How Often Should You Re-Optimize?

We discuss:

• How to optimize so you have to re-optimize as little as possible
• How better columns directly translate into more robust strategies
• Hold a rule "in reserve" so you can apply it when needed without full re-optimization
• Documenting previous optimization decisions will make the next strategy better
• And much more...

---

Before you go!

The next time you find yourself in a drawdown, I want you to stop what you're doing and head on over to betterbacktesting.com to sign up for my free backtesting email course.

I hope to see you there!

Creators and Guests

Host

Dave Mabe

Host

Michael Nauss

What is Line Your Own Pockets?

A weekly show for systematic traders who want to make more money from their trading strategies.

Michael: 00:00

All right, everyone. Welcome to another episode of Line Your Own Pockets. This is a question I know Dave's gotten a bunch. I know I've gotten a bunch about re optimization, but I think this is spurred on by a more recent question from a client or listener. So I'll let Dave take it away with chats about re optimization.

Dave: 00:19

Yeah. So it's a very often a question I'll get. How often do you re optimize your strategies? And I can tell that my answer is very surprising to people because of how they phrase the question. Like the most recent one I got about this was, do you reoptimize every day, every week or every month?

Dave: 00:39

And

Michael: 00:41

No. No minute or hour, like every every losing trade, every every suboptimal exit, you know, is that it's funny because I don't know. It would be very uncommon, but I would not be surprised if somebody out there, maybe even someone listening to this does every day where they go through and they're they're just tweaking and kind of playing all the time. But why don't we have your answer first, and then we'll we'll dive into it a little bit.

Dave: 01:10

Yeah. So I knew my answer was gonna probably shock this guy, but I try to never do it. I mean, ideally I'm creating a strategy and applying rules that are consistent across a large period of So that's always my goal is to not have to reoptimize or have it be unusual that I have to reoptimize. So that's always my goal is to do it so well at the beginning so that I don't have to reoptimize. And there's some things I do about that process to make that happen that I think are pretty unique.

Dave: 01:52

So I think it's a good mindset to be in to create rules with that in mind.

Michael: 01:59

Well, I'm the same. I try not to reoptimize. So I guess that's the episode and we'll we'll end it there. But no, there's I think two major kind of branches. When I heard this question, this topic, I thought I would take it that I think makes sense.

Michael: 02:13

And the first is just what you do to try to avoid it from the get go. And then the second that we can go back to in a second is what will actually make you, you know, get to work and potentially reoptimize the strategy. But let's just start with, you know, if someone's finding they have to reoptimize very frequently, what do we think went wrong? And, you know, what would you kind of suggest to that person to say, if you're feeling the need to, or if you actually kind of have to because the strategy is not performing well anymore, what would you tell that person to kind of avoid that going forward when they're building their next strategy? What kind of steps can they take to make sure or do the best they can to make sure they don't need to reoptimize that frequently?

Dave: 03:02

So, yeah. So the first thing I think, I get a lot of traders that are back testing maybe a futures contract or one instrument. And when you do it that way and when you create a strategy for that, there's just not that many signals that are going to occur. Your starting point is going to be a small number of trades anyway, relatively speaking. So, in that case, I would try to add more trades to the strategy or go back further in time for your backtest.

Dave: 03:45

The goal will be to try to get a lot of instances of the signal and the trade and use that as a starting point. That's one thing.

Michael: 03:55

So step one is you probably didn't have enough data to begin with essentially to break that down. If you're finding that you, you know, did your strategy and it looked like it worked well, but it just continuously is breaking right off the bat, or you feel the need to go and kind of change things right off the bat, I would agree. Right? There's there's I I know where you're going next, and this is where I'm I'm gonna agree as well. But, yeah, step one is, you know, if you only have a limited number of signals, then you don't have enough data to really do that much optimization at all.

Michael: 04:27

So when you're optimizing, part of it is going to be, I wanted to say guessing, but we'll use a better word of like intuition of things that could affect the strategy, because you don't have enough data to really just kind of let the numbers decide of this is what's important and and this is what's not important. So that I think becomes a little bit more of an art and less of a science. You know, if you're back testing something that has, you know, maybe a couple 100 or a thousand signals on like the ES or or some sort of futures contract, You just kind of have to look at your data points and say, you know, let me think of what I think potentially could affect the strategy and go from there, which if that's all you got, it's what you got to do, but it's certainly not optimal.

