In this segment of our Algorithmic Trading System videos, we walk you through the details of the The Swing Trader Package. This package trades the — Treasury Note (TY) and Momentum  (ES) Trading Algorithms.

Trading futures involves substantial risk of loss and is not appropriate for all investors. Past performance is not necessarily indicative of futures results.

Results discussed in this video are based on back-tested & walk-forward models that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under-or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown.

Swing Trading System

[Video Trasnscript] In this video I’ll be covering the Swing Trader trading system. I’ll be going over this package covering the back-tested and also the walk-forward results since live trading began on the two algorithms that are contained within this package.

Before I do that though, I’d like to go into the risk disclaimer. Algorithmictrading.net is what’s called the third-party trading system developer. We are not registered commodity trading advisors so we develop trading systems that are 100% automated and algorithmic for use in the futures market. They can be traded on the Trade Station platform, or auto-executed with best efforts by one of the NFA-registered brokers. Keep in mind that trading futures does involve substantial risk of loss, past performance is not indicative of future performance. Also note that in this video we’ll be talking about back-tested data, walk-forward. When we say walk-forward, we also mean to imply in some cases, live trading as well. But it’s safe to just assume that all the trading that we’re talking about here is from hypothetical accounts, in other words the 15K account that we’re going to show is assuming that $15,000 was traded on these algorithms back when they were last optimized, then walked forward to show how those algorithms have done. The algorithms have traded in live accounts throughout that period, but because some of the fields that we show are from the hypothetical Trade Station account, the simulated account, so we have to have that disclaimer. But we will be, I’ll try to talk more about that as it comes up, but feel free to pause this slide and read this more carefully, but just keep in mind that trading futures with automated systems is, it’s really not for everyone, so just remember that there’s risk involved, no trading system is perfect, there’s no such thing as the Holy Grail. But we do think that we have a great system here that we’d like to show and highlight some of the good and bad.

Swing Trading Algorithm

So, with that, I’d like to talk about the Swing Trading algorithm. First, we’ll talk about the Momentum ES Algorithm.

ES Swing Trading Algo

The Momentum algorithm is one of two algorithms that are traded in the Swing Trader package. To illustrate that, I’d like to go to my Trade Station account and just show you what I mean.This algorithm on the left is the S&P 500 and many futures. You can see that the Swing Trade algorithm is applied to this chart, or the Momentum ES algorithm is. If I kind of zoom out, you’ll kind of see what I mean. Recently we’ve been in this Trump rally, although the last few days, the market has been selling off some. But just to give you an idea of what we’re looking at here, this is the S&P 500, starting in November, all the way through March of 2017. What happens is, when you code a Trade Station algorithm it allows you to apply it to the chart, and that’s when all these trades populate. So if we zoom in, you’ll see a few of the more recent trades we’ve had. So today is actually Tuesday the 22nd. Yesterday, coming into the weekend, we were long the S&P, so on Friday we were long the S&P, and on Monday morning we had a gap up, and this algorithm got out with a gain, and then it sold off. So the little blue dotted line that you see here is a winning trade, the red dotted line is a losing trade, like we had here. What you’ll see is that this algorithm has actually done really well in the last few months. If I kind of zoom out, or go back in time, you’ll see the start of the Trump rally, and it’s really done well. If you go forward in time, you’ll see that there’s been a decent number of winning trades. I don’t want to go into too much of the details of this algorithm, though, because we do cover it in another video. But really it’s just to show you that the Swing Trader trades two algorithms, the Momentum ES, which is this one here, and also the Ten Year Note, which is this one here. The Ten Year Note algorithm is the one that does well when the S&P goes lower, typically. There’s a negative correlation of about negative 35% between the Ten Year and the S&P. Which means, in the long run, on a weekly basis, closing week to closing week, going back since I believe ’03, if the S&P is lower, as it was on Monday, then the Ten Year Note will be higher, as you kind of see in this rally here. Now they don’t always trade inverse to one another, but overall, if you average it all out, that is how they trade. So by trading both these algorithms, we try to take advantage of both up-moving and down-moving markets.

