How Far Back Should Your Backtest Look Good?
Okay, everyone. Welcome back to another episode of line your own pockets. We got another user question. Funny enough, before we get into what it is, I think it was like three weeks ago, Dave proposed this to me of what do you think of this? And both of us probably have incomplete answers, but we haven't.
Michael:I don't know if we haven't decided on an answer. We haven't decided on the best way to articulate an answer. But it's been one of those kind of noodle scratchers that I've I've caught myself thinking about a little bit. So I think it'll be a really good episode because I think what we're gonna do is we're just gonna bounce the potential answer off each other and then probably by the end of it come to come to either a disagreement or an agreement. Think either way it'll be super fun.
Michael:So, Dave, why don't you go over what the what this user brought to you and and why it's so complicated for us?
Dave:Yeah. So this is this came from a a new trader I'm coaching now. We started up a couple months ago. Really sharp guy. Like, he's his name's Tyler.
Dave:He a lot of people that come to me, like, they seem to have a lot of success in some other area, and they're come to trading now. Mhmm. And I like I like people like that because they're super smart and they're motivated. They know they can do it, and they're just fun to work with. So Tyler asked this question, and I'm I'm I'm just gonna go ahead and read it here, and then I'll tell you what I thought as soon as he as soon as he asked me.
Dave:So is it okay. Dave, at a high level, for the trading edge, you know, for trading edge in general, if something is working year to date way better than previous backtests, why do I care what it did years ago? So he's basically saying, hey, I've got a back test. It looks great in this year, so it's been, like, four months or at this point, I think it was probably three months into the
Michael:year. Mhmm.
Dave:Looks great for this year, but when you go back further back in time, it doesn't look good. But why does he care? Like, what does it matter? Like, the last three months are great. So the first thing I thought was, okay, I sorta answered this for myself many, many years ago, and I haven't really thought about it in a long time.
Dave:So it kinda took me aback when he asked it. And I didn't have an immediate answer that I could just copy and paste essentially Mhmm. Out of my brain. So, yeah, I thought it was just a great question because because of that, the way I felt when he asked it and what I thought about. And then I knew it'd make for a good podcast episode because when I mentioned this to you, you were basically like, hey, let's don't talk about it until now.
Dave:Like, let's let's wait and record it because it's going to be such a good topic to discuss, I think.
Michael:Yeah. And I think this is another one of those two that will show the difference between day trading and swing trading a little bit like long term trading and short term trading. But the reason that I started, I think my brain just started kind of circularly thinking about it is that the two schools of thoughts, the two things that came to my brain right away was one, you don't have enough data to robustly test across different market environments, whether your system kind of works over that short timeframe. The other thing was, well, maybe you maybe you found something that is a relatively new phenomenon, and that you're you're just getting right now. And, in which case and you have to and I think this might be one of those kind of intuition or type of things.
Michael:But because if you've, you know, say discovered something that's a new phenomenon in the market, that's something that happens just now in the market and didn't happen a couple years ago. One example is I know a guy who's running these double and leveraged triple ETF strategies. These are fairly new, right? They're in the last couple of years for individual stocks, they've started existing. So if you found an edge there, that's great.
Michael:But what you're kind of saying is that at least in the past, you're able to test this edge and it doesn't work. So I don't and again, it's just it ends up circling around my head over and over again is that you want as much data as you can to be able to test and to optimize and to do all that kind of stuff. But at the same time, what if this thing just started working right now and will continue to work in the future? So, yeah, this is why it's this got me pretty stumped. So I'm excited just to kinda go back and forth with you on it, I think.
Dave:Yeah. So this this goes to the core of trading for me. Like, this this question goes to the core. And because like I've said many times on this podcast, I want traders to have a path to confidence. And that's what I work with traders to create.
Dave:You what do I mean by that? It's easy to trade with small size when the size doesn't matter. Trading is easy, like it doesn't really matter, right?
Michael:Yeah.
Dave:If you're trading small size. But you should be on a path to trading with bigger size. And you should always have in your mind questions that you need to answer to get to bigger size. And, like, that doesn't happen overnight. You're gonna have to gain some experience at the news at higher size.
