most exciting period to work in the financial markets obviously the excitement was both good and bad but the learning was good the learning was excellent yeah yeah no absolutely now of course you're been very busy writing a Book and and and that you know I can only imagine how much effort and time that takes but when you are not trading when you're not writing what what you like to do nowadays that sort of takes you

your mind off those two things I enjoy cycling mm-hmm so and one of the reasons why I wanted to stop working you know full-time capacity with the the two-hour daily commute that they evolved was also to spend more time with my young family So those are my other interests and in the summer I also enjoy doing a bit of sailing except so we're I full disclosure I was actually a world champion in one particular sailing class when I was younger so Wow Wow there we are that you are my first world champion on the show

again in full disclosure on I'm nowhere near as good as I used to be insane fantastic excellent Rob now you've written a book about systematic trading Which normally involves a lot of math and equations in the real world but you've managed to write a book with very little meth being used so why was this important to you to do it that way the credit for this really has to go to my publisher okay when I I sent the first draft of my first chapter to my publisher and he came back saying this is great but

by the third sentence you lost me but if you want to write a book that only you know perhaps a few Thousand people in the world can understand then that's fine you know we're prepared to do that obviously you know the Condors is a kind of an equation that publishers use to do with size of audience cost of book length of book publication costs and that all these things kind of get factors in them together so at one extreme you'd have a you know a highly specialized option pricing book that might sell you know you

know in a good day months other Thousand copies are its whole lifetime yeah it'll cost you you know three figures to buy and at the other end you might have investing for dummies which is probably going to be ten dollars and could could well sell hundreds of thousands or even millions of copies truth so he said you have to think about whereabouts and spectrum you want to pitch your book and I very much felt that I wanted to speak to as large an audience as possible but so then but Then the publisher can came back

again and said you know you you're going to struggle I think to to bring your ideas down to the level where anyone on the street could understand them so let's try and picture somewhere in the middle and and that's where I tried to do it I'm not sure how successful I've been because there's a couple of reviews on amazon.com one says this is the first book you should buy when thinking about systematic trading which another one Says this you should probably read at least 10 other books before opening this one opinion differs as to whether

you know it's it's it's a straightforward as you say but I so need to try and make it as accessible as I could and incorporate maths only when I thought it was absolutely necessary I think you did a great job and by the way you know having also put myself into these public light with this podcast you know comments and opinions and reviews say they are what They are and completely uncontrollable and and whatever people whatever have people had in their mind when they wrote and you know that's how it comes out so I I

would focus on the on the good ones even though you can have a hundred good ones and one bad one and it's the one bad one you focus on isn't it it's that's true but in fairness both of those reviews were good reviews sure and it's just that they disagreed about how I'm gonna take that as a kind of Sampling error issue and assume that on average I'm somewhere in the middle which is something I'm comfortable with ya know it's a great book absolutely now I want to start sort of digging into the book a

little bit and and in the beginning you you start out by defining three types of investors namely the acid allocating investor the semi-automatic trade and the stall systems trader I want you to tell me a little bit about each of them and why it's important to Define them but I just want to be completely you know open here and say I actually I'm not English by backgrounds I didn't know what the word stones meant and so I I looked it up and I mean there may be other people on the show today you know listening

and who have no idea and so it comes out as saying very loyal and committed in attitude you know and it has all these things so so now at least I I learned something new very early on in the book which is stones Means that you are very committed to you to your strategy so anyway that was a digression here let's go back and and you explain a little bit about the three types of investors as you see them okay so again in the interest of making the book more accessible I didn't want the book

to be read by people who just purely wanted to to do what would normally thought of a systematic trading in other words meaning you have some rules that look at Normally technical price patterns and then in a completely quantitative way you know decide what positions you should have as a result of looking at those prices and then make the trades often automatically so that that's what a staunch systems trader is is someone who that has an end-to-end process with you know trading rules at the beginning that are completely systemic and then a position management framework

that translates those into positions and Trades and does all the risk management which again is completely systemic and then the trading at the end it can be automated but the point is that you know you follow those trades religiously and you never deviate from them so it's a system that that could be completely automated now that's quite a narrow set of people and not everyone is in a setup where or kind of placed themselves where they're comfortable with that and it depends on whether you're comfortable With the idea that's a relatively simple hopefully set of

trading rules can can actually predict what will happen to prices and financial markets now not everyone signs up to that and I kind of identified two other groups of people who are out there the first of people who think that the humans are better than computers at predicting price movements and to be more specific they think that they personally better at predicting price movements than the Simple rules are and this is what I described as the the semi-automatic trader okay so the idea behind the semi-automatic trader is someone who still wants the freedom to say