The sideways-moving markets are always the trickiest to trade, but these two algorithms handle it really well. In particular, the Ten Year Note will usually do really good in an S&P sideways-moving market, and the Momentum algorithm just won’t trade a lot through those periods. If we zoom out, you’ll see what I mean. As the market was going lower back here in January 2016, this Momentum algorithm was just on the sideline, because you don’t see any trades throughout here. But let me go back to the FOIL and we’ll keep talking about this Momentum algorithm. So this algorithm began trading live back in October 2015. At that time we called in the BullFire, but it’s the same algorithm as the Momentum ES algorithm, we just changed the name.

This is an equity curve of the walk-forward results, meaning the results that we’ve seen, in the hypothetical account, since we last optimized. That’s on a blind walk-forward out-of-sample data set. So what that means is that, not to get bogged down in too many details, but most of you are probably familiar with back-tested results, and how those are always going to be the most optimistic results that any developer can show. The walk-forward is always going to be a little bit more pessimistic, and we rate them based on efficiencies. So how efficient is the walk-forward compared to the back-tested, and in this video I want to really highlight the walk-forward results of this package as opposed to the back-testing.

If somebody started back in October of 2015 with a 15K account, trading only this Momentum algorithm, this one algorithm that’s contained within the Swing Trader, then they would be up about $9,100, the percent profitable is about 76%, and this is the equity curve. So it gives you an idea some of the ups and downs that it sees, but overall the trend has been higher, so that’s good, on this equity curve. We’ve actually recently been hitting new highs as well on it. So this algorithm has done really well since it went live. And again, this is one algorithm within the package. I can also go back to this chart, and show you the Trade Station report. Now this is going to include the back-tested along with the walk-forward. So what it’s doing is looking at the entire back-tested data which is starting back here, going all the way through, around in here in October. Then this right here is all the walk-forward.

One thing to note is that we do have over 10% of trades that are out-of-sample now. Once we hit that mark is when it’s a lot easier to make judgment calls on how good an algorithm is. We can look at the trade list as well, though, and you’ll see the old name that we had is still in here. This trade list though, is where we get all the trades that I’ll show you as well, as we keep looking at this system. So again, that’s the Momentum algorithm.

Treasury Note (TY) Algo

I’d also like to talk about the second algorithm that’s in this package though, which is the Treasury Note algorithm. Okay, so the Treasury Note algorithm is actually, I believe it’s one of the longest-running algorithms we have, that we still sell today. It was last optimized back in December 2014, and it was an add-on to the NQ Legacy package at that time. At that time we referred to it as the P2 Push-Pull, but we’ve also called it the TY Algorithm, you’ll see the Swing Trade TY. But right now what we call it is the Treasury TY Algo, but it’s the same thing. And what this is showing is also the walk-forward. So basically from 12/15/14 through 3/22/17. This just shows how well this algorithm has done. Again, this is the second algorithm that’s in the Swing Trader package. A couple things that I want to point out though. So it’s had 27 trades since we last optimized it. It’s up about $7,000. We also have the per-unit size set of 15K on this. I’d like to note, though, that this algorithm, overnight margin requirement to trade the TY is really only about $2500, so this is a very, I don’t want to use the word conservative, but if somebody had 15K, in theory they could’ve traded six contracts, this is showing just one contract. So of course, if somebody did that then the max draw-down would’ve been too much and they would’ve had to add to their account. But that’s why we show the 15K. So even though it’s 15K here, really only about $2500 is used per trade, with the overnight margin, but its percent profitable has been 70%. This algorithm has done really well. Here’s the equity curve going back to December. This is again, walk-forward. Still hypothetical account, but it is a blind, out-of-sample walk-forward data. Now, I think another thing I’d like to do is just show you the chart and talk about the Ten Year. So again, the Ten Year is basically the Ten Year Note, the futures for the Ten Year Note. It typically does trade inverse to the S&P. There are periods when they trade closer together, meaning that the S&P goes higher, the Ten Year also goes higher. There’s periods where they could both go lower as well. But normally, they trade inverse to one another. The last couple days is a great example of that. Like I pointed out, I think, when we were talking about the Momentum algorithm, on Monday we had a big drop on the S&P, I think it was the biggest drop since October, I believe, and it’s mainly due to the political concerns over I think the healthcare vote that’s due, but again, I’m a technical trader, not a fundamental trader, so it doesn’t really matter to us. But I just want to point out that you had a big drop on the S&P yesterday, and let’s just look at what the Ten Year Note did in that period. So this is what the Ten Year Note did in that same period. So when the S&P gapped higher, like it did on Monday, the Ten Year actually gapped lower. Then as the S&P sold off, the Ten Year went higher. I want to highlight that, because I think it’s pretty crucial thing when we’re discussing our trading methodology with algorithmictrading.net, which is to be market-direction agnostic, meaning we don’t care if the market goes up or down.