Dave:You're gonna have to, you know, see how the strategy works. You're gonna have to get acclimated to the new size. Sort of like when you go to like when I go skiing in the winter, it's high altitude. Right? Mhmm.
Dave:If you're not used to that altitude, you you get a headache. I mean, I I you basically get sick. And if it's bad enough, you could get really sick. Yeah. So what we started doing is staying at a lower elevation the first night, and then when we go skiing you know, the the the second day we go.
Dave:So that first night allows us to get acclimated to the higher altitude. It's the same thing when you're trading with higher size. You have to sort of there's a period where you have to get kinda get yourself acclimated just for a whole variety of reasons. Yeah. But so so why is this question so fundamental to that?
Dave:If you're trading something now and it's worked well for three months, but the backtest before that looks bad. Well, it's giving me it's giving you a signal. It's saying, okay. This this this has happened in the past. What's gonna happen in the future when you're in a drawdown?
Dave:What is gonna be going through your mind?
Michael:Mhmm.
Dave:That and the reason it's so fundamental is trading through drawdowns and having the confidence to trade through drawdowns is what makes traders successful. And that because that's when you quit. That like, we talk a lot about how high a failure rate there is among traders. The only time the reason that is, is because they can't trade through a drawdown because they don't have confidence in the strategy.
Michael:Yeah, I would go even further than that. Like it is I think it is everything in trading is is how you deal with because I would in this from past experience, the the kind of death cycle of trading is a little bit of a drawdown going completely off script or sizing up or doing all these dumb things and, you know, taking the ship that way by by having that drawdown, not something that you can understand. You know, for also, from example, right, we came out of a pretty wild couple of months there. It'll be interesting to see if someone watches this back, how tired I look then versus now because I've just got my account out of an equity drawdown and up up pretty nicely on the year. It's yeah, the mental toll there is really, really high.
Michael:So I totally get what you're saying that if you can't for me, it was very easy to go back and say, let me look at the strategies during 2020. Let me look at them through 02/2008. Let me look at them through the year February and say, Okay, this what I'm have what's happening to me right now is perfectly in line with everything that had happened before, so I can I can muscle through, I can go through, and I know that even if it gets a little bit worse or a little bit better, then that's fine? But so that, again, makes total from a psychological point of view. But what about, again, the other side of that?
Michael:What if the person has actually discovered something that is a new phenomenon in the market? So I guess that just comes to the question of how long must that phenomenon exist in order to give you enough understanding or enough confidence to then trade that? Because like the example of of the buddy of mine who's doing this weird stuff with these two x single stock ETFs or even better example, crypto traders, right? I'm building crypto strategies right now. In 2018, there was no data, right?
Michael:And even now, I've only got about five years, six years of really good data. So I guess the more important thing is when is enough? Like, how much is enough? I think that's the question and the answer, right?
Dave:Yeah. I think most traders base their answer to that question on too small of a time period. So and just think about it. Like, yeah, you're suggesting that, hey, maybe there's something new here. Well, it let's say it is.
Dave:It's still too new for me to really have a path to confidence because, like, how do you know that it's gonna last? The only way you know it's gonna last is if you really, really, really understand the fundamentals of the market structure around what's happening and you know that it's going to last, but that's so rare, a trading strategy that works like that and that you would have the insight to understand it. The real way to confidence is to see, okay, how long is this thing at a really good looking equity curve? How many trades are behind that? How many markets does that cover?
Dave:What are the worst drawdowns in it? And what are those gonna feel like? This is the main reason that the number one thing I look at, it's not the only thing, but the number one metric I look at is looking at the equity curve. Yep. Determine if the strategy is gonna be successful because it tells such a vivid story about what you're gonna be feeling when you trade the strategy.
Dave:Look at the drawdowns. Those are pain. Right? That is pain looking you in the eye. You're gonna be miserable there, and you're gonna wonder if this thing even works.
Dave:Is it gonna come out the other end? And, like I said, that's why traders fail because they don't have the confidence when they're in a drawdown. How else would you know? Like, how else would you know if you're going to come out of drawdown? Just because somebody said this is a good strategy and that's why you trade it, that's not good enough.