I think that that Apple is a good buy but I should be sure Google but who then wants to take that and put it into a systematic positional management framework that will then decide how big that position should be you know when what size they're stopped Should be when they should open positions when they should close positions and positions they should have open how do you know how to manage holistically the risk of that because I I believe that the if you have a good position management framework in place then how good your your your

ability to forecast or how good your systemic trading rules are if you're the first kind of investor is much less important right to the extent that you can Actually run simulations with completely random entries so you you basically simulate the trader who is no better than flipping a coin and if you then feed that into a position management risk framework and set up correctly that guy will still make money and that's partly because in the past we've seen trends and markets and if you're setting up a system where you've got big stop losses then they

will naturally tend to pick up on on trends but it's it's still An interesting finding and if you then add in trading rules that do a good job of predicting where the markets going to go that does a performance to your system but not as much as you might expect sure so that's the second kind of person now the third kind of person is it's a very miserable cynical person who thinks that no one can predict what will happen to so priced in financial markets and I call them the asset allocating investor because a kind

of long term buy And hold mentality is we often say well you you can't pick stocks you know you the best portfolio is to to buyer so a selection of ETFs they give you exposure to different asset classes and then you you basically say you know what I've got no idea what these things are gonna do I'm just gonna buy all of them so that that's a perfectly valid point of view and I run part of my own portfolio on that basis but then again I still think there's value in using a position Management framework

to say well that's fine but you know how should you account for the different risks of the different assets you're buying how should you account for the correlations you know how should you trade that portfolio how should you rebalance that portfolio given you've got a set of costs so the challenge for me was to create what I call the framework which is that sort of thing that in the middle between either a trading rule or a Qualitative opinion or a stubborn you know buy and hold mentality and then that takes all of those opinions and

kind of processes in the same way and produces set of positions and that's what I've tried to do sure no absolutely oh thanks for doing that now Rob many people in sort of the money management business are very focused on explaining how they do what they do in order to convince investors to let them manage part of their money but in my opinion a More important question is why so let me ask you why you should start a systematic trading strategy today because most people are not as good at trading as they think they are

that's pretty much it I mean I talk a lot in my book about what behavioral finance people and psychologists call cognitive biases so essentially you know our brains are kind of wired in a way that made sense when we were no wandering across the plains of Africa a few Hundred thousand years ago and trying to hunt whatever we were hunting though when you're trying to make decisions based on complex information and those emotional biases that come through that mean you end up often doing the wrong thing now I know with myself that this is true

and with the the little Barclays anecdote that I read out to you that could read out many many more of those bad decisions that I've made you to emotional problems and I believe these Affect nearly all people I think the you know and the biggest bias of all is overconfidence people think that they are much better than they they are and people think they are much better than average than they are and only someone who thought that they were better whatever that means I'm not trading than don't anyone else would actually actively trade the financial

markets and you know all the People who are actively trading the markets can't all be right there must be some of them whoever though averaging true I believe that if you use a system with relatively simple rules you can actually have overcome these biases and and then even exploit the biases that other people have now a slightly question about sort of because I mean we always have to sort of look look at ourselves and and in what we do and and early on in your book you you you you Explain why people should be skeptical

about trading systems that you can buy off the shelf or books that you could buy or blog post that you can you can read and and we all know that there's a lot of that out there but can you maybe explain why you think your book is different probably because I spend most of my book explaining how bad I am trading and how you should be very you know I for example I say a lot you know you you really shouldn't expect a Sharpe Ratio of more than well the absolute maximum you should expect is

one right and and actually for other people you should expect a lot less and actually you you know you should you should do your kind of position sizing as if your Sharpe ratio was half who you expected and in contrast you know you if you're trying to sell a trading system the natural human instinct is probably to do some kind of back test and fit this thing until you have a Sharpe ratio that Looks attractive and they you know is unlikely to be realizable in real life because of course you've ever fitted to get it

yeah but it's very hard and I have sympathy for people who are in this marketplace true I'm not saying they're bad people they're evil people who are out to scam everybody but but it's like being in a market where you're selling cars to people and the people who are buying your cars have no way of knowing how how good the Car is there's no way of knowing verifying for example what the top speed of the car is sure and you have to say oh my car can go 250 miles an hour because another guy saying

or my car can do 200 miles an hour and when you know people don't buy the car and unlucky they drive it and it's a trading system so of course the returns you get are random to an extent and the car then in only just fifty miles an hour all crashes and and they can't Really complain because of course you know that that's the way these things work so I think any anything you're reading you have to ask yourself why is the author presenting me this information and why what's their motivation for doing it I

mean my motivation for writing my book was obviously to sell it sell books sure of course your economic perspective writing a book and selling it within getting a Percentage royalty you know in terms of hourly wage I have to sell quite a lot of copies just to get to the point where I'd be getting the same as I went working in McDonald's for writing so I'm I am genuinely interested in in educating people and trying trying to explain to them that they need to be more realistic and that's a completely different market place from where

you were trying to compete with people who are trying to make the most outlandish Claims to stand out from the the pack of people making similar outlandish claims no I agree with that completely and I think it is evident from from reading your book that that is the the fundamental motivation now you you you touch already upon the point about sort of the the flawed human brain in in your book and you end up talking a little bit about sort of the the temptation of taking profits early and letting our losses run that's really how