We definitely prefer it to go up, because we love this country, and it just is more beneficial to more people, but if the market does go lower, then we want to make sure that we’re equally balanced, or as close to it as we can be. So that’s where this Ten Year Note comes into play. So if somebody is trading the Swing Trader, the entire package that we’re talking about. Then they would be trading the Momentum ES plus the TY. The idea is that as the market goes higher, the Momentum ES does well, and as the market goes lower, then the Ten Year Note algorithm does well.

The key in designing a system using this methodology is that you want to do your best to prevent losses in the contrary market moves. Meaning, if the market, the S&P that is, is going higher, we want to avoid losses on the Ten Year Note, otherwise you might make gains on the ES, but then you just take losses on the Ten Year. So the key is, really, to try to prevent the losses in the contrary market conditions. That’s where, and that’s really important I think when we’re talking about the Ten Year, because since December 2014, the S&P has gone higher. It’s been in a pretty strong bull market, especially the last four months. But the Ten Year is actually up in that period. So the fact that the S&P has gone higher, and we’re up on the Ten Year in that period is really good. We would be happy if this was just break even. Because we would have gains on the S&P side, on the Momentum algorithm, this one here, because market’s going higher, and this is a long ES, or long S&P algorithm. The odds are that we’re going to do well in a bull market for sure on the ES algorithm. If we can also do well on the TY, then it sets us up really well, because in theory, if the S&P goes lower and we go into a bear market, then the Ten Year should do even better. That’s our expectation.

Complete Swing Trading System

So I think with that, what I’d like to do now is talk about the combined results of the two algorithms combined, which is the Swing Trader, and just talk about how it’s done. Okay, so now what we’re looking at, is when we combine the two algorithms together, this is the complete Swing Trader system. What you might note is that we start the walk-forward on October 2015. The reason why we start in October is because the Treasury started in December 2014, but the ES started in October 2015, so we have to take the largest period to show a consistent walk-forward. If we went back to December then it would be a hybrid of back-tested plus walk-forward. So what we’ve done is we just started with October 2015 through the 22 March 2017. Basically just combined those two algorithms. This shows you the blind walk-forward results of the Swing Trader package. So a 15K account would be up about almost $12,000. We’ve had 148 trades on the ES and the TY combined. Percent profitable is 74%, and the max draw-down is about 6700. If you average it out, average gain per month is about 4 1/2%, and then the average per year is about 52. So overall, this algorithm has done really well on the walk-forward, and remember, this is not back-tested, this is walk-forward. Still on a hypothetical account, so make sure you consider that disclaimer, and still, past performance is not indicative of future performance, but it does give us an idea of how well this algorithm has done. And what a lot of people ask is, when do you know if an algorithm is still good, or bad, or how do you really analyze it? So now that we have about 10% of out-of-sample data we can look at, then we can start making judgment calls on how well this algorithm has done compared to the back-testing. So if I go to this chart, what we’re looking at now is the back-tested, for the Swing Trader. So it goes back from October 2003 until 2015. A few things that we will do to analyze how well it’s done is we’ll look at things like the slope of the equity curve, we’ll compare the slope of this equity curve, the walk-forward Swing Trader, with the back-tested equity curve. We’ll also look at things like percent profitable. So in the back-testing it was 77% profitable, and in the walk-forward we’re at 74%. That’s all within very reasonable expectations.