Dave: 05:09

Yeah. And I'm glad you brought that up because it is a good way to think about it. Like if you don't have a lot of data, you are using your intuition. That can be valuable and people develop very good market intuition over time and can do that. But it does take a special kind of trader to have that be your path to confidence, like really being able to trade a lot of size based on your intuition and to really have that level of confidence.

Dave: 05:43

I know that I don't have that. If I'm going to scale something up, I need to have some hard numbers that I can believe in. I just don't trust my intuition. Know some traders can get to where they could do that. I just And maybe I could, but I found it just so much easier to have confidence when you look at data and you have these hard numbers.

Dave: 06:09

And the more data you have to look at, the less you have to rely on intuition. So I've said this before, backtesting is kind of a shortcut. It's kind of a cheat code to a good strategy. It's very, very valuable to be able to do it.

Michael: 06:27

And I I guess also what we mean by intuition, just to break it down when people are are figuring out, like, say, I'm I'm building an open range gap scenario on on the queues. Right? And that's that's the only thing I wanna trade is the Nasdaq version of the queues. You know, that in by definition can only happen really once a day. So then you take the amount of days you have in backtests, and and that's your your data source, which if you can go back a long period of time, that's that's great.

Michael: 06:56

But still, you're not getting multiple signals a day. You're not getting across different assets. So the strategy you have, you're gonna have a little bit less confidence in. And then what you're doing is when you're optimizing, you're kind of noticing things and then saying, oh, well, you know, maybe when it gaps up, it works better. When it gaps down, it works better.

Michael: 07:15

And that's all well and good, but I've, in my personal experience doing that, if, like Dave said, if you don't have the confidence, you're kind of chasing the dragon where you're looking at just maybe because you need now need to wait every day for one more data point. So maybe once a week, you're looking back and saying, well, every trade lost this week, so it must be something that's going on. You just don't have the confidence to and the data set to go back and say, well, that's normal. That just happens and and move on. Yeah.

Michael: 07:43

So when you're in that drawdown of you've lost every single day this week, it just becomes really, really hard to go in unless you are been trading the cues, I guess, for so long that you have such an intuitive feel on it that you're like, yes, I know I'm in a drawdown now, but I've just seen this work so much and I've traded it for so long that I can go on. Which, again, it can be done, But, you know, it's that someone who has to be kind of intimately aware of that right from the get go and then loses, I think, a lot in the process where, you know, if you were to say, hey, I want do an opening range breakout on everything, well, now you're talking that you could have, you know, up to a thousand triggers a day to to then run datas and optimization on, and that becomes just a way easier to say, okay, well, now that I've got all of that data and such raw data, I don't need to rely on intuition. I can just let my filters or let my columns that tell me what's most important.

Dave: 08:40

Yeah. And also, you have it's almost a really probably the best path to confidence is having some intuition and then creating a back test, gathering some data that confirms that intuition in some way. You have both aligning in a way that's really powerful and that can be really confidence producing. So I think that's a really powerful way to do it. And we've talked about collaborating with discretionary traders as quant type traders.

Dave: 09:20

That can be a really powerful thing to have discretionary traders that typically have a lot of intuition joining forces, collaborating with, sharing ideas with quant traders. And if both sets of those traders can have empathy with the other one and understand how the other one thinks, you can do some I mean, could just have some really magical collaborations and really come up with some good stuff.

Michael: 09:47

Well, and I I think you you find that a little bit naturally on your own when you're building systems. You know, the the best systems that I found come from you either see something in the market or you're reading something or whatever, and there's that kind of instant, like something just triggers in your brain. You're like, you know, that's a really good idea. I should then go kind of test that as opposed to you see some things come through and you're like, maybe it will work. And then you test those.

Michael: 10:14

There's I find those will work way less. But when I see something and it just slaps me in the face and I go, I need to test this right now. I think that is kind of just the inbuilt because there was some generally, that's not just because two moving averages cross, right? There's some event that occurs in the market that you understand why it would happen, and it's not just that the chart pattern exists. You go you understand kind of the psychology behind the market or or the participants, and then something just hits you and goes, okay, I have to do And then if you test that and it turns out to be wicked right away, then those that's, I think, the way a single person can combine those two things when it really you you understand perfectly about why, you know, hey, I'm I'm trading junky stocks when they break down after a gap up.