We as human beings are wired tell me a little bit more about that and and and what it really means when it comes to trading if you get this balance the the wrong way around so to speak and what you found when you when you test these kind of human behavior so to if I could call it that so the the natural human instinct in when you see the position rise a little in price then your long is to say is to want to take a profit and this comes Down to essentially that that kind

of strong feeling lee wants to kind of lock that in and the reason you want to lock running is you want to prove that you're right it's the overwhelming human emotion to to prove that you're doing the right thing it's called confirmation bias in the literature now when when the stops falling if you sell at a loss then you're going to be proving that you're wrong I nobody wants to do that true so What you actually then want to do is hang on to that position hope it goes up in value and as it keeps

falling of course you have the same conversation with yourself and so you're forced to sell perhaps because you know you've ran out of money sure so it's really about the way the human brain is treating a lot realize unrealized losses and unrealized losses differently if you're thinking about them differently even though they're Exactly the same thing and you know there's this kind of mindset that it's it's not a profit until you you've sold it and it's not an it's not a loss until you've you know taken a lot of it is completely wrong it's very

easy to think about a sort of pattern a price where it would actually make sense to buy on a small profit and that would be if the market was trading in a small range now the problem is that most of the time markets don't do that they trend now This isn't the time for there on a theological argument about whether trend-following is a good thing or a bad thing that certainly in the past and people like you know Wynton have done tests over hundreds of years of data where it's available markets having the past exhibited

a behavior where they're trending mm-hm so yeah if markets are gonna trend in this the behavior we're just gonna sell in a small profit and you know cut only When you've got a huge loss is exactly the wrong thing to do you should do exactly the opposite of that sure which is what a trend following system will do so this is a really good example of where there's a human bias in our brains creates exactly the wrong kind of behavior and you can then write a really simple rule that not only corrects for that bias

but will actually exploit it and if other people are doing are doing this and you know this trend following System will effectively be taking money money off them sure social now of course it it kind of goes into the debate has also been about you know you know different kind of strategies you know convergent strategies versus divergent strategies we know trend-following is a divergence strategy then you have a lot of relative value strategies on the convergent side and and I guess I mean I guess in in you know part of your conclusion is of Course

that's you should have a little bit of everything and that's that's probably true but when you did your test I mean from from memory you did a test with two these two different rules on thirty one futures contracts what what did you find in in in that and if you don't remember I I have to fight I have to finding in front of me okay well I did I did find that I think he was 26 out of 31 markets or 27 sure probably seven out of thirty all markets A very simple rule which took

losses early and let profits run yeah it wasn't actually a kind of classical trend falling well it was something much simpler than that and did better in in 27 out of 31 markets you know that's not a huge surprise because you know firms that have been trained for have been profitable for many decades so it's not it's not a big surprise but as you say I'm not saying that trend Falling is the only way to trade exact just for this is a really nice example of where a human bias produces a behavior in the market

which can be exploited by a simple trading room sure entamoeba there are others on a develop you know on the convergent side as well yeah yeah no no true and actually you use the phrase simple trading rules that there are two things you kind of highlight also early on in the book and that's you know the importance of having Simple trading rules but also the importance of sticking to a plan tell me why this is crucial in your opinion well if you're not sticking to a plan then you aren't really trading systematically the the whole

point of having a system whether it be you know the full-on store what I call the storage system trading where you running with systematic trading rules and the position management framework or the more qualitative semi-automatic where you're Making your own forecasts but then putting them kind of binding yourself into this systematic framework to actually trade and manage those positions but the whole point of that is that you you gain the benefits that you can get from doing that which you know you're going to lose if you if you start meddling with your system and making

changes and you know this is something that everyone does from the the cummock that the guy is a retail trader and Who's using an off-the-shelf charting package and is looking at the signals that are coming off it and saying well I don't really like that signal I'll ignore that one I'll do this one I won't do that one but even in large institutions like AHL and I'm not criticizing them specifically because I I know happens in in all institutions the trade systematically you still have debates about whether you should override the system cut the system's