Michael:And again, I totally agree with it. I think we're probably doing an episode soon optimize for, and I think that equity curve discussion will be something we'll we'll dive into. But I totally agree there as well. The the question still remains unanswered, Dave, like again, so so say so three months, that's okay. Now assuming just based off one of your clients, this is a day trading strategy.
Michael:Yeah. Okay. So that to me is a little bit more interesting. If someone came on and said, I got this swing trading strategy for the last three months, tell them, Get out of here. Right?
Michael:It's just it's nothing, But the market's probably just either going up or down for the last three months and and you're you're lined up with that. But there has to be that that moment in which you can be comfortable. And I think probably, like with most things, the answer that we're both going to kind of land on is further testing is obviously required. So that is either in paper or small, right? So if I say I had a new strategy, if I was in this scenario and I had, you know, my four or five strategies that are working and they're doing fine, and then I think I've discovered something and the recent backtest looks good, and I have some reason why I think that this change has taken place, right?
Michael:I think that's the other thing is that if I just come across something and and the whole back desk sucks and now it works and I have no idea why, then it's it's nothing. But if I have some reason to believe that, yes, there has been some fundamental shift in the market or or I've discovered a new asset class or something that hasn't been been around before. For me, the first thing I'm doing is I'm giving it a small account, giving it like $10 or something, and I'm throwing over there and I'm going to start forward testing, which I don't know, I never know why we call it forward testing. It's just just watching it in real time. But that idea of gathering more data in that way.
Michael:The other thing, I think, is to get just more, you know, psychological about it. What I love about this market is I think it's just opened up another question is if you can find out why it's working now versus not working before. And there is an answer there because there's it could just be random noise that might be more important than this entire discussion, because if there isn't something that flipped, but maybe, you know, the markets like right now, in the last little bit, the markets insanely volatile. So maybe it's that. Maybe it's, you know, the the way the market has been moving.
Michael:So I think for, you know, talking directly to the the person there, if you're really kind of stuck on this and you want this to work, you have to, like Dave Salkmo, get to that path of confidence. And by doing that, you have to figure out and you've got to hack that. Why is it working now versus not working before in a way that's not random? You know, yeah, volatility or, you know, the mood of the market or whatever it is.
Dave:Yeah, I think that so a strategy like that, the one you described, I think that's the right way to approach it. If you really want to trade it, forward test it, trade it with small size, and see how it goes. I mean, that that's another path to confidence. There's basically two paths to confidence. Just raw trading experience where you create a strategy in and out every day and and come to know it and have experience with it.
Dave:That's probably the most valuable. Yeah. I mean, it's definitely the most valuable. The hardest to get, though. But that takes yeah.
Dave:It takes time. It takes a long time. And a shortcut to that is backtesting. Those are the two paths to confidence.
Michael:Mhmm.
Dave:And for something this short, what would happen what I I can I can know what would happen if I started trading this? I would not I I I do not see a path to trading it with large size. I I don't see that path. It's gonna take too long. And so if a drawdown came, I wouldn't know.
Dave:I would be looking back previous to this year and saying, well, there's no question I mean, there's there's nothing that gives me the confidence to trade through that drawdown. So what what I did with this trader said, alright. Let's I understand you found something here. Let's find something that gives us a good equity curve for a longer period with the same strategy, and that's what we've done. So instead of looking back to the beginning of the year, we look back to 02/2021.
Dave:So we've got four years now, a completely different set of parameters to look at. Basically, the same strategy, just taking a different combination of the trades. And if that and and the fundamental reason why that's important is then you have that is enough period for me and for people I work with to get some confidence to actually trade it with size, to be able to trade through a big drawdown because you know or you're never 100% sure, and that's what makes trading so fun and hard. You're never 100% sure, but you can have confidence to trade with size and trade through a drawdown, which is the hard part of trading, and that's the most fruitful part of trading. And that's what's going to keep you in the game longer if you're able to trade through drawdowns.
Dave:And and like I said, the only way to do that is to believe in your strategy, and that takes time and effort to really truly do.
Michael:Yeah. And, you know, the another way to do it is just look another way to look at it is just shelf it, right? You know, you could you could take the strategy and say, Okay, I may be on to something. I'll see you in another three months or whatever it would it would take to go and then work on work additional strategies. Because, you know, it's funny, every single one of these conversations generally comes back to you got to trade multiple strategies.
