Okay, everyone.

Welcome to another episode of Line Your Own Pockets, where we're continuing to go down this kind of basic backtesting and systematic trading mini series.

This question actually comes from a user.

So as always, we love we love users' questions.

And essentially, you're talking about using watch lists or stock universes and whether or not that makes sense.

So when you're building a system, should you look at every stock in the market or, you know, just the S and P five hundred or the Nasdaq one or Russell three thousand or any of these individual sides of things.

Now, in the last episode, me and Dave did a lot of time agreeing with each other.

So that's why we did this one now because I think there might be some disagreements when it comes to the difference between day trading and swing trading.

So first of all, why don't you give your answer, Dave, again, knowing that this is from what you do, which is kind of exclusively day trading.

Yeah.

So this question comes from a guy that I've been working with named Roger, like the thesaurus.

Super interesting guy, very technical guy.

He's all about optimization.

That's his background from a different field.

And he's really getting into strategy development.

And he had this very good question, should I be filtering my universe based on inclusion in the S and P 500.

He started off that way and then he moved to the Wilshire 5,000, which I don't know much about.

I don't know much about any of these lists really, which is I think very interesting to talk about because I think that's a really important thing.

You don't know, you're outsourcing the inclusion of these things to somebody else.

So you don't really know why they're in there.

So that in itself to me is a red flag because I talk a lot about what's your path to confidence.

And that's almost like a little black box.

You're using a starting point that you don't really have full control over, you're basically outsourcing the logic, you can't really know why or why the stock isn't in there, then that's already a red flag for me.

How can I have full confidence to trade with a lot of size when I'm using that as the starting point?

And it's not only just a rule, it's the basic starting point for a strategy.

It's a big red flag for me.

So I'll take a little bit of a different side of it.

So first of all, if you are going to do any of this, just a caveat right off the way, you have to have data that includes survivorship bias, which means that if you are testing, say on the NASDAQ one hundred, just to make it simple, right?

It's gonna be the one largest 100 tech names at the time.

If you're testing five years ago, the NASDAQ one hundred is gonna look slightly different than it does today.

If you're testing twenty years ago, it's gonna look wildly than it does today.

So a lot of people don't even account for survivorship bias, which is I'm going to test my system on all the things that were kicked out of that Nasdaq 100.

So that's huge, right?

I use Norgate for that.

They handle it really, really well.

And for when it comes to kind of systematic trading, and when it comes to testing, I think that solves a lot of your problem right off the bat, because you're testing against a universe as it was at each moment that you're you're testing it at.

So the SP 500, just gonna be your largest 500 names in the SP 500 or in the in The US market at any given time.

And making sure that I'm I'm including that bias certainly helps.

But I totally agree.

The part I totally agree with is that this is the whole myth of people think that there is such a thing as passing investing.

Right?

There's not.

It's the S and P 500 is a system, and it's it's kind of like a really good argument for systematic trading where they're saying that they think that the most impactful value of a company is its market cap.

So when your market cap grows to a certain size, you get brought into the S and P 500, and if it shrinks to a certain size, you're you're kicked out.

And even on weighting, it's weighted higher the higher your market cap is.

So they're just using a, almost like a single variable system at that point.

It's like a rotational system that uses a single variable.

So that's kind of I'll just get started with saying, I think we are gonna, I think, have some disagreement basically around time frame.

But yes, if you're going to go down this road at all, it's an absolute must.

You've got to have that survivorship bias data or you're Yeah.

You're fooling yourself.

So I totally agree with that.

That is an absolute requirement to have.

So and on one hand, hey, think that's the requirement that you have that is already going to weed out a bunch of people, like a bunch of your competition.

And just that there's that hurdle there sort of makes that somewhat attractive because you just know that some people aren't going to realize that they're going to just use the S and P in the current way and they're going be inefficient traders by doing so.

So you bring up another good point there.

With the S and P 500, it's weighted as well as just inclusion.

So it's not a simple yes or no that you're in the list or not.

It's that you're in the list with a certain weighting, correct?

Yes.

For the S and P 500, there's the RSP, which is equally weighted.

But, yeah, so to get into the S and P 500, there's a couple criteria, but the main one is, are you one of the 500 largest stocks in The US?

So then I would So you're already going from a starting point that You're already reducing We talk a lot about a continuous variable versus a binary.

And using a symbolist is 100% a binary.

And you're already going to be losing precision in your strategy.

And I think that's it's just going to be an inefficient way to create a strategy.

Now, can you create a strategy that way?

Can you use that as a starting point?

I think you can.

But you're leaving a lot of money on the table by doing so.