Risk because of something that's going on in the world now the key point of course is if you've got a purely systematic training system that you tested in the backtest all kinds of stuff happened no one was was there in the backtest to to override it the system just runnin and did what it did and assuming you accountable with a backtest and comfortable with its behavior and will do what training it was doing in the backtest Yeah then you should really be comfortable with letting that thing run now without interfering with it because that's exactly

what happened in the back test so there are actually a very very small number of circumstances in which I believe it's its right to to meddle with the training system is where override it and unfortunately this is something that comes about with with longing experience and I think the danger is if you're ever in a situation where you were spending Too much time looking at what your system is doing and following the financial news and all this kind of stuff all these things feed into a an environment in which you're you're more likely to try

and second-guess the system and override it that's why I I don't sit at my computer all day watching a train I spend a lot of time setting it up so that it's fully automated and just reports to me when think things look like they might be Going wrong sure but in an institutional setting you know you it's much harder when you're you're trading with other people's money because you have this fiduciary duty to look after their money and as you know yourself and if something happens and you don't override the system then there's always that

question of whether that was the right thing to do so it's it's a culture in which is much harder to to sticky exactly to what the system's doing so I Mean it's interesting I mean you and I are talking you know in in in November of 2015 and actually I would say the things you just touched upon is very real right now in the debate of of investors which obviously I spend a lot of time talking to because people are worried that the coming changes in interest rate environment meaning we've gone from a from a

bear market in bonds to a you know a sorry from a bull market in bonds to to a bear market at some Point when the interest rate cycle turn which of course we know the US central bank has the duty to now a few times this year already and at some point it probably will come and there is definitely fear out there that all these track records that we have been able to produce and can document and show they're not going to be worth a lot when when interest rates suddenly start going up because not

many CTS F in this case CGS have traded through a a rising Interest rate environment and so there is this fear that oh it's gonna stop working and then it's not so and of course there is a point to it because most testing will have been done on data for the last three years thirty years three decades not a lot of people and a lot of not a lot of data is available going further back I wanna I wanna I wanna to talk about something related to this in a second but I just want to

hear your initial reaction to to this kind of Concern that the investors clearly have so my last job was managing the fixed income portfolio okay so yeah I was thinking about exactly this problem okay and I came to a number of conclusions the first conclusion is people often forget that what makes we're not actually we say oh we're looking at prices and we assume that if er a bond price falls that we're going to lose money but actually what we're exposed to is the total return Mm-hm so if you're if you're owning bonds then your

total return is going to come actually this applies to all futures but is going to come from both the the movement in the spot price and also any any carry or all down that you're getting and the carrier rule down essentially is sort of telling you what the market expects what happens to spot price over the period whatever period it is so what that means in practice is if You're an environment where interest rates are very low we're expecting to go up then the interest rate curve the yield curve will be quite steeply upward sloping

and that means that the caring you'd get on owning bonds further out in the maturity space will be relatively I so in a nutshell if the mark if the interest rate moves in the way that the the forward prices expect it will move you won't actually make all those any money there's only that the the rates Change unexpectedly so if they if they rise too early or too fast that you lose money so there's that's the first thing to say is that people kind of I think a lot of people miss that we did a

lot of kind of simulations and tests looking at different and you know interest rate environments and we came to the conclusion that you know that there wasn't as much for problem as you might think sure the second thing is is diversify Mmm diversification if you're running a CTA and forty percent of your assets are in US bond the futures then you're some kind of crazy guy right and this this is true we know regardless of what you think Janet Yellen is going to do hmm you know you should have a diversified portfolio so you know

I probably I look at my own portfolio perhaps twenty to twenty five percent is and is in bondage true and if it was more than thirty Percent I'd be sort of thinking well that seems a bit high regardless of what I think's gonna happen to interest rates it just seems quite high and given all the all the asset classes that are out there in the CTA space yeah and also you know why have you got so much money just just in one country of course all one prices will reacts what happens in the US but

the what happens in the US will be the most significant Thing so you know if you've got a reasonably diversified portfolio then you exposure to anything unexpected happening the US should be relatively small the third thing to say is you know we I we were having this debate for the best part of three years before I left AHL and it's now 2015 so this is a debate we've been having for five years yeah and if you've done if you've done any kind of meddling in that period like reducing your exposure to fixed income You would

have been hurt because being long fixed income and also trend-following fixed income as being one of the the the greatest trades over the last five years so for example last year and was an excellent year for CTAs and most of them made most of their money in bonds actually mostly in European bonds but you know so if you kind of cut your exposure in bonds too much to you know to say just 10% you'd have seriously Missed a lot about the term yeah so my message really is don't panic yeah no exactly yeah yeah I

mean if you you you can go back to probably the last time we had an interest rate rise that kind of panicked the markets in the similar way was 1994 when resource was when you know for example Orange County happened a lot of people got really got caught shop so you know you can kind of go back and look at CTAs for trading back then in' or look at back tests and simulations You know in as much as they can be trusted and look at what happened then so it's not like there isn't any data