Michael:It always comes back to that. Yeah. Because if you were, you know, this was your first strategy and your only strategy, then there's a big psychological bias to hope that it works and to kind of force in your brain to say, no, I really did find something that didn't exist before and to push that going on. If you already have profitable strategies that you're running, it becomes much easier to say, oh, that's interesting. And then save it in a folder and set a reminder somewhere to to come back to it in three months.
Michael:If you are on to something, then the equity curve should should look great. Then maybe that's that's a little bit more of the paths of confidence. But unfortunately, all of this kind of stuff takes takes time again, unless you can you can replicate that. Other thing I was thinking while you're talking there about testing across like other markets, right? And again, I need you need to know more about the individual strategy, but it might be interesting to take that from the equity side of things and test it in, I don't know, different markets out there, like crypto or futures or something like that.
Michael:And then if there's edge, because you can either go back in time or you can go wide across different instruments. And that's another way to do it as well, where I've I've taken strategies before where they work well in the futures market, but I don't have there's not enough securities in the futures market to make me really confident about it. So I expanded that market out to futures and then another whole bunch of liquid equities because it is something that would really need that that liquidity. And that gave me a lot of confidence with it to say, okay. If I do it on, you know, Nvidia and Apple and, like, you know, the most liquid names in the market by going kind of wide like that, that's another way to take a look at it as well, I think.
Dave:Yeah. Totally. I mean, that's that's the the the perfect thing to do, and that should give you more confidence. Did did you have to make any adjustments to make it work across all those instruments, or was it basically unchanged?
Michael:It's basically unchanged. The the differences of, you know, the ticks or this, you know, how futures trade versus that. But that was, yeah, that was about it. Then I've I looked at the same. And what I did is I filtered out, you know, top S and P 500 names by volume, because again, the whole point is it has to be very liquid for these indicators to to calculate properly.
Michael:But yeah, so going broad certainly certainly helped with that. It still didn't it didn't do as much, I think psychologically as being able to see through different market periods and drawdowns as it did, but it was it was an okay kind of substitute. And what you may discover there is you may discover because what I end up doing, of course, the the first thing was to go through the most liquid mega cap, you know, billion shares a day traded names. But then you learn something about the strategy because, of course, the next test is okay. What about the next couple and the next couple?
Michael:And then confirm that as you go down the liquidity and the volume scale, the system degrades. So sometimes again, going broad gives you that because at the end of the day, what I think we're looking for is just the most amount of data we can, right? And so we have to figure out different ways to try to get that data in some way. And then those are really the only two that I can kind of think of off top of my head without
Dave:Yeah. I I think maybe. I think going I think going broad is good, but I think going back in time further is even better because and here's why. Basically, I'm lazy.
Michael:Good.
Dave:Laziness is a virtue. Like, programmers like to say that.
Michael:But Yeah.
Dave:Creating a strategy that you can have confidence in is hard. And Mhmm. What I don't wanna do is have to reoptimize and recreate basically, redo the strategy every three months. What I want to what I'm gonna do, I'm trying to create strategies that work for a long time so I don't have to reoptimize and think about it because it's deep thinking. You have to really concentrate.
Dave:It requires a lot of focus because you're answering questions like that, that you're convincing yourself and the more skeptical you are, the better you're convincing yourself that, Hey, this is something that I'm willing to put money on the line for. And that takes time. So I don't want to be reoptimizing all the time. I've got strategies that I've got one strategy that I've basically traded almost unchanged for, like, fifteen years. And that's that that's long for my strategies, but I love it because I don't have to constantly I mean, I I look at it, reoptimize it every few years.
Dave:I don't wanna be doing this every few months. Right? But that's that's my goal when I create a strategy.
Michael:Well, and that's that's more important and compounded on, again, what seems to be the theme most podcasts, which is multiple strategies. Very easy to do if you wanted to reoptimize a single strategy. You do that every week if you know, time wise, if you really wanted to. But when you get five, six, seven, hopefully ten, twenty strategies down the road across different timeframes and different market environments and maybe different securities and all of that, it becomes a huge amount of work. And, you know, if you're taking 10 strategies and you're going have to go through and reoptimize them all the time, you're you're the amount of work and burden you're putting on yourself is just huge, where if, you know, your reconciliation process that you do every day or every week or whatever it is ends up being that you don't have to do any of that, then that's great.