So two different worlds.

So let's separate them out because I'm going to be full in agreement for the day trading side of things, right, where you're essentially you should if you're running a purely or, you know, 99% kinda, let's say, technical setup, right, you don't care what the fundamentals of the company because it's a day trade, you're in and out in in a day or an hour.

Who cares anyway?

That should be done on as many instruments as possible.

So you should probably just use the full equity list for that.

However, you're always going to be filtering by some metric.

Like, I'm sure you have in all of your setups some volume requirement.

You don't wanna be looking at stocks that do ero shares for a day.

So you're doing some filtering yourself.

So the question I would kind of ask that person is, what is it that is appealing to you about S and P 500 names?

Because if it's just that you think it's gonna work better on larger, more liquid names, you could kinda do that filtering yourself, right, just by adding a a market cap and a volume filter as well for for the day trading sides.

Now for the swing trading side of things, the thing that I like about running it against universes is that I like the fact that there is I am, in a sense, outsourcing some amount of quality filters.

And if I'm gonna be holding something for a week or a month, I understand that I am I'm outsourcing that that universe filter.

So for it to be in, say, the S and P 500, I know that there's a whole bunch of other nerds that are out there that understand this whatever, you know, the profitability of the company and the market cap and all that.

I'm basically saying you guys do that work because I have the data going back to 1950 of everything that was in the S and P 500 at the time.

I'm not so worried about the fact that I don't fully understand exactly how they do whatever because it's it's there.

At the end of the day, it's again, it's gonna be different for both, but it's gonna be one of those things where the, you know, the solution's gonna be tested.

Like, wanna test it on both universes and see which performs better in the long run.

Yeah.

Well, a symbolist is essentially a filter.

Right.

And by definition, this filter is ero or one.

Now, we've talked a lot about this exact problem and why you shouldn't be using a binary filter ero or one that you're losing a lot of precision by doing that.

But yeah, here we're talking about, yes, we're using this as a ero or one to whether it's on the Symbolist or not.

And you bring up important point, I think, which is, is this the best way to is this the best filter to apply?

And when you go back to your system, I don't care what timeframe you trade on.

Swing trading, investing, you're getting in for a reason.

And that reason is the signal.

And if you're telling me that that signal, the most important factor is whether they're in the S and P 500 or not, I'm not buying that.

I think your system should be better.

Here's an example.

Put it in your backtest.

Just like we talked about before, take a backtest, do no filtering, put a column in there for one or a ero in the S and P 500 or not, or whichever one you want to use.

Put it in the Strategy Cruncher.

And that darn well better come up as important in the strategy crunch.

If it doesn't, then that would be a big red flag for me.

I would think, okay, why in the world would I be using that?

That's another factor that's obviously more predictive.

And then that's to my point of testing.

So there could be limitations, for example, of going back twenty, thirty years and getting the market cap for each individual stock.

Like that could be a a problem.

Right?

So, again, so you're kind of outsourcing market cap requirements to, you know, the Russell 3,000 or something.

That could be something that you're doing.

And, yeah, it should be as simple as let me run this universe on the entirety of stocks and get results a, And then, you know, let me return it on, you

know,

the S and P 500, whether it's not it's in it and get results b and see if that improvement is that significant.

But, again, at the end of the day, like you mentioned, you're filtering names.

So the question is, a, do you have the ability to filter names by market cap if you're if you've determined that market cap is important to your your, your system?

Because I I do have some that are that I use a dollar volume for a kind of a way to say, hey, it's liquid and people care about it.

Yeah.

So there is that kind of premise as well.

And then also, the likelihood and the other reason people may use filter sets like S and P 500 or Nasdaq inclusion is the overnight hold and overnight risk.

So if I'm holding any stock and I can't be I can't look at the market cap of it or I can't I can't test the profitability of the company or or something like that or cash amounts over long periods of time because that data is hard to get by outsourcing and saying, okay.

Well, if it's in the Russell three thousand, you can just do the math and you can see the studies way more likely that a company outside of the Russell three thousand goes to ero overnight than one that's inside the Russell three thousand.

Right?

Not impossible.

Then way less likely again that a stock that is in the S and P 500 goes to ero overnight as opposed to a stock that's outside the S and P 500.

So just to show you, that's why people generally speaking for swing trading and long term investing will use these filters is you're essentially just saying, I can't get the profitability and the, you know, the profitability type data and market cap data myself very easily.

So I'm just gonna grab the Russell three thousand.

But that's still gonna include, I think, the Russell three thousand, like, 7% of the investable universe.

And you you talked about the Wilshire's 5,000.