In other words, in my opinion, it’s almost impossible to have an algorithm trade as well as it did in the back-testing in walk-forward Once you have at least 10% out-of-sample. So there’s a way that we rate them using an efficiency metric, and it’s basically saying how close to the back-test is the walk-forward. The fact that we are so close on the percent profitable is very good. The number of trades is 148, in the back-testing it’s 12,444, so we’re over 10%. When we hit the 124 trade mark, that was the 10% mark, so we’re above that. The average gain per month, walk-forward, is 4.41%, back-tested it’s 6.3%. So it is down little bit on the average per month, but still really good numbers. So overall, we are definitely very pleased with the results of the Swing Trader algorithm. It, for the most part, has lined up very well with the back-testing. Even though, again, the past performance does not dictate that it will continue to do that well moving forward, it does at least give us kind of a reference point.

Advantages of Swing Trading

So with that, I’ll just quickly talk about some of the highlights, now that we’ve talked about the back-tested and also the walk-forward. All right, so as I’ve mentioned, we have more than 10% out of sample, and enough trade data to perform analysis and to really kind of dig into the details and compare how we’ve done. The average gain per trade, percent profitable, average gain per month, they all match up, within reasonable expectations, to the back-tested. A few other other things I haven’t really mentioned yet, but it trades the TY and the ES futures, and those are, I believe those two are the most liquid futures that are traded. So what that means is that we are not trading the EMD or the DOW or something that doesn’t have a lot of liquidity. We’re actually trading the ten year note and the ES, and the bid ask on these is always really high. The ES usually is about at least 500 by 500 on the bid ask. The Ten Year is even more, usually 1000 contracts by 1000. What that means is that with a company like mine, that offers these algorithms, is that there’s plenty of liquidity to meet the orders that the algorithm is generating. It’s 100% automated trading system, it’s 100% algorithmic. I do not ever go in and modify any trades on the algorithm. In other words, it’s 100% technical, I don’t ever say, well I think the market’s going lower, so we’re going to modify a trade. That’s also important, this is 100% automated. The reason why that’s good is it helps remove a big part of the emotions of trading. Anyone that’s traded knows that the biggest battle that you face is yourself, to be honest. You’ll run into the case where you might start moving your stocks or your limits, or there’ll be conflicting technical indicators, so you don’t really know which one is best to follow, so you’ll go with your gut, and nine times out of ten your gut will be wrong, which is why most day traders fail. So the fact that the algorithm is making the decisions helps remove some of the emotions. It can be traded on the Trade Station platform. In other words, our customers, some of them prefer to have Trade Station running, and when that happens we load the algorithms on their platform just like you see here, and they let the algorithms run.

Automated Swing Trading

It can also be auto-executed by one of the NFA-registered brokers with best efforts. We have a list of brokers that support his system, and you can look them up on the NFA website to make sure that they’re okay. I think those are the main highlights. So with that, I’ll quickly go into the correlation analysis between the Ten Year Note and the S&P. Okay, and the reason why the correlation analysis is important, I think, is because a lot of people might say, and rightly so, they might question some of the walk-forward, and they might say things like, well the market has been going higher, and so everyone’s doing well. So I do want to highlight that half of what’s traded in this Swing Trader package is trading the Ten Year Note, and the Ten Year Note does have a negative correlation to the S&P. So in the analysis that I did, I wrote some code that basically bought the S&P on Monday morning, and sold it on Friday, and then also bought the Ten Year Note on Monday, and sold on Friday.  I basically just ran a correlation analysis between the results on a side-by-side comparison. The correlation coefficient is negative 35.19%. What this implies is that there’s a weak to moderate negative correlation between the Ten Year and the ES, the S&P. In layman terms I guess what I’m saying is that on a weekly basis, so if you pull up a weekly chart of the S&P and the Ten Year, if the S&P is up, then normally the Ten Year will be down. Again, it’s not 100%, but that’s kind of the normal, that’s kind of the standard. So there is a negative correlation between the S&P and the Ten Year Note, which is good for us to know, because it helps to answer the question that I also have, which is namely, okay how sure are we that if the market goes into a bear market, which everyone’s expecting, how well will the swing trader do. Because there’s a negative correlation, we do expect that the Ten Year will do well. It’s not to say that the Momentum algorithm will not take losses, for sure it will. It’s not 100%, here’s a loss right here that we took. Here’s another one right here. But the point is is that the gains from the Ten Year, our expectation is that they will offset the losses that you might see on the Momentum algorithm. But I also want to do one quick thing, and show you how this Momentum algorithm did in 2008, which was the last really big bear market that we had.