Michael: 11:02

It makes sense. They're crap, and people are selling them off the open or whatever it ends up being, but you can kind of have a little bit of a story and then the data confirms the story, then I think that's like you're you're really onto something there. That's the for me, the best kind of strategies.

Dave: 11:16

Yeah. I agree. So so I think we're getting a little let's let's bring it back to the original topic because I think it's a really interesting thing to think about. So I have another trader that I'm coaching that he prefers to remain anonymous, really great trader. And he has a really good strategy.

Dave: 11:41

He's trying to optimize more frequently to get rid of some of the worst losers from the system. And so he's trying to find what I would call sort of local minimums or local maximums to figure out how to avoid some of these big losers in a way that's more optimal than looking at the entire trade set. And I think that I've gone through this before. It's really hard to do. Whenever I see I was just thinking about this this morning as I was thinking about this topic.

Dave: 12:28

Whenever I see inefficiency, even if somebody else is doing something like, gosh, that's so the wrong way to do it, it just bothers me to my core. Oh gosh, the world could be so much better. Your world could be so much better if you did this differently. And I know that his life would be better if he just looked at the broader picture. Because it's so hard to gain confidence in something that might be like, let's say you have a backtest and you backtested it off ten years of data.

Dave: 13:07

And now you're trying to look at smaller segments to figure out different rules to apply in the now. Well, that's just a really hard problem to make those adjustments. Each time, I mean, every time you do an optimization, you're thinking about those rules and you really need to understand the why. Why now? Why are you applying this rule now?

Dave: 13:33

Does it make sense? Is it a temporary thing or does it hold up over time? So by definition, if you're using a smaller time period, you're inserting yourself into this world where the world is gonna change a lot based on what you're choosing. It's gonna be really hard to have a path to confidence for this type of setup.

Michael: 14:01

Well, and I think you've hit on the crux of the problem with optimizing too frequently, right? There's two ways or two reasons that I think you could really reoptimize a strategy, and one is that you've learned a lot more since you built the original strategy. So when you're going back to that original dataset, there's more that you want to test on it or more that you want to do, but you're still playing with that really original bulk dataset. I don't think that's that bad. I I don't think there's any real problem with that at all.

Michael: 14:33

It's like, oh, we talk about it all the time. You add a new column to your library and you're like, oh man, I really just want to go see what this affects my original strategies. That's one thing. But I think where people fall in problems when they're reoptimizing is they're only doing so because more data has come in. And kind of by definition, if you're doing that frequently, then not much more data has actually come in.

Michael: 14:56

If you're re optimizing even once a month, well, you're basically saying, hopefully you've tested your strategy out a number of years, and now you're saying because one more month of data has come in, now I want to reoptimize. And that to me doesn't make sense. Now, again, if it's I've discovered something different that I want to apply to the whole dataset, that's one thing. But the person you're talking about kind of just seems like, so they're incorporating, you know, every couple days or even once a month or something, the amount of data that has come in, that's just not enough, right? You're weighting probably that most recent dataset way higher than you are your full dataset.

Michael: 15:36

And to me that seems like a bit of a fool's errand. And like you mentioned, you spend a lot of time. That's that's the other thing is you always gotta look at your time as being finite and saying is reoptimizing something that I've already done. Is that the is that the best use of that of that time or or is building something else a better use and and a better kind of profit potential in the long run?

Dave: 15:59

Yeah, I think a better use of time is building out your column library, watching trades that come through your system just to really try to understand what your strategy is missing, what columns you've got to find a number of columns. It's capturing the movement of these trades and the situation, but it's not capturing all of it. There's something there that's probably predictive that you're not looking at. And the more you look at the charts and watch the charts as they happen in real time and in the review each day, you're gonna notice stuff and you're gonna be able to apply that to your column library. And then you can add that and like you said, go back and test it and see how it compares to the other columns in your library to see.

Dave: 16:51

And the goal is to come up with something that's highly predictive. And once you do that alone can turn into different strategies in and of itself. Having that ability to have a column that is highly predictive.