at all but and you'll see losses of course but they shouldn't be too large as long as you you know having exposure self too much to one asset class in one country sure and the other thing I would I would add I think all your points are very good and of course one can add one more and that is of course that systemic traders today can be you know short the Pawns as easy as they can as long so so there's no there's no but there should be no bias there but you know what what

is interesting to me is that there are a few maybe a handful of these managers who were around in the last interest rate sort of hiking cycles so from 1977 to 1981 for example interest rates went up dramatically and I can see on the firm I work for which happened to be around back then you know it was a very profitable period for for this kind of Trading so to me at least it looks to me that that when the bigger interest rate cycle is up then I think there should be good opportunities but when

people refer to the period as you mention 1994 which was a difficult year for for the CTA space well actually what it was he was a correction in interest rates going up in a much bigger down move so to me it was more of a counter trend situation than it was the fact that the interest rate cycle had turned Hasn't really turned it was just correcting against Topeka to it anyway let that be for for a minute I'm gonna I'm gonna give you a pause to drink a little bit of tea because I want you

to I want to point out something that related to what you mentioned earlier which was the the importance of sticking to a plan and I happened just to be sent the other day a link to an article where a queue was which for those who don't know is one of the very big firms in in Our business the founder of a QR Cliff Asness was recently interviewed by Bloomberg and this is what he said about you know investment you know success and and and and and also in relation to to to warren buffett he basically

said that genius is still good but more and more i think about doing something reasonable that makes sense and then sticking to it with incredibly fortitude through the tough times maybe i metal that up a little bit in my reading of it but Basically what he's saying is sticking to a plan is possibly more important than just being a genius now the other thing he goes on to say when talking about sort of the quote unquote greatest investor in in the world warren buffett and he says that and this is about a study that was

done about him he said of course they found he was fantastic but not quite as fantastic his track record was phenomenal but human phenomenal what was beyond human was him sticking to Sticking with it for 35 years and rarely if ever rarely retreating from it so it goes very much to the point you made before now of course at this stage i have to do a little bit of selfish promotion and that is to to put things into perspective a little bit and that is with my own that the founder of the firm I work

for Bill Dunn who has essentially been running his investment strategy for 41 years with an annual return of more than 15% which put him of Course along with with our you you ownership of Marty Bergen right up will with will Warren Buffett yet you will never hear these kind of rosy descriptions in in the media of how they describe Buffett's achievements but anyway that's a little bit of a of of being sidetracked but I just think the the sticking to a plan and being exceptionally disciplined over the long run I think it's such an important

point for people to realize absolutely and Understand I mean I think I completely agree with Cliff to the extent that I don't think I am a genius and therefore sticking to a simple plan is absolutely the right thing for me to do hmm now let's talk a little bit about the difference between a subjective and an objective system and why objective systems in in some ways are better can you talk a little bit about that because you mention it in in in in your book as well Sure I'll probably start by telling you that often

when I bump into people who you know that trading I'll say oh you you do technical analysis and write your fundamentals trader and I say well mainly I'm doing technical analysis because I'm using a system that just uses price data and then they they kind of either get very excited and start talking to me about double bottoms and of course his heads and golden crosses well they're kind of back away in and Make the sign of the Cross to ward off the evil through hanging on which camp they're in so you know is this kind

of this assumption that the technical analysis is this obscure on this voodoo like a Russian where you you know you kind of stare at lines on the on a screen and and then make a judgment as to what's going to happen now all these kind of classic technical analysis methods which I I would describe as pattern matching okay because generally Speaking there's a there's a pan that you have in your head and you're looking to see that pan and then you make a decision on that to me they're all extremely subjective you know generally speaking

you know you you can't write an algorithm that will identify those patterns and if you try and do so you usually won't end up with a cyst that is profitable in a backtest so they're subjective they're subjective systems there's no rules you could write Down there's no computer program you can write to identify them they're purely in the eye of the beholder now there may well be there probably are actually people who who can do this who can look at charts and and and see these patterns and make decisions and make money but you know

that that's not the world of systematic trading to be trading systematically you need a purely objective way of identifying given some data you know what your position should Be and that means you can't use you know all these weird and wonderful patterns but you can use things like moving average crossovers you can use things like you know break outs out of a range assuming that you know you can identify the range with with a purely objective system and now an objective system has a number of advantages first of all there's a kind of you need

to make one assumption which is that the future will be at least a bit like the past if That's true then you can back test them and then you if the future is like the past then you will have a system that will definitely make money there's that repeatability of that you can also analyze the properties of the system I mean the risk properties that the leverage how fast it trades and use all these things in your system design which you can't really do with a subjective system so to me at least anyone who says