There's not even 5,000 stocks in the Wilshire five thousand anymore because they're not 5,000 tradable liquid names.

Yeah.

So I think it's really different, a really different conversation if you're talking about the Russell three thousand, which is 97%.

You're basically excluding, like, OTC and and penny stocks out of the Russell three thousand versus someone who's saying, I'm only gonna trade, like, the Nasdaq 100, or I think we can both agree.

The absolute cardinal sin was I'm gonna pick a basket of my own stocks to test again.

I think that is you've really done something bad then because I've seen people do that all the time.

I'm gonna test my setup on, NVIDIA, Tesla, Apple, whatever.

It's like that's taken to the extreme that I think you're really, really kind of hurting the data at that point.

Yeah.

I if I can maybe understand if you're holding for a long time, the closer you just get to being an investor, but there's so much more money to be made on a shorter timeframe and there's so much more of your account that you can turn over by making more trades.

It just doesn't make a lot of sense to me.

For the the world I'm in, the inclusion in the S and P 500 is never gonna be important.

Just never.

Well, and I'll let's just go.

So I have I have three buckets of systems that I run.

Right?

And I try to run all of them.

So I have day trading systems, which I'm I'm building out and and trading actively now.

They include everything.

Right?

And then the swing trading systems, they also include everything, but with more stringent filters on things like dollar volume and interest and all that because I don't wanna buy a stock that goes nowhere for a couple days.

And then I have investment systems, and I have those.

One is a a basket of ETFs, a universe of ETFs, and then the other one is based off the, the Russell three thousand.

So they're never super limited universes, but as the time frame goes out, the filtering comes in.

And I wouldn't use any of those filterings if, for example, for the investment ones, I could go back and look at things like market cap and and, you know, profitability and and metrics like that that I don't think are predictive in nature.

They are just more protective in nature, if that makes sense.

If I'm gonna hold a stock with you know, I'm gonna put $20,000 in a stock and hold it for a month.

I'm gonna be way more comfortable if that company has a couple bucks to its name or, makes a couple bucks as opposed to something that is, you know, just this OTC whatever that could be a complete scam.

But for a day trade, I don't care because the odds that it's gonna go bankrupt intraday are are much, much smaller.

Yeah.

So there's a couple of things I think about here.

One is, you know, when you're creating strategies, we know that you want to minimize the number of rules.

If you set out to curve fit, that was your goal to overfit, what would you do?

You would add a whole bunch of rules.

So you know that to avoid curve fitting, you need to add as few rules as you can.

So right away, we're adding a rule here that we've already said is not super predictive.

And so it's already somewhat arbitrary and you're basing your whole strategy on that.

So why isn't this not like you're setting yourself up to curve fit right from the very beginning?

Because it's just, it's another filter set.

So just like right from the very beginning, I'm sure you put some ero, non ero volume amount on it, right?

It's just the same way of saying I'm putting some non ero market cap amount on it.

That's the way I look at these filters is that, because if you look at it, there's the IWM, which is Russell 2,000, that's from two to 10,000,000,000, and the mid caps is, what, 10,000,000,000 to 50, and then the large caps 50,000,000,000 up.

So all you're doing is instead of bucketing by volume, which everybody does all the time, you're also just bucketing by a certain amount of market cap and profitability.

So, again, it answer is always, like, test it.

But if you're finding that your test runs way better on the S and P 500, then you're probably saying, well, for my whole time anyway, some amount of of profitability or stability of the company, you know, cash on hand, whatever it is, that leads to that market cap is important for that.

And quite often when it comes to things like trend following and and longer term stuff, you'll notice that that's the case.

It's not your your junky penny stock that jumped up 200% this day and is gonna fade its way back to ero.

That is something you wanna hold for a month.

So that's why in different universes, it makes sense to, right, potentially look at different names.

If, again, you could get the data to say, okay, which one of these?

Is it market cap?

Is it profitability?

Is it, you know, cash or whatever?

You could do that.

That would obviously be better.

You'd you'd split those out.

Just, again, for long periods of time, for long periods of history, I think fundamental volume or fundamental data is super hard to get compared to even technical data.

Yeah.

I think that's true.

So you do bring up another point, which is even said if it's not in the Russell 3,000, there's a much higher chance of it going to ero.

Well, you just stumbled upon a very good trading strategy there by excluding good companies.

That is a case where I could actually get behind something even for day trading is excluding a list of stocks that you know But I would say typically that's on the short side, but

I would argue even on the long side.