Michael: 17:08

So that's, I guess that's kind of explaining people who just want to make a good strategy better. I think we should also address the time that you want to re or that you have to re optimize a strategy kind of even when you don't want to, right? And what's basically what I'm thinking about is that you have a strategy that's kind of falling apart and you want to go and explore to see if some sort of reoptimization could benefit it in some way. Is there, you know, we're no longer talking time because this could be ten years before something like this happens. It could be six months.

Michael: 17:43

You don't know. But what is kind of that first initial threshold where you say, okay, something is going on here, and I think I could go back and maybe strip it to its base and then start to build it back out to see if there's maybe something in the last period of months or or years that I've learned that I can now kind of fix what's going on with that strategy. But what's that first signal to you to say, okay, it's time to time to dig into this one. And this one deserves my attention a little bit right now.

Dave: 18:12

Yeah. It's a common question that traders will ask. Do you you know, When do you turn a strategy off? Is there some threshold? There's not really a threshold.

Dave: 18:25

Is I can hear you saying, Dave, a client. You should have some sort of threshold that you meet. There should be some sort of rule for this, but it's not. There's definitely a lot of feel here and a lot of emotion and intuition about it. And it is just, is this strategy performing like it used to?

Dave: 18:50

Does it need some work? Am I getting as much profit as I was before from the strategy? And is this the strategy to be working on right now? Is there another one that I should be working on or has more potential or makes more trades that I see has more potential for profit that I should be working on now. So it's a complicated question.

Dave: 19:14

And a lot of this comes down to how you prioritize your work. We've talked about it a lot. New traders are starving for ideas that they feel like that is the bottleneck for new traders. Once you get some experience and have some strategies that you've been trading for a while, all of a sudden this avalanche of new ideas comes into your head and it's like a constant spigot of new ideas. So the real problem for experienced traders is prioritizing your work, figuring out which thing to work on now.

Dave: 19:53

And like you said, strategy doesn't tell you, Hey, I'm gonna work for two months and then I'm gonna fall apart and I'm gonna need your immediate attention then. Like it's always immediate and you're like, gosh, I'm working on this other thing. All of a sudden this needs my attention. So, okay, which one do I do? And yeah, it's a tricky thing because yeah, the timing is never good for these, for when the situation happens.

Michael: 20:21

No. And like we talked about, the main hedge against this is just multiple strategies, right? If you've got, you know, 10 strategies going on and one starts to suffer, you don't you don't feel it in the way if you have one strategy going on and the other one suffers. But yeah, I couldn't agree more there where I was in the same boat when you you have a hard time thinking about what you're gonna build and and now I am a 100% convinced I will die before I get to everything on my list because it's just not it's gonna grow as as kind of time goes on, which is both the exciting part of it and then also kind of a bit of a burden because you're right, what you got to look at is how much is this strategy degrading And is it worth your time versus just going through and building kind of a new strategy? And I think that would be the difference of whether or not it is just performing worse or whether it's actually hurting and losing significant amounts of money.

Michael: 21:20

I think just a quick and easy rule of thumb just to before you do any alarm bells at all, I just say, look at your prior data and if the drawdown is within, you know, some variance of what your normal drawdown is, right? Say the max drawdown before and the strategy was 20%. If you're within like 30%, I don't even think you look at it, you know, it's only when that starts to exceed that level, and then you can't really figure out why there's not some sort of kind of structural change or difference to the market or something that you think may may kind of pass through. But it is outside of whatever your testing date is responsibly. And that's where I think most people that I've talked to, they get into some problems is, you know, they backtest something and, you know, say it has an average drawdown of 10% and has a max drawdown of 20%, and the thing's drawing down 5% and they're already looking to go reoptimize.

Michael: 22:17

And that's that pain avoidance thing that we talk about all the time. There's no intellectual reason to why to do that, but you're going through and you're saying, well, I don't like being a drawdown, this sucks. And it's way different to see the 20% drawdown. I think Marsden Parker talked about this when we had him on the show is that there's one thing to see the drawdown in paper and say, Oh, I could deal with a 30% drawdown. Yeah.