Michael:You're reducing work. And that's kind of the point of systematic trading right across other things is that, you know, we talk about this all the time, you only have so much bandwidth throughout the day, and it should be spent on thinking of new strategies or improvement to existing strategies. It shouldn't be spent on just backtesting and optimizing of over and over again.
Dave:Yeah. So, you know, if you're asking yourself this question about, okay, is three months good enough for for this particular strategy, and this to provide a little more color here, this was a strategy that traded the NQ. That's a futures contract. Yep. So so, really, what I'm sort of reinforcing with this trader, with Tyler, is you need to be a little bit more paranoid.
Dave:You need to be a little more skeptical about what the data is telling you. So I've worked with some traders that, you know, we'll work on strategies. We'll we'll come up with rules, and I'll I'm guiding them so that they can come up with their own. That's a hard process, and it's it's one that I've gone through many times so I can help people with their path to confidence. One of the questions I ask them when they suggest a rule or where they look at the data and say, okay, I think this rule is good.
Dave:Asked them for, okay. What's the what's the clear story? What's the coherent story that this rule is telling you? And imagine that I'm you know, imagine you're explaining this to a skeptical experienced trader. Are they gonna buy the coherent story you're coming up with about, okay, you know, relative volume greater than X, that's a rule we should apply to the strategy because it improves it.
Michael:Yeah.
Dave:Some of the traders I work with will like, I remember this one guy. I'm not gonna mention his name. It was basically like he was, like, answering, like, answering a test or something, like, checking the boxes, like, coming up with this the most surface level story about why this rule would work. And I could tell, like, immediately looking at this, that he's not thinking deeply enough about it. You're you're not skeptical enough about it.
Dave:You you need to be thinking, like like what you said earlier, why? Why does this work? Mhmm. And imagine you're having to explain it to a skeptical trader and convince them. How are you gonna convince them?
Dave:Well, it needs to be it needs to have some validity and not just, hey, the data shows says to do this, so I believe it. Like, it's gotta be more than that.
Michael:Yeah, and it really it's the the word sceptic, I think, has gotten, you know, bastardized a little bit recently, but that is that truly is it. Like you should And this is no different from any other scientific endeavor, right? You go in with the assumption of of I have this hypothesis, and I'm going to work damn hard to disprove it. And the way I look at it is the hypothesis is just the base strategy that, you know, when X, Y happens, there's some sort of positive expected value. And you really need to spend a lot of time and energy trying to break it and see under what scenario it breaks.
Michael:And I imagine, again, not a not a dev, that that's a huge thing with software developers. Right? It's like, here's my software. Beat the hell out of it. And if it survives, then it's it's probably good, and it's probably good to ship out.
Michael:And if it breaks, then okay, then then we go back. That's a lot of the way that I look at at building strategies is I I don't want to prove to myself that the thing works. I want to try to disprove that it works. By doing so, if it survives the test, then I know that I'm on to something. You know, a lot of that could just mean robustness testing of, right, say you've got a, you know, like your ARVOL, right?
Michael:If ARVOL one works, but then arvo 0.9, it just completely falls apart and it's garbage. Then, you know, it's doing this kind of stress testing and pushing on it from as many sides as you possibly can. And then if it survives that, then that really gives you that confidence to say, Okay, this is something that that I got to trade.
Dave:Yeah, the software development example is a good one. I think I remember multiple times dipping some software to you at Trade Ideas and then you would immediately break it. Well, it's just like, as you're a developer, you go through and you can try to imagine what the user's gonna do. As soon as it gets in the hands of a real user, like sometimes they do something that you would have never crossed your mind. It's the same way with the market, right?
Michael:Yeah, we're seeing that a lot. I don't know if you saw the videos in self driving cars now where people like a dude would just for, you know, shits and giggles was walking down the street and he had a jacket on. Under the jacket, he had a stop sign on his shirt. You just open up his jacket and the cars would just grind to a halt. And it's yeah, it's a lot like that.