I know a lot of traders, for example, will do the same thing with, and not necessarily again, if you're doing it short term, not necessarily because of an inclusion to a list, but they're just they're all it's every low flow trader you ever talk to.

They're not gonna be trading Nvidia or Apple or whatever.

They specifically filter their universe by, let's get the big stuff out of there because we know it's more efficient.

It will, generally speaking, be less volatile.

Now sometimes they're not even doing that by putting in a multi market cap filter.

They're just saying, you know, rank the stocks by how volatile they were that day and naturally kicking those names out.

But, yeah, I've met a lot of traders say, I just only trade low cap stocks.

So by definition, they're putting a filter on a certain amount of market cap, and they're just doing the opposite of what we talked about with the with the day trading or the swing trading side of things or investing.

Well, I think that a lot of low float traders aren't actually putting the filter in.

The definition of their strategies just automatically gets rid of anything.

Like no really big name is ever going to move like 100% in the day.

So it's just never going to show up in a lot of these strategies that people trade.

So yeah, they're not specifically doing it, but they wouldn't turn their nose up at it if Nvidia went up 100% in one day.

That's probably too high.

That's probably a good short.

Yeah.

And that's why I'm saying they but, like, let's say they had three stocks that went up a 100% that day, and they could only pick one.

It may be not a first filter.

It could be a second or a third filter down the list.

But and that's the same thing I would say.

So when I kind of talked about the the dollar volume as a way for me to find kind of established names, that is a way, again, to kind of try to accomplish the same goal that this this guy might be trying to do, but just do it a little bit differently.

And by dollar volume again for the audience, it's just the share price times the volume, and you just take an average of that.

And the idea is that if you're a very high vol dollar volume stock, just means that there's a lot of money changing hands in your stock every day.

Generally speaking, you'll be just by definition or by nature of it, you'll be a higher cap name.

Right?

Apple is always Apple and Videol, these are gonna be the highest dollar volume names every day.

So, yeah, it's one of those for a proxy, again, okay.

Yeah.

If you if you had an alternative, I would say it's okay.

The yeah.

So it's one of those things that know, there's probably a better way to do it, but for a proxy, don't think it's bad at all.

So have you thought about you know, we talk as we mentioned before, we've got a binary ero or one.

It's either in the list or not.

So does Norgate or any other service provide the weighting over history?

And is there a way to backtest on the weights?

I I think you can get that.

I don't know why that would matter that much unless you were looking for because there's subsectors.

So say you only wanted the 100 largest names.

There's the S and P 100.

Right?

There's the S and P 50.

There's the S and P two fifty.

So you could definitely filter, by, you know, quartiles of of the largest names out there throughout histories.

And if that really mattered, then, yeah, I guess.

But like we talked about, every time you do that, you're going be limiting trades.

Right?

Yeah.

But I think about the edge of the list, think about the bottom edge of the list and how arbitrary it is.

And if you've said that, Hey, the list is set up, here's the definition.

And on the very edge of that list, and you even said like the Wilshire 3,000 or whatever doesn't have, or 5,000 doesn't have 5,000 names in it anymore.

That's completely arbitrary.

Like just think of the bottom part of that list.

It's completely arbitrary whether it's on or off the list.

And so it has to have some difference, like the top end of the list versus the bottom.

I mean, that's the whole reason you're using the list at all.

So it seems like to ignore the waiting would be it just seems crazy.

Right?

Well, think of it again.

Let's use the volume example.

Right?

Again, in your strategies, I'm sure you have an arbitrary volume cut off.

You you had to pick a number at some point.

Right?

Now we're talking like Loki's wager stuff.

Right?

At some point, you have to pick a a number to say, you know, you know, if it did less than 50,000 shares, it's off the list.

And if it did more than 50,000 shares on the list.

So at some point, there was a cut off that that should have happened.

So it's one of those things where you have to you have to do that cut at some point, and they're just doing that cut based off a a metric like market cap as opposed to something like volume.

Right?

Yeah.

But I think that I I look at the I I I would call that like a universe filter where I'm I'm It doesn't actually come into play that much, but it is like a floor below which there's a lot of risk there.

I know there's a lot of risk.

So I'm saying use a floor for that so that your risk isn't too high below that.

I think that's fundamentally different.

I think it's fundamentally different, though, from the S and P 500.

That's a completely different type of filter because you're saying, hey, this filter is absolutely gonna come into play every time.

In my mind, if you're using a filter like that and it's doing a lot of the work for the strategy, it better be dang predictive.

Yeah, predictive or again, like we talked about protective, right?

You know, making sure that you're not going to ero overnight.

But I would say it's it's kind of essentially the same thing at the end of the day.