that they're trading systematically but Then couldn't in theory at least write a computer program that would essentially replace them isn't really trading systematically that they're de something different so and then of course there's the whole element of trust that you have to be able to trust your system and you have a great quote in your book where you say something like a system which is fully automated but not completely trusted is potentially lethal yes yeah that's true I mean there's been a few high-profile cases over the last few years and normally high frequency in firms

who have a problem with their software or something unexpected has happened of course they've then lost a lot of money very quickly because that and that's the problem with high-frequency trading you know you you can get it written if you do something crazy like buy high and sell low you know a thousand times a second you can Get through a lot of money very quickly yeah it's not just enough to you know if you've got a black box on your desk and it's running and it's doing its thing you need to be completely confident in

what it's doing otherwise anything the moment starts doing something that's slightly unexpected you're immediately going to want to change it and meddle with it and that's why I have a huge preference for systems that are as simple as possible because if a system Is simple you know you'll know that given something's happened in the market today it should be buying or selling and if it does that then you you can kind of be relaxed very occasionally something it won't and then you can investigate and hopefully there'll be a valid reason for that but so you

should have something that you know ninety nine days out of a hundred is doing exactly what you expect and if you got something that's very complicated then you know It's got a non-linearity in it then that's much harder to do it's a shifty little bit and talk about the things you need to avoid when creating a trading system because this is actually something that I often meet when I sit down with investors because they they want to know how you as a systematic fund manager avoid these things and the things I'm sort of the 3no

knows that I that I think you wrote about and that is sort of overfitting over trading over Betting it's kind of the passwords do you use maybe you talk a little bit to each and why these are things you really do need to to stay away from okay so over fitting is it so most people are familiar with although some people use the term curve fitting so this is the situation where you you know you're developing your trading system and you naturally want your trading system to look as good as possible so you kind of

got a couple of ways of Achieving that one way of achieving that is just to try lots of things until you find something that looks good out so you know you might think well I'm gonna try and I don't know a moving average crossover and and you put it in and it does okhane you think well maybe it looks like the same situation is it which doesn't do so well so now I'm going to add a kind of a layer to it and change it slightly so it does a bit better in those situations and

you you Would straight this a few times and you wind up a thing that's really good and the second way is a more quantitative way where you you when you do that process but essentially in an automated way so you do some kind of automated fitting you have a number of parameters that describe how the trading system the haze and then you you kind of search automatically for the combination of those parameters that produces the best performance but it doesn't really you Know that there's no real fundamental difference between two these two procedures so they

both incredibly stupid and dangerous it's just the one is probably slightly more respectable than the other so you know that there's a couple of motivations as to why people do this one is it seems harsh to call it greed but but essentially it's a desire to have a higher fat test performance than is perhaps realistic right and that might be because you know you you're Working in institution they might say well you know don't bother coming to me unless you've got a back test in the shark of at least one I mean that might not

be realistic depending on the combination of assets and the style of trading you're doing so you know you push and push until you get to that the the second reason is I guess comes down to overconfidence you know I think is a fundamental human flaw that affects most people most people think that they know They should be able to get a good bat test and they should be able to make a lot more money than then it's perhaps realistic so for those two reasons overfitting is is rife really it's it's really hard to get away

from and there are there are kind of three three sort of ways of overfitting if I can just yeah so this the first one is what I call explicit overfitting so that's where you've actually got this set up where you know you've got parameters Varying in some automated way the other way which is a bit more insidious is implicit overfitting and that's where you you know you're not doing a formal fitting process but that you you you at least once look at the results of your back tests and there may be some kind of change

and it might it might be something as simple as saying oh I'm not gonna run this system on the US 10-year because that doesn't do as well as the u.s. five-year trouble and the third way About fitting is what I call tacit overfitting and that is where you only ever try things that you already know will work okay and there is absolutely no way to get away from this because you know you're not going to probably sit down at the computer and develop a trading system without having done some research or having known something about

the lives of Industry sure and what that means is you know if you if you plan to be down in front of a computer tomorrow With a completely blank sheet and said right wing's above a train system the first thing I would do instinctively is probably tests and rules that were trend-following I wouldn't test rules that weren't trend-following because I'm you know that weren't kind of trend-following with a minus sign in front and if you like because I know I know instinctively and already and they don't work from and that's the form of overfitting even

though I've done no you Know I've not even looked in an account curve done no kind of statistical optimization it's just from inside my head sure so moving on to overtrading sure yeah let's talk about that so over trading I think is the the Cinderella of these three problems it's the one that gets the least attention and most people know about overfitting and most people know about over betting but as I think relatively few people have a good handle on how fast their trading system is Likely to trade and what proportion of their likely performance

is probably going to be eaten up by costs right and perhaps you know also many people have a realistic idea of how much it actually costs to trade mm I mean it's very easy to say all my Commission's gonna be whatever $1 per lot whatever it is but you know obviously when you then go and execute you're going to have to pay perhaps half the spread if you just cross the spread Straightaway or perhaps more or perhaps less depending on how you you know you do your execution how large your orders are and that that's