Michael:We don't know what the future environment of the market's gonna be. We don't know if this is gonna be an incredibly volatile market or a dead market or an upmarket or a downmarket. So what we need to do is we need to try to, and that's I think, you know, we keep coming back to there's no substitute for a lot of data, because if you go back far enough, you've seen every kind of environment that there is. And without that, right, again, it becomes really, really hard. So yeah, going into the process of I'm going to break this thing, I'm going to prove to myself that it doesn't work.
Michael:And then if it stands, then then great. I think that's by far the best mentality you go to. I see too many people doing the opposite where they're, you know, they're trying to make just the equity line point up and not not understanding that you're just, you know, you're torturing the data till it tells you whatever whatever you want.
Dave:Exactly. Yeah. So there's a really good quote by one of my favorite quotes by John Feynman, famous scientist from, I think, the sixties or seventies or eighties. It's something like your goal is to not fool yourself, and you're the easiest person to fool. Mhmm.
Dave:So it's
Michael:one of those reasons why I can't I can't do fundamental trading at all or or story trading, I think is the kids are calling it because I'm the kind of guy that I would listen to an earnings call and all the the BS that the CEO is always going to talk about how great everything is and how they're growing on all aspects. I go, this sounds really good. I'm going to hold on to that stock. And I just I can't do that because I would be too easily convinced, I think, by by, you know, pretty words and whatever. And, you can do the same thing with data.
Michael:If if it looks nice enough, you can convince yourself that you're onto something when you're really you're really not.
Dave:Yeah. There's I don't know if we talked about it on the podcast before, but there's this so in in economics and in other fields, in the social sciences, there's sort of a sort of a crisis because there's a lot of publication bias in the field. So people you know, you see these studies like the marshmallow study from the sixties. It's hard to recreate that. Right?
Dave:And part of it is because there's this publication bias. Like, people you know, you don't get articles published unless they say something that sounds cool. Right? Or sounds interesting and believable. That's the incentive, is to say something good no matter what the data really means, right, or what the truth really is.
Dave:Well, the one the things I love about trading is there is no publication bias. The only thing that matters is P and L, and you learn pretty quickly that there's no sense in fooling yourself. You have to really truly understand it to have a path to confidence to trade with size, and that that's hard work. And and when you're looking at the data, it's easy to be fooled, and that's the the last thing you wanna do.
Michael:Yeah. And I don't that's why I always you know, the biggest thing to talk to someone at the it's like, at the end of the day, and I always say, when you put your grandma's money in in this strategy. Right? Because there's one thing to say, right? Oh yeah, you know, built this thing.
Michael:I'll put my own money in it, but some people are flippant when it comes to their their capital and they're Okay. You know, maybe we all owing or taking some risks. So I always go, Okay, have you built this strategy in a way that you're Okay putting your grandma's money in it? Then all of a sudden it's like a great little stick up the butt there a little. Oh, Okay.
Michael:Well, then no, go back to it, right? If the strategy is good enough, you should be totally Okay, you know, putting, you know, to get around all the it's probably not good investment for your grandma to be to be day trading the market in her age. But the yeah, if you wouldn't do it with someone else's money, if you wouldn't do it with something that's, you know, a little bit more risky, you could say their kids or something like that. Then, yeah, because your goal is to reach truth. And, you know, again, it's very easy to fool yourself with with numbers and with data.
Michael:But at the end of the day, hopefully, you're gonna load this into a robot, and that robot's gonna do everything that this system tells it to do. And you're right. At the end of the day, you're either up money or you're down money, and doesn't doesn't matter what the equity curve said at that point. You can't eat an equity curve, right? You can only eat the money that comes out of the account, so.
Dave:True. Yeah, that's good. So there's it reminds me of this can't remember the site. We'll find it, put it in the show notes. It's basically a site that shows various correlations between different data points.
Michael:I love that. It's like the amount of what was it? The amount of gold medals won in The US versus like the amount of people who get choked by their own blankets at night or something. Like, two things that make absolutely no sense, and they make the lines just perfectly right? Because there's yeah.
Michael:There's all these yeah. I I know which one you're talking about. I love that side.
Dave:Yeah. There's tons of them, and and it kinda brings it home. Like, it's if you were just following the data, you would think a whole lot of things that just weren't true about the world by looking at that. So definitely brings it home.