Michael: 22:41

And there's another, it's another side of things to see your account down 30%. So yeah, I think if you're inside that boundary, then you just have to kind of sit back and take the pain, take the

Dave: 22:54

I think it's really good when you optimize your strategy, when you basically, you know, when you create your strategy and get it live, there's a couple important things that I do that I think traders could also do, and that helps in this situation. One is I have some additional rules that I hold out for the case where, okay, do I need to improve this down the line? So instead of starting from scratch or re optimizing with something when the strategy starts losing money. I've got a couple rules there that I've already pre planned to apply that will make the strategy better, although make it trade fewer trades. That's extremely valuable to have.

Dave: 23:47

Mean, so you're not because I've been in that situation. You're in a drawdown, you feel like your world is ending, you feel like you're at a complete dead end and you're completely unprepared when that happens and that sucks. So to have that, to prepare for that ahead of time is really extremely valuable. The other thing to do is kind of along those same lines, have some preplanned sort of notions for when you will reoptimize, when you will change sizing. It's especially good for sizing.

Dave: 24:25

You know, we've talked about scaling up strategies on this podcast and another good thing is to have a plan for scaling down that, you know, if you, you know, if you've been trading a strategy for a long time, it's often, it's often psychologically difficult to scale down just because you feel like you're kind of giving up. And it's not all or nothing. You don't have to go from trading something with full size down to zero. There's a whole range that you can do, like scale back. And a lot of the emotions that you have about the strategy will ease a bit when you trade it with smaller size and you could, it's it's like a better it's like it's like a better mindset to have when you're trading a bit smaller.

Michael: 25:14

It's it's very right, but it is it is very hard because it's it's almost like, you know, if you buy a stock for an investment and it goes down a bit, your inclination is never to sell part of it when you're underwater because you're like, well, you know, if I just buy more, it only needs to bounce a little. So it's yeah. The natural human inclination is to do the exact opposite. Right? And just to get into that more of that kind of pain avoidance thing, right, is when you see a drawdown, it is one of the hardest things to do to down your size because just mathematically, it's it's likely gonna take you longer to get back to even and then and then back to to making money.

Michael: 25:57

And like a lot of things in trading, it is it is totally counterintuitive and certainly not fun at all. But it's, yeah, it's a 100% right, is that as soon as you can, you know, the old trading adage is that the first thing you do to get yourself out of a hole is stop digging, right? Go to the point, just lower the size where you can kind of have your brain clear. And then you start to work on things when every you know, you're no longer getting kinda slapped every day if you're a day trader, every swing, or every week if you're a swing trader, you're just getting little kinda taps. Right?

Michael: 26:34

You're not getting hit. And then either one or two things were happening. Like it was just a normal drawdown, the strategy will recover on its own, and then you size back in, and then six months out, you don't even remember what had happened, right? It's just so far in your rearview mirror. Or it continues to suck, and then you stay at a small size and at least you've kind of given yourself way more leeway in order to fix it before it becomes something that's like mission critical by doing it.

Michael: 27:02

But yeah, just very, very a 100% correct, but that's very, very hard. If you're doing that, I kind of applaud your will because it's correct and you should do it. So forcing yourself to do it, think is ultimately a great thing where you can go and say, yes, it's not working, I'm gonna size down as opposed to going on tilt like a lot of people do.

Dave: 27:23

Yeah. Well, this is incredibly difficult even more when you have a small number of strategies like one. So the worst situation I've ever been in, I almost quit trading. This is back in the early twenty teens. My strategy was working really well.

Dave: 27:46

I had a drawdown one year for part of the year. And I scaled back and reviewed my results afterwards. After I came out of this drawdown, I realized, man, if I had not scaled back, look at how much money I would have made. So, that wasn't the problem. The problem was the next year when I was like, I am not scaling back this time.

Dave: 28:10

I've done this. I've made the mistake before. So, I was ready to keep my size really large for this strategy. And I ended up in this really big hole and finally cried uncle. Mean, that was totally depressing.

Dave: 28:29

It was really, know, drawdowns can be incredibly motivating, but they could also be incredibly dangerous. So you have to be able to have the right mindset to turn a drawdown into positive thing. And that's a really hard thing to do when you're losing money. And that's the only strategy you have. And it's worked so well in the past.