that's harder to quantify so you know I'm I'm completely obsessed with with trading costs and one reason for that is that they are a relatively stable and controllable thing to look at so you know it's not really a lot of point in getting excited about your day to day performance because that's essentially a random number and You poke over a long period of time that that will have reach a positive number true but so you know your trading costs are much less variable than you you come and I've analyzed them with with really small amounts

of data and say well am I paying too much have a pain too little and you know this is an area that I think most people just don't even think about if I look at the number of people who are proposing that you should day trade you know trade and multiple times Per day trouble and do so with financial instruments that are relatively expensive like spread bets or you know OTC F at retail FX you know that those I run the numbers and I can't make those systems work on a basis where you're trading multiple

times per day because they are just so expensive to trade well you know it's only really with a few relatively cheap futures markets that I can I can I would be able to trade that quickly So when that's the second mistake drop and I think he's really intact yeah the third mistake over betting is something most people kind of know exists which is just very simple taking on too much risk so you know we can measure risk in many different ways but I I used the expected average annual standard deviation of returns and you know

if you if you look at some kind of the octa space you know that number will be are usually around 15 to 20 25 percent Printers now if you look at the trading systems in a lot of books or on websites or other people people claim to be running you know that number can be the hundreds of percent in you so they're running with with five ten times as much risk as what you know institutional managers were considered to be prudent and you know that's clearly to me and it's way too high and the only

way you can justify that is if you have extremely high expectations of all your Returns or these so again it all comes back to overconfidence sure absolutely now Rob we've hit the one hour mark and I hope the listeners are making lots of notes from what you're saying because it's it's very valuable but I just wanted to let everyone know that actually right now we've only covered chapter one of your book so so that's 25 pages out of 300 so you know if we're going at this rate we're gonna have probably the another world record

for You and that's gonna be the longest podcast episode ever produced so I'm sure we're gonna have two and that's obviously my job is now to focus on maybe some some some few headlines from the book because I also want to try and put it into the context of some of the usual questions that we have but but this is great stuff and and and very very interesting indeed now in Chapter two you go into more of the trading rules and what makes a good rule and Also you talk a little bit about the Sharpe

ratio which you've already touched upon what's realistic and and and much much more so so let's jump into some of these ideas I mean you talked a little bit about it already but when you come up with a trading rule that's kind of two approaches I guess there's one you know you come up with the idea first or you look at the data first tell me a little bit about that and when what people should be be aware of in in in in This instance yeah so I call these two approaches ideas first and data

first yeah so I think people assume that that systematic traders all use data first and so a classic data first approach would be if you used a machine learning algorithm where you you know you get a huge amount of data when you dump it into this algorithm and now magically comes the the trading rules that you should be using mmm The so that's one approach the second approach is ideas first and that would be where you have an idea and then you create a rule to sort of this will be expresses that idea and you

then back test that that rule in in in a similar way so these approaches are both kind of equally valid they have their advantages and their disadvantages I personally prefer the ideas first right there yeah and that might be because of my background so you know I have two Degrees in economics so I'm more used to thinking in terms of how the world works and that there are these kind of underlying economic drivers driving asset returns and that that's kind of what we should be trying to capture but also because I've seen more kind of

misuse of the data first method and more kind of overfitting so you know you you have to be careful about overfitting in both cases and it manifests itself slightly differently in each case but I've seen I think people see data first as a kind of magic bullet and you know you get people coming along who say well I'm gonna just use this big data like method which is the buzzword now and I'm gonna I'm gonna discover something that no one else has discovered before and will be incredibly profitable and you know a big problem with

that line of reasoning I think it's no it's possible and there are firms that probably do use things on your network machine learning Big data successfully but it's it's probably much harder than the new thinking is yeah yeah I like another quote that you have in your book or another sentence you have in your book and you say profitable trading comes out of careful research done by thoughtful and knowledgeable people who seek to understand where profits come from and I think it goes to the point about the ideas first that you kind of have to

know you know why why Does what it does so to speak yes I mean for me that provides an extra layer of validation because you know what I like to know why I'm always profitable hmm because it gives me more of a conviction that it might be profitable in the future hmm and also because it gives me an idea of what the risks might be around it you know so what what you know what what what's that exposed to I mean let's take an example which is probably close to you your heart which will be