What you’re looking at here are the results on the Momentum algorithm in ’08, and the reason I want to highlight that is mainly to show that, and this is back-tested, so keep that in mind, but in 2008, when we were in a bear market, this long S&P algorithm still eked out a gain. And so on a $15,000 account, this algorithm still was up about 10%. Now if I click on the Ten Year Note chart, it’ll show you the results that the Ten Year Note saw in that same period. And so in 2008 this algorithm, back-tested, was up about 17K, almost 18K. So if you combine those, then the Ten Year Note was up 18K, and the S&P Momentum algorithm was up about 1800, so back-tested at least it was up about 20,000 on a 15K account in 2008. So I think that mostly covers what I wanted to talk about. Let’s see if I have anything else. I guess I don’t. So I think the last thing I’d like to comment, is just in general, the advantages of the Swing Trader over something like the S&P Crusher which is another algorithm that we have on the website, and then I’ll also comment on the disadvantages. All right, so if you’ve visited our website, or if you’ve been following us, you know that we do offer another package as well, we offer the S&P Crusher and the Swing Trader. The primary advantage of the Swing Trader is that we have over 10% of out-of-sample data on it, meaning the algorithms contained within it have traded live the longest, if you compare that to the S&P Crusher. Now the S&P crusher also trades these two algorithms. So if I was doing a presentation on the S&P Crusher, basically everything I said would apply for the most part, because it also trades these two algorithms. The difference, though, is that the S&P Crusher also trades five other algorithms. But the disadvantage with the S&P Crusher is that some of those algorithms have only traded live now for about, well since October of 2016, so about five months, roughly. So you have the Swing Trader, which has traded live for well over 18 months, and then you have the S&P Crusher, which if you take the shortest period of traded live, which would be from the Morning Gap algorithm, and Covered Calls and the Iron Condor, then we really only have about five months of live data on that, the rest is back-tested. Now the S&P Crusher recently has been doing really well, where the Swing Trader has consistently done well throughout that entire period. So if you’re torn between the two, we cannot give advice, we’re not registered commodity trading advisors, so it’s the kind of thing that people just have to decide for themselves. Now what some people do is they’ll trade half on the Swing Trader and half on the Crusher. When they do that, really what they’re doing is their trading double on these two algorithms, and then one on the others. But then some people really only care about the walk-forward or the live trades, and if that’s your preference, then you would probably consider the Swing Trader. We’ve always like the Crusher, because it does trade seven algorithms, and it trades the Iron Condor or the Covered Calls which do really well in sideways-moving markets, and then it also has three day trade algorithms.

Top Performing Swing Trading System

The bottom line is that the Swing Trader has consistently done better in the live trades. It’s not to say that that will continue, but that’s really the data that we have to go by. Right now we do have the majority of our customers do trade the Crusher, but there are some that trade the Swing Trader. It’s really just something that everybody has to decide on their own. You can talk to a registered investment adviser, or since this is futures, a commodity trading adviser would be more appropriate, and they could answer questions unique to your situation. I mainly, my company takes the position that we want to just be as transparent as we can, just show the data and let people make decisions on their own. So I think we’ve been true to that in showing the data that we have on the Swing Trader. We’ll continue to do the reviews moving forward showing how the algorithms have done. But you know, the Swing Trader, we just can’t deny the data, it’s done really well. I think that’s all I had. So if you have any questions at all, we would love to answer them, we can do a live demo with you, where basically I’ll show this screen with you and walk you through trades and give you any updates you want to know.

With that, I think I’ll sign off. So give us a call today, and we’d love to do a demo, and just remember, I’ll just sign off with the final disclaimer, that trading futures and options is not appropriate for all investors, it does involve substantial risk of loss. The max draw-downs that we showed are measured closing trade to closing trade, consecutive trades, so that it’s a peak to valley, but on a closing trade basis. So that’s good. And I think that’s all, so have a great day, and thank you for watching, and give us a call or shoot us an email, and request a demo and we’d be happy to give you one. .

< BACK TO VIDEOS

DUE DILIGENCE

Reviews & In The Press

Our Design Methodology

History of Algorithmic Trading