The only thing that I kind of see your point on is that the the S and P 500 by definition only lets in 500 stocks.

Right.

So it's not that it's it's cutting off at a hard market cap number.

It's cutting off at a number of of stocks.

So that's why I say it would be better if you determine that market cap is important to your trading strategy that if you could get that data, again, with survivorship data going back far enough that's interesting to you, because then that may fluctuate anyway from, you know, 500 to a thousand, right, or whatever.

It could fluctuate around a number.

But, again, at the same at the end of the day, at at some point, we are always kinda putting our foot down and cutting off and saying above this number, good, below this number, bad, whether whether we want to or not.

Yeah.

It's just something that for a day trader, don't think should matter at all, which is market cap.

For somebody who's holding for a long period of time, it should.

And then the question again is always, what is a proxy for that?

Right?

So I would kinda go back to the user and say, what?

I bet you there was a comfort feeling of potentially trading only the S and P 500 names that we got from, okay, these are big names.

They're real companies.

They're not gonna do anything, you know, insane.

And just saying, okay.

Could you replace that?

Could you just add a whole bunch of volume and then, like, a low ATR and get roughly the same, you know, 90% of the same comfort while still allowing a little bit of a more dynamic list?

Yeah.

I'm sure it fundamentally comes down to day trading versus swing trading because just when you say you want to make sure that a stock doesn't do anything insane, that's the whole point of creating a strategy.

You wanted to do something insane and then figure out how to trade against that or find the stocks that are going to do something insane.

That's where you really make money.

To me, I could never use the S and P 500 for day trading just because you're limiting yourself arbitrarily and there's a whole bunch of like your strategies are necessarily restricted and you're closing yourself off to a whole lot of money by doing so.

Yeah.

And again, I think it's we've kind of nailed in the big thing is that there are certain strategies that you do want your stock to go crazy.

And there are certain strategies or certain timeframes, I guess, defined better that you won't.

Right?

You know, if if I'm holding a stock for a period of time, you know, I'm placing a trade once a month, and I've got a bunch of other stuff to do, and I'm just placing a trade once a month, then I better be you know, you talk about all the time, path to confidence.

I better be confident that I can hold that particular stock for a month and not have to check on it every day.

Right?

You know, a lot of the traders I deal with, I know you're kind of exclusively dealing with pure professionals that are trying to do the trading for a living, and I deal with some of them, but then I deal with a lot of people who are dentists, doctors, plumbers, lawyers, whatever.

And if they have a portfolio of stocks that they're gonna trade and hold for a week, they wanna make sure that it's not, you know, a company that like, all of these companies now just it's gonna date the podcast a bit.

You're seeing these companies triple in value because they're taking out giant loans to hold Solana.

And, you know, there's one company I think that's up like 500% in the last two weeks, they're doing it.

If I may, you know, there's one gentleman I know who's a firefighter, and he's like, I can't be on the job putting out fires worrying about this this crazy company, so he has to trade longer and slower.

So for someone like that, he might just have that path of confidence to be better at, you know, holding something that he's got some confidence in as a real company and is not, you know, inflated value because they're y'all owning fartcoin.

And the with to do that systematically, sometimes the best way to do it is just to say, right, give me the the Russell three thousand or the Russell one thousand or or some giant basket of stocks that are at least legitimate companies.

Now for day trading, again, who who cares?

Okay.

So you mentioned earlier in the podcast how it's really bad, you believe, to come up with your own list and just trade against that.

Let's talk about that.

Why why do you think that's so bad?

Well, that's just all kinds of biases.

We could we could list them.

Right?

So there's survivorship bias and regency bias and right?

There's all kinds of biases in that.

And the reason I think for a lot of that, unless it's something that you are absolutely sure has, a, existed for a long period of time and, b, will exist as something you're going to trade for a long period of time, I think it's a bad idea.

So, you know, an example would be Apple.

There was a time I know a couple traders, and that's all they did.

They just traded Apple.

That was their whole Yeah.

Thing.

But then Apple became this giant mature company that stopped moving, and, you know, their systems would have degraded away to ero or or kinda negative at that point.

So I wanna differentiate that a little bit with people who just trade, like, the S and P 500 as a whole because that is always going to kinda be around and has already been around.

If there's every day that we wake up and the S and P five hundred's gone, we it's we should have spent our time learning to grow potatoes as opposed to trading the market.

But yeah.

It so you're you're kind of you're bringing in all these biases and you're building a list of of stocks that were volatile in the past and were kind of the in crowd at the time, and you just have no way of knowing systematically if they're gonna continue to be that way or if it's the day that you include them in your your little system, they're gonna stop.