Michael:Well, and I think it also shows why people why there is it's very important to have some market understanding before you get into systematic trading, you know, not necessarily that you you have to, you know, discretionary trade for a long period of time, but some observation of the market or some, you know, like core fundamental reasons, because then if you do something like there's some people out there that legitimately wouldn't know that, you know, the quarterly dividend growth of a company is going to work for a five minute scalping strategy. There's some people that really just wouldn't have that that core understanding. So I always think it's like that. It's like, you know, the same as I wouldn't want to put somebody in the driver's seat of like a self driving Tesla if they've never driven before, because, you know, what if they needed to do something or what if I yeah, I think you have to. And it's one of the reasons and I know you're going to highly disagree with this, but I keep my own discretionary trading account open.
Michael:And it's tiny, and it's literally me just going through charts. Like I still like to do the, you know, macro analysis of the market and all that kind of stuff. And I take these I try force myself to take one bet a week Just completely, I see this happening, you know, this thing's breaking a trend line, just completely kind of discretionarily. And A) it's to kind of not ever get myself lost in the data. It pulls me back to what's moving the market is certain certain events that are that are happening out there.
Michael:And I find it a good way to come up with new ideas as well, because sometimes you get a really good winner out of that. Natural gas was one that I had, for example, and the basic idea was it spent tons of time under the two hundred day moving average and like three or four years, and then it finally broke through and broke through on like good volume and a good candle, I bought some there and I ended up making like 50% on the trade. Was an amazing trade. So that instantly your systematic brain goes, oh, well, let me go test that. Turns out it's not very good, but it's like one of those things that, you know, you have to be able to do that.
Michael:But I think by grounding yourself in that or pulling yourself out of the numbers every now and then, I think is important. And again, I know you're probably not gonna agree with it, but I think it's a good exercise to continue to have.
Dave:Well, I think so a couple things to say. One is I do see a lot of traders get stuck in back I call it backtest world where
Michael:Mhmm.
Dave:They're not looking at the market. They're looking at a backtest. And you're totally right that you you need to watch things in real time and watch them play out. I don't think you need to desperately just trade discretionarily, although you can. I think that's a great idea.
Dave:But you need to watch things in real time and see how that how they show up in the backtest later to really kinda get a sense of what the market is and how the the it connects to your backtest and what flaws the backtester has and how to come to grips with that. But, you know, I I will say, I think you sorta underscored the whole point about confidence there because that account you're trading, you said, was small compared to what you're doing with the other stuff. Right? Yeah. It's because don't have a path to confidence with that kind of strategy.
Dave:If you did, you would like it would take some work to trade something in the way you're doing there with a lot of size compared to the other stuff you're doing.
Michael:Yeah, because it's one of those. Yeah, I can't test the hypothesis. It's just more of having having a feeling and wanting to take a trade off that. And I'm just I'm shocked that you said it was a good idea. Thought I was gonna get I was ready for my to get hit with a ruler or something.
Michael:But yeah,
Dave:I think Market intuition is really super important. And yeah, the more you can get at that, the better. And especially for systematic traders, I think that's very important that like it's gonna make you way better when you have the skills to backtest and you have the market intuition to go with it. I think that's really powerful.
Michael:Well, I think it should if you know, if we are talking to any discretionary traders out there right now, I think it also proves the path and the end goal should always be systematic trading, right? So every time I take a discretionary trade, if it ends up being a great trade, the first question I always ask myself is, is this something that I could I could then, you know, automate and test and go through that? And sometimes it's not. Sometimes it's just, you know, I know a geopolitical event's happening and a chart looks like X and, you know, I take the trade and it does well. And yes, those are relegated to smaller accounts because that's kind of I look at those as flyers, right?
Michael:They're not really going to be consistent, right? They're in the long run, that account does make money and it will make money, but it's just not something that I could it has too much to do with my brain and how I'm feeling and interpreting news and things that I can't quantify and put in a box and then, you know, put that box on the counter. But yeah, I do think there's some help there. I also like what you said too about about watching it live. And that's something that I think is very, very important where people, they want to set the bot, they want to go, you know, do the beach or something every day.