Dave: 29:00

I put my kids through college with this strategy. To give up on it was just so brutal or to even scale back, especially because I'd done it before and I felt like it was a mistake. So yeah, a really hard thing to do.

Michael: 29:18

And yeah, and again, I know we're hitting this on the head all the time, but the most important thing is that multiple strategies, right? Because you know, if in that same moment you had two strategies even. Right? And and it's it's really it's kind of, I wanna say, like, inverse exponential. Right?

Michael: 29:37

Because you think about it. If you have one strategy and you're applying all your buying power and time, whatever, it's all one. Soon as you get to strategy number two, well, that's all that split in half, then strategy number three is gonna split. So, you know, there's there's a point in which each individual strategy isn't making that big of a deal. But if you're starting out, that first, like, two or three additional strategies to get yourself down makes all the difference in the world because, yeah, in that same moment, even if you just had two, which still isn't a whole lot to go off of, then at least, you know, hopefully one is treading water or helping with some of the losses of the other, or even best case scenario, it's making up first, you're still somewhat positive, even though it's falling apart.

Michael: 30:23

And, you know, it goes back to the it's very easy to get very lazy when you're on a hot streak with a strategy. You got in the hot hand, you just open up your computer every morning, you turn on the bot, and, oh, look, I made I made money today. And you close the bot, you walk away. That is the easiest time to get incredibly lazy and incredibly complacent because you're not doing anything anymore. You don't have to do it.

Michael: 30:46

You're recording your trades, you're working for like ten minutes a day, you're making a whole bunch of money. And that's a big trap that a lot of systematic traders fall into is because that's gonna be great for a couple months, maybe even a year if they're lucky and everything's fantastic. But that debt of being lazy will get paid at some point in the future. And you'll be sitting in that moment saying, man, if I just took half the time I spent on the golf course and built another two or three really good strategies, you could have really avoided a whole lot of pain. So, you know, when it comes to when to reoptimize, you know, I think the answer we came to is, again, as little as possible, but doesn't mean you're not doing You should always be doing stuff.

Michael: 31:32

It's just your time should be sunk in different areas.

Dave: 31:37

Yeah. So, as you're talking there, it reminded me of something I have done in the past, which people might find useful. If you think about your backtest, let's say you've got a ten year backtest and you want to try to simulate what reoptimizing would look like. One of the things you could do is take all the data up until say four years ago, like some period of time, take that set and do your optimization on that. How would it have looked since then?

Dave: 32:13

And you can take that to the next degree. Like if you do feel like you want to be optimizing a lot more frequently, one thing you could do is automate the optimization piece or like at least systematize it and then go through and simulate those optimizations over history of your backtest. And you could collate that as path one for optimizing the strategy is re optimizing every so often and compare that to what you get if you optimize across the entire set. And you could do different slices of time back for that part too and compare. And I mean, I think what you'll find is I think it would be really hard to find gains over the default easy approach of just using the entire set.

Dave: 33:11

But there is a path here to see. So it sort of puts it out in the clear there. Here's how you could do it and how it would have looked like if you were doing this frequently and compared to how it would look like across the easiest approach. I think that's a good thing to try.

Michael: 33:32

You know, I think the theme is it's working in parallel, right? You should always be not going back to the strategy that you had and really fine tuning that, but kind of like you mentioned there is have a whole bunch of parallel work going on because this was actually something I was gonna ask you is how much people who reoptimize too frequently or even are asking us this question all the time. Do you think it's like an itchy trigger finger? Do you think there's just, they're looking to do something and maybe they, cause I've always found that when I present this answer that we both kind of came to on our own that just says, you know, as little as possible, but do something else in the meantime. There's almost like a bit of relief that they have that I think that they are, they're worried that even sometimes when things are going too well, that they know that the the other shoe is going to drop.

Michael: 34:26

So again, people are optimizing their strategy, not because they have to optimize their strategy, but because they think that they have to be doing something and they just don't have a plan to be doing something else. And they just need to have that energy that they want to be using on this very frequent re optimization. And they just we need it needs to be redirected into into better uses of their time?

Dave: 34:49

Yeah, I think there's probably a couple of different camps, at least here, that's kind of people that ask this. One is they're trading something like futures contract where there's not a lot of signals. So yeah, you're gonna feel like you're gonna have to re optimize a lot because there's a big risk of curve fitting where you have a really small set of trades. So yeah, pretty bad. It's gonna feel like things are changing and you don't know why.