Okay.

So here's another question.

Do you and I totally agree with that.

I know some traders that will even day traders will look at their results and say, Nvidia just doesn't work well in their strategy.

They've seen it show up.

The shit list.

A bunch of times, just doesn't work.

There's an inordinate amount of losing trades for one name.

And so they take it out of the system.

What do you think about that?

Yeah.

Well, it's a shit list.

Right?

I've I think everyone at some point has done it.

Now from a a systematic point of view, it doesn't make any sense.

Right?

It's just not, not mathematically.

The, I believe, and I think you'll agree with me on the right answer is try to figure out why.

Right?

Does that stock have some characteristic again, market cap, we're always hitting on.

So is it is the market cap too big?

Is it, you know, is it maybe a specific industry of stock?

I know people, for example, who hate trading like commodity stocks, because they find they don't have the same move.

So they're looking for stocks that move faster and that, you know, like, is there a different way to filter it out?

You know, path to confidence, being a human being, you know, mate with one stock.

If it makes you feel that much better, maybe.

But it's you have to know you're just you you're fooling yourself at that point just because it's you've touched the stove too many times, and and the next time you touch the stove, it will probably be cool, but you're just kinda mad at that point.

So Mathematically makes no sense.

I tend to agree with that.

Totally agree.

So there's one scenario that we haven't talked about where I do use symbolists.

And I think this is going to be pretty common.

But if not, you should think about this.

I use it to exclude ETFs.

So, a lot of the strategies I run, I'm specifically excluding ETFs from that list.

So, in that case-

It's gonna be a long episode because we're gonna be fighting back and

forth on

all this stuff.

I see you don't like that.

Well, like we talked about, the guys who do the low float trading, but they don't necessarily exclude low float.

They're not putting a float and they're not putting a market cap in a scan.

It's just they're doing it by subjects of it.

Who cares if there's there's two things that I think that will hit on with this.

But the first one is who cares if oil, the oil ETF is gone crazy because something absurd has happened and it hits your scan.

If it has predictive value, if your idea has predictive value, maybe a commodity has just gone nuts that day.

Uranium, I remember after Fukushima.

Like, you can look at that candle.

It was crazy.

There could have been some excellent day trading opportunities or even short term swing trading opportunities in the uranium play.

So by removing ETFs, you're potentially removing, you know, you'll probably get some companies that are involved in it, but you could be removing something that on a very rare circumstance, I agree for most of time, you could be excluding something that's that's awesome.

And shouldn't you just like you said, you shouldn't filter your universe by market cap.

Shouldn't you also not filter your universe by whether it's an ETF or not?

Yeah.

I that's I don't do that for every strategy, but there's some moves that there's some strategies I have where it does come up.

These names do come up in the strategy.

And it's a little a juiced ETF for the Juiced Nvidia or whatever.

That was actually my second point.

Why not?

If it's juiced and it comes up on the strategy in my criteria for a certain strategy, it might be simply because it's juiced.

So I don't want And it's gonna be juiced by That's going to be sort of more normal behavior for that name.

I with some could just see with some of my strategies it's like in baseball when you're hitting foul balls.

You're not hitting home runs with it.

You're hitting foul balls.

You're making contact, but you're not hitting home runs.

And it's kind of hard to explain, but I feel like some strategies I see sort of make contact, but they're foul balls.

Think about if you didn't know the rules to baseball and you were up there and you started hitting and you hit foul balls every time, you would think like, wow, I'm doing great.

But you're not even like, you're pretty much striking out.

You're not hitting the ball in play.

So I like in these cases with certain strategies that I trade, it's good to exclude ETFs for that reason.

I wouldn't say it's what makes it profitable or makes it successful or not, but I do feel like I exclude some foul balls by doing it with this particular class of strategy.

So two things to think about.

Like one is, so if you're doing a good job normalizing your data, shouldn't the fact that it's a juice ETF be just taken accounted for?

And it will only show up if it was an exceptionally large move on that juice ETF that day.

And the other one is, okay, what if this the stock is doing something?

So for example, I I just brought up a HIMSS.

So they they mail enhancement, let's say, was down, like, 30% the other day.

K.

It showed up on on one of my systems.

Himz, which is the double leverage one, which I wouldn't have seen systematically, had a 70% move.

Right?

So, you know, if I had excluded ETFs from that, maybe I would have traded the underlying that was that still had a a massive move that day.

But by excluding the ETF, I missed out on a double percentage move, which either means more profit or less capital allocated to get the same profit.

Right?

Yeah.