Michael:And I think you can I think you can do that sometimes, right? I think it's but I think it's important to have an exercise of I'm going to watch a trade comes through. I'm going to click on the chart. I'm going to watch the thing as it develops. And the beauty of it, it's not intense work because you're not doing anything.
Michael:You're just you're just watching. So being able to put yourself kind of in an observer mindset and have that detachment of it's not my trading, it's this weird nebulous robot, even though I'm the one that that made it. Yeah. It does a very good job of saying, Okay, let me see what this does. And you can just kind of watch it detach.
Michael:And again, it's it's really chill work. You could have a baseball game up or something on the other monitor and be totally fine. But yeah, I do think watching the market day after day is very important because you never know where your ideas are going to come from.
Dave:Yeah. And, you know, there's there's some simulators you can do that sort of simulate that, but it's just not the same. Right? You have to see it in real time and watch it play out.
Michael:Well, and with the I think with the P and L associated to it, right, because that makes it way more way more engaging anyway. At least at the very least, I would say a little bit around when the the trade is entered and a little bit around when the trade is exited. If you have limited time, those are going be the two most important parts. But I do think there's something to having it going and watching it any anytime you can. I know a lot of people get into systematic trading because they have other jobs and careers and that's fine.
Michael:But as much as you possibly can, I think it's good to watch the trades play out?
Dave:100% true. Yep.
Michael:Well, I think we arrived at an answer. I think it was a roundabout way with the basic answer was no, it's not enough data, but at least hopefully we gave you some suggestions on on how to potentially deal with it. We came up with a bunch of them, from going broad to, you know, testing it small to, right, shelving it for a bit and bringing it back out. Lots of different ways to do it. But it doesn't mean abandon it entirely if it's only working for that period of time.
Michael:But like anything, this is yeah. Go ahead.
Dave:I'll say one thing, but I just wanna make sure people are clear. I'm not saying that three months is too short a time for any strategy. For this one, because it traded one instrument and the the equity care before that was bad, then that by definition is not.
Michael:When I even
Dave:think But there's other strategies that have multiple equities where three months is actually plenty. So it all depends. There's no hard and fast rule that we're coming to here that people should take away from it.
Michael:Right. And I think that the main, even for me, the main takeaway is the previous equity curve was bad, right? If you had no equity curve and you only had that three months period, that might be a little bit different. But the fact that you had way more evidence that the strategy didn't work than the strategy did work right again, the either you discovered something new or this is just like a glitch in the matrix and it's kind of working right now. So yeah, 100%.
Michael:But again, it's the the thing I love about systematic trading is that. Some people would come in with a question like that, they'd expect a yes or no answer. And then we gave like seven different ways you can get to test and ask more questions and dive into the data deeper. Because we we don't know, right? It could be that, you know, it could be a new phenomenon.
Michael:It could be this could be that. But the only way to know those things is getting further further testing. It's funny, it ends up being I think the majority of answers is, I don't know, test it. I've no idea, you know, go back in. Think that's going to be the way to do it for me personally.
Michael:I think it would either if I was very confident, I'm trading small. If not, I'm shelving it and I'm checking on it and I'm setting a calendar, right? I'm going to my calendar and I'm scrolling a couple months away and I'm writing a note to say, Hey, check back on this and see how it's doing. So I hope that helped. Do you have any other words of wisdom before we end it?
Dave:I think that's I think that's a good place to end it. The the the thing I'll underscore, what we said here about the path to confidence, it's not gonna be the last time we mentioned that here because I think really grading fundamentally comes down to that. And and that's why systematic trading is so enticing because you can get a path to confidence, and you can get it quicker. You don't have to wait for years and years of trading experience to go by. It's sort of with back testing is kind of a shortcut to confidence.
Dave:Can think about it.
Michael:Well, awesome. Well, again, you can tell how much we appreciate comments and questions and all that kind of stuff because sometimes you stump us and we've got to come online to hammer it out. So I really enjoy these where where it's not a predetermined answer, right? We have to kind of pound our heads to get together to get some ideas around it. So I really, really like that.
Michael:So as always, I'm Michael Noss.
Dave:And I'm Dave May. Thanks for joining us. See you next week on Line Your Own Pockets.
Creators and Guests