Dave: 35:20

So that's one set. The other set is I think people just really feel the need to be right. So anytime they're in a drawdown, they think, okay, well, I didn't do it quite right. Let me go back and maybe I can improve it a bit. The win rate's better.

Dave: 35:37

I just feel better about it. So the answer for the first situation where the trading futures is generate a whole bunch more trades, apply it to The US equities markets, and you're gonna have more trades than you can deal with. And your path to a profitable strategy is so much easier doing that. So that's the answer there. The answer for somebody who feels the need to be right, what I would say is focus your energy on reconciling your backtest to your live trades.

Dave: 36:15

There's usually a lot more to learn there and apply to your strategy and to your execution that can improve things. There's always something to learn when you dive in and do that comparison. Some of the most valuable things I've learned from trading is from doing that.

Michael: 36:37

Well, like that kind of reframing about what is right. And that's something that I try to hone in on just a little bit, and I think is important is that for people being right, I I always find the new trader is comparing being right to making money. If I make money, I was right. If I lose money, I was wrong. And then what you kind of mentioned there, and the more experienced trader, is being right is kind of how well did my real world results in the bottom using all this follow the system.

Michael: 37:10

And it's another superpower, I think, that systematic traders have is because what you're trying to do is you're trying to pull randomness out of out of something that is always going to be in some degree random. Whether or not you make money any given day is more or less random, right? You could have an amazing strategy, but it could just be wrong this one time. But if you look at it and you then rerun your backtest, and you know, you should have lost money and you did lose money, if you can start to rewire your brain to say, I was right because I did the thing that I should have done, as opposed to tying that kind of self worth or correctness feeling into whether or not your P and L is up or down at the end of the day, I think is is perfect. And and, you know, it's something I I see a lot of professional traders do is, you know, they'll they'll walk away, they'll lose 10 that day, and they'll have a big grin on their face.

Michael: 38:03

And it's because they did this is everything they should have done, and they did it absolutely perfectly. And to them, that is that that correct action as opposed to, you know, I did things and then I ended up making money. And again, that rewire it. It's not easy, but it's definitely something you should work on.

Dave: 38:19

Yeah. And as you're talking there, I think another category of trader that would ask this question is somebody that's having a strategy that works well except for some big losses that are scattered throughout. And in that situation, I think the answer is like a lot of times you're not going to be able to avoid the big losses. Some of these strategies, you're not going to be able to avoid the big ones. So the answer there is not figure out ways to try to remove the big ones.

Dave: 38:54

I mean, you should try, but a more valuable thing to do is add more trades to the system. So the big ones are just less important. That's really powerful way to improve a system is to just sort of overwhelm the losers with a lot of smaller winners. That's, that's a good approach that a lot of people don't think about.

Michael: 39:18

So, I guess a, I'm just looking a, forty minute episode short is when do you reoptimize as little as humanly possible? Is that probably the one line way to sum it up?

Dave: 39:33

Think that's the takeaway. But to be able to do that, you need to get better at optimizing in the first place. So think about rules that do have staying power, that have more staying power than what you did before. So I would say the big lesson here, be very discerning about your optimization process as you go through it. Keep track of the decisions you make, which rules you apply, why?

Dave: 40:00

Did you choose this one and not another one? So that when you do end up having to reoptimize at some point, you can very quickly get back into that same context, that same mindset, and you will have learned a lot in that period. And if you go back and see exactly the steps you took, you can make some changes there and apply what you've learned since then to improve your strategy and improve your, the more important thing, the more powerful thing, improve your optimization process. When you end up with a stable strategy that has a lot more staying power, that's really, that's powerful.

Michael: 40:43

Awesome. Well, as always, we appreciate people tuning in. We appreciate the questions because, you know, that's, it seems like a lot of the episodes can just be spurred on by someone asking something that we go, Oh, that's cool. You know, I've been asked that a lot, or some of them completely kind of change the way we think about things. So, yeah, super appreciated.

Michael: 41:02

Another great episode. And as always, I'm Michael Noss.

Dave: 41:06

And I'm Dave May. Thanks for joining us on Line Your Own Pockets.