So those two things are are kinda what came to mind when you were talking there is that, yeah, normalize the data correctly should be the same.

And then if you're getting more juice for for less of the squeeze, then who cares if it's an ETF or a or a stock?

Yeah.

I'll have to think about that.

Because if we Yeah.

I'll have to think about that.

But my my I I do know that it's not a huge mistake.

Like No.

No.

It's just to exclude them.

Well, and now

If it is at all.

Now we're gonna go we'll circle it all way back around.

Right?

Is it's just like it's probably not a huge mistake to look at the Russell 3,000, which is 97% of the investable universe, or the Wilshire 5,000, which is 97% of the investable universe, as opposed to the entire stock market, you're probably if you run both of those simultaneously, it's you're you're not the difference isn't gonna be one strategy kills it, the other strategy is awful.

The the question you're always asking yourself by excluding anything is, am I excluding something that could eke out a little bit more profit?

Right?

So I'd say excluding the ETFs is the same as including something else.

Right?

Yeah.

I'll have to think about that.

I'm gonna go back and look at ETFs versus not ETFs on a couple of strategies and see.

Because it has been a bit since I've looked at it.

So

Well, and this is Yeah.

We should talk about these single stock leverage ETFs are a newer phenomenon and something that I'm exploring heavily because it is a newer phenomenon.

So we we've talked about this privately that for, like, bots, I would love every time Tesla, if it comes through the system, is replaced by a double leveraged ETF with, like, half the size.

And right?

Because I think these are these are amazing tools that are just there's not gonna be a huge amount of data on because this whole, like, this mega degenerative stuff has happened in the last couple years.

So it might be one of those without as you're doing this data, you're spending a little bit more time focusing.

Leveraged ETFs are not new, but leveraged ETFs in individual names are way more.

Yeah.

That would be interesting.

I'm not sure do you know if there's a way to categorize whether something is a leveraged single name ETF versus just a regular ETF?

That'd be interesting to see if you could categorize those and and get those quickly without going through and examining each one.

Is there a list of those that that does that categorization?

That'd be interesting.

Yeah.

There would have to be something out there, but I think there there is.

And the problem is they're coming up all the time.

So these guys are chasing chasing the dragon with this.

Like, there's now one for quantum computers and, like, quantum computers aren't a thing.

Mhmm.

But the companies are on fire.

So then every time you get volatility in the company, they launch a leveraged ETF.

Because one of the things that I was exploring that I thought would be interesting is by the time the leveraged ETF is launched, is that the end of that particular run-in that name?

The yeah.

The idea is, like, could you end up going through and saying, hey, this is one of those cycles because, like, for example, they delisted the coal ETF, and that was the bottom of the coal market for, like, the last five years.

And it's

the Mhmm.

It's the classic magazine cover indicator of by the time, you know, the news is talking about it, right, that it's it's the end of the move.

So it's a very interesting universe.

And again, it might be one of those you built the strategy before this existed.

Now you you you go back through and and you find that over the last few years, you would have done better.

But, I would encourage everyone just to take a look at those as well.

But, you know, it's funny that it's whether you're excluding or whether you're including stuff, it it we're kinda coming back to the same idea, which is likely suboptimal.

Not necessarily bad, but, you know, if if you could figure out a a different way, it might be a little bit better.

But if it if it works in backtesting, it's probably not the worst thing in the world.

Yeah.

Yeah.

I'm curious now.

Yeah.

Super interesting.

This is, I knew this is going to be a good discussion, but I didn't think it would be this good.

Yeah, this is great.

Well, again, we go to a different episode of the podcast we did for a while about building your own trading group.

I think this is the the you just saw a little bit of the value of that of someone who just doesn't you know, there's mentor mentee relationships where the mentee just usually listens to what the mentor says and maybe ask some questions.

But then there's finding people that are your peers and they might give you a little bit of pushback and or say, you know, have you really thought about this?

Have you done that?

And, doing it off social media where you can do it in an amicable way where you're not like, right, instantly hating the other person because they disagreed with one small thing you said.

These these kind of trading groups, I think, are are important because, yeah, we you know, me and Dave talk a lot offline, as well.

And, yeah, quite often, you can go, you know what?

I didn't think of that.

Let me go back and give that try or change some things here or there.

Yeah.

Yeah, kudos to that old podcast we did, which I think I still think is one of the most valuable ones we did.

I think it's called Create Creature and Trading Group.

Mhmm.

We'll we'll we'll link to it in the show notes.

But yeah.

Yeah.

Definitely a a good one to to go back and listen to.

We stomped Dave, guys.

Come on.