The evaluation of the Swiss franc in in January oh yeah I remember that so it would have been very easy to have run a data first approach and looked at the the you know the Euro Swissy rate and saying oh this thing always stays in this narrow corridor between 120 and 125 or whatever it yes you know well that's all we need to do is create a training wall that essentially does the mean reversion between these two levels and it will make a lot of money and it would Have made a lot of money and

then in about five minutes it would have lost you know thousands of times their profits in one go yeah whereas an ideas first approach would be to say okay well you know I know that there are situations in which central banks you know want to keep their currency depressed and therefore you know a carry role should work because you know the classic of course being the the Japanese central bank for decades now as one of The key beginners as low as possible so that's why funding in the yen and the carry traders as generally works

that's the reason why that trade works now if the the the Japanese central bank then turn around and say well we're not going to do this anymore then you may will want to question whether you should carry on doing that rule but the point is at least you you can have that conversation you understand what you're doing whereas if you just look to some Data and so the listing goes up and down and with regularity let's just without understanding the reason why which is that the the Swiss central bank is desperately trying keep the rate

hmm no from breaking that threshold and if you'd understood that rather than just looking at the data first and going no further then you know you hopefully would have realized that doing this trade was an incredibly stupid thing to do Sure very true now some people may have observed and rightly so that we live in a world of constant change and they may wonder you know systems how can they cope with sort of a an ever-changing world can you can you explain that a little bit in your usual good educational manner then then so that

people don't confuse systems with being something that is completely sort of static and and and and needs constant you know change in order to adapt how How does sort of systemic trading adapt so a key question here is what kind of trading system do you have so if you run in something that's trading very quickly then you probably will and will want to have it dancing on a very regular basis right and that's because generally speaking the the kind of short-term behavior of markets what you the economists would call a market microstructure is something that

that's changing very very constantly and fairly Quickly so a high frequency trading wall that worked well a year ago may well not work anymore this other sure now the good news is because you're trading so quickly you have an awful lot of data points that you can look at and you can actually look at six months of data and six months of data maybe you know terabytes or petabytes of data that you know can analyze and come up with a you know the new kind of pattern or system that that's in the markets trouble now

At the other end of the spectrum is the kind of trading that I do where we're ascent I'm essentially trying to pick up on what I think are mainly human behaviors that I that probably haven't changed the hundreds of thousands of years and in that situation you probably aren't going to want to change your system unless you have to so you one are going to want as much data as you can to fit in your system so I'll go back to two Winton which is a big CTA I think actually the biggest CTA is still

true try to get hundreds of years of data and to to sort of fit or at least validate their systems worked mmm the so the depends where you are on this too extreme so one one view is to say yes the world is changing but if my trading system is a relatively fast trading system then I can and should adapt it to cope with that change yeah yep and the other extreme is to say yes the world is changing but there are Certain things about the world that I believe will will not change or at

least will change incredibly slowly and therefore I do not need to adapt my system so it's a question of where trading system sits on that continuum and then you should you should act appropriately so you know if you appropriately means for example if you have got a system that it's trading relatively slowly you probably shouldn't be true trying to adapt it every single Year to the new you know the new world that exists right because you don't have enough data to actually tell you syste cailli that the world appears to have changed you know you

should be using thirty years a date at least if you can if you're using thirty one is for 30 it's not really gonna change the parameters they're coming out at you all you'll get some prices very much at all sure now I have a question that I mean in a sense I I sense from you that that I mean you have a lot of experience in sort of the trend following and you seem sort of to to have a certainly a liking for for trend falling amongst other things but you also talk about trend following

and and and lottery tickets and in both cases you know there there there are small losses are waiting but there are some few much much bigger winners now we as human beings we love to play the lottery but I found very few investors in in my 25-year career that Uses the word love in the same sentence as trend-following so do you have any explanation as to why that is when when these two things have something in common yet they seem to be received very differently yes it is an interesting problem because you're right people have

a strong preference for trading systems that that generally make money every month yeah but every now and then lose lose a lot of money which is the exact opposite to trend following and I think I can have a lot of them a lot of emotional kind of bonding with these people I feel like you know I really understand their pain because from one you know when I look at the returns to my own system inevitably if I'm in a drawdown I feel more unhappy than if if it's doing well and if I've had a day

when I've lost money even though I know from an intellectual perspective that this is just a random number from some unknown distribution of returns that I Hope has a positive mean you know when I have a down day I feel slightly less happy than if I have an up day and if I if I have an up day every day for six months then I'd be much happier and sympathy for that point of view I think the the reap they're sort of reason for their dichotomy between this notional response and the emotional response to buying

lottery tickets is to do with the the size of the pain and the size of the payoff right so when you know if you've Invested all of your money into a trend following fund and you know you lose money for two-thirds let's take an extreme example you know two out of every three days are losing money you could be losing you know maybe half a percent of your net worth two out of three days whereas no one's probably going to buy that many lottery tickets if I was to buy enough lottery tickets the represent half

percent of my net worth that he was you know quite a Few lottery tickets so 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 insure our show continues to grow please leave us an honest rating and review on 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] you [Music]