That's all we can give around.

We got him.

We got him thinking.

He's he's

just He's got me thinking.

I I love thinking.

You know?

So there was I I just heard from a a listener.

I just got an email this morning from a listener.

New listener, he said, man, I he said, wow, I I can't believe I didn't know about this podcast.

I've been listening to this on my commute.

I don't always agree with everything, but a lot of what you said is confirmed what I'm already doing.

I was like, oh, let me know when you don't agree with what we say.

The most interesting discussions come out of that.

Time that happens, send me a note and I want to hear about it.

Talk about it.

Also, because neither of us pretend to be these mega gurus that have it all figured out and we're not, that kind of thinking, I think, kills systematic trading.

As soon as you're like, okay, I got it.

This is my setup.

This is the thing I do.

You're you're you've just stopped any further progress.

And when that thing that you're doing ends up stop working or, you know, performs a little less optimally, or even if it keeps going forever, you're still leaving a bunch on the table, not doing different things.

So Yeah.

You know, I yeah.

I'm the same way.

I love questions and and pushback and that kind of thing because I'm not gonna be one that I'm gonna fight and say, oh, you're wrong or you're dumb or whatever, unless what you're saying is dumb.

But it's it always anything that's brought thoughtfully should be considered thoughtfully.

And and like we talk about all the time, idea generation is kind of the hardest part of this whole game.

So, you know, the that little conversation we just had, now Dave's got some ideas, and we'll go test them.

Turns out could be right, and excluding ETFs is the way to go, and that's fine.

That could open up a whole other door of questions that he needs to ask, or he could find that with recent changes to the market and with these new single stock ETFs that there's a little bit of unexplored edge there, and that answers a whole bunch questions as well, right?

So, yeah.

Yes.

The first thing I'm going to do is go look at the ETFs.

And here's a good thing to point out.

If you find something that is improved excluding something like this, like we're talking about, let's say that that does work.

There's usually some edge on the opposite side for that excluded list.

So very often, it is a kernel of when you're able to exclude something and improve what you're doing, there's a kernel of a strategy there on the opposite side with what you've excluded.

Is exactly how, like these kind of tough questions that you're asking me and that you ask yourself, this is how you get better.

And this is how you turn one system into two systems.

And you continue to do that and just get more ideas.

The more strategies you get like that, the more ideas you will generate for yourself.

It self perpetuates.

Yeah.

And, yeah, I think the perfect example of that was when we were talking about the stocks that are not in the Russell 3,000, those those junky stocks that can't even make that that right?

They're probably like that for a reason.

Right?

So, that might lead a whole path down to what if I only look to short stocks that are not in any known index because they're such crap.

They can't even make it into the into the worst possible ones, then, right, that ends up opening this whole other other door of things that you could do, and then you can either take that and apply it to existing.

Maybe you already have a short strategy, and it'd be very easy to say exclude everything from the Russell three thousand, test again, see if it's better, or building a new strategy.

So, yeah, it's always conversations and questions lead to more questions lead to more conversations lead to more question.

And that's why, again, I think we love this style of trading because it's way more of an intellectual pursuit than let me guess what's gonna happen next on this one stock and this one trade and either be right or wrong mostly because of luck.

And then, right Yeah.

Go and and repeat the same process.

It's not nearly as cerebral, I think, as, you know, talking about the whole theory and edges of the market and things like that.

Yeah.

I you said the word luck, I'll use the word gambling.

When it comes down to it, that's what a lot of people do that aren't systematically trading whether they admit it or not.

That's a big part of what draws me to systematic trading is that I know it's not gambling.

There's some luck involved, but it's not gambling.

I can tell you that.

That's probably a whole episode we could talk about.

All right.

We'll write

it down.

No, that was a great one.

Again, thank you all for the question.

Again, summarize, probably not ideal to filter anything, but, if you don't have any other way to do it, right, if you can't find, a metric in order to do it, then, you know, give it a shot.

Right?

And at the end of the day, a lot of these answers are tested and tested both ways.

Yeah.

That will either answer the question or lead to a couple other questions about why.

For me personally, if I tested something and it just completely fell apart on the entire universe of stocks, but did really well on the S and P 500, that's not an answer.

That's another question of, right, why and then, you know, where can I stretch the boundaries?

What can I play with?

If I can't find the data that makes up that difference, then maybe I'll I'll use that constituency just to as a as a data proxy for lack of a better term.

Yeah.

Love this.

It's great.

Great chat.

And until next time, I'm Michael Noss.

And I'm Dave Mabe.

We'll talk to you next week on Line Your Own Pockets.