Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Just a quick update, it's a little hard updating my journal here while I'm so entrenched with keeping my spreadsheets up to date, my hand written journal, trading, and testing new trade ideas.
After two wins on friday, and then two losses today I am pretty much treading water. Discipline has been very consistent, so just waiting for results.
Alright so an update on my progress. Last Thursday after three losing trades in a row, I hit my max stop loss point....again. So back to the drawing board I went. Fortunately the combine is forcing me to take a hard look at my entire process and maintain a very professional point of view, along with giving me some great practice.
After the last reset I was pretty convinced that I could spread my risk between two markets, and end up with a better result with a lower risk profile. The truth is that if I had a deeper drawdown limit that it might have worked. But the current limit pretty much kept me out of the YM for the last two weeks as volatility soared, and I was forced into trading mostly the Euro. Another key problem I hadn't fully thought through was my max daily loss. My original rules had me quitting after taking two consecutive losses. Well how does that factor when you double the markets you are trading? Yes I had reduced my leverage, but would have to take twice the losses to stay consistent with my rules, which exceeded my daily max loss limit via the combine.
So now to the meat. What I learned after reviewing my process was that I obviously had not back tested enough market cycles even though I thought it was good enough, but also that I had ignored a key component in the systems long term success.
Adam Grimes is a big proponent of the pull back trade, but when I was doing my initial back testing, my results were showing sub par performance for any kind of measured pull-back or momentum rollover as compared to a simple breakout with a large stop. And as you will see in my equity curve model, the system worked great until the beginning of August go figure.(This was when I first started the combine, and it was the only month that the month ended lower than it had started)
I was also missing the fact that buying on retracement limited my slippage and risk quite a substantial amount. That reduced risk goes directly to the bottom line for every unsuccessful trade. So now I have modified my plan to take into account all that I have learned so far, and have added a simple limit order entry on pull back after the initial breakout. The result has surpassed the original system performance with less than half of the drawdown.
You can see where the 20x25PB on the equity curve sliced right through the period of drawdown that the 20x 35 breakout system was experiencing and continued to reach all time highs. So I am trading the new plan starting today, and only the Euro this time with two contracts.
I will continue to back test while I trade the system looking for any new hiccups. I am committed to the process, and reducing errors in all aspects of my trading. Success should not be far off. The pic below is the simulated system performance since the first of the year.
I'm sure you're already aware of this....but I wouldn't get too caught up in those great-looking backtested results. If you have not already, I highly recommend reading @kevinkdog Kevin Davey's book "Building Winning Algorithmic Trading Systems" and, of course, what I consider the bible of systematic trading, "Evidence Based Techncial Analysis" by Aronson which provides answers to the problem of why there's deviation between backtest and future results. Kevin's book is one of the best in detailing how to approach the aforementioned problem.
Thanks I will check it out. It's been a while since I've cruised through Kevin's stuff, but I need to make it a point to re-visit it. Please understand that I am not placing all my expectations of future outcome on my back tests. I am using it how I feel like one should use it, and that is to identify how the system reacts in different market cycles, so that I might notice any common nuances or pitfalls I might be able to easily avoid, as compared with no chart analysis.
So far I have been able to trade the system pretty much exactly how I back test it, and with my simple rules and consistent trade management I see no reason why I should not continue to look at more data. Even if I only get a third of the results that I have seen so far then I will be a very happy camper.
If you have anything else to add please feel free to do so. @kevinkdog as well. Thanks guys.
My main back test goes back one full year. There have been some large swings in direction and volatility throughout the last year in the Euro that I thought it was a good place to start. There were a little over 200 trades which I thought was a decent sample size to start with. I am also in the process of targeting specific environments that are not as common in the past to see how the system performed during those as well.
Specifically the doldrums of last fall where volatility fell off a cliff. And what I am finding is that my current minimum ATR threshold has kept me out of the market during almost all days of super low volatility. So while I may not be making money during that time period, I am also not losing money which is encouraging.
Just finally posting a screenshot of my current TST combine since I haven't done so yet. Just figured out how to use the screenshot function last week. You can see the previous two combines collapsed at the bottom of the picture.
Alright, so time to update some things I've been working on. Its neat how ideas are flowing all the time now since I am focusing on the right things.
So I am in the process of more back testing of my current system, and continue to find great results. Loving the lower risk profile that the pull back aspect has added to the system.
I am also currently developing a systematic, but discretionary approach to swing trading the Euro. It actually seems to work on most markets since it is 100% derived from price action. But it takes advantage of turning points as the market swings from counter trend to trend, which keeps risk lower and maximizes profit. All good things.
Now onto my current project. I was thinking after looking at several trades, that there might have been a flaw in my thinking when I set my current stop at 4 points outside of the previous market swing, so that my stop might not be picked off. Well, after realizing that the 4 points x 2 contracts = $100 (on the Euro) adds quite a bit of drag to my overall performance, since this affects my P&L quite often. Even with a high win rate, since I do not take discretionary exits(or try not to anyway), every trade that does not hit all targets is stopped out at some point and is affected.
So I got to thinking, how many trades out of my entire sample size requires a 4 point stop, and do those couple of trades I might win offset the constant savings of a smaller stop?
Well after doing my initial run, I have come up with another idea. Even though the one point stop does not seem to be the best choice right off the bat, what if I re-entered on all trades that were stopped out by 1-4 ticks, and then broke market structure back in the direction of the trend? And it looks like there is some gold there so that is my next pursuit, but for now here are my findings with a 1-3 point stop, and how they compare.
You can see how the market goes through cycles of running different size stops, but the 3 point stop seems to perform the best. This still saves two points on a large amount of trades vs my current 4 point stop. Next I will post the new equity curves for all three vs my current, and see how it affects the bottom line.
So already I can see that the minimal impact of the 3 point stop on the current system makes running the 1 and 2 point stop numbers as they stand a moot point.
So two thoughts cross my mind immediately. This is something I can implement immediately with little effort on my part, and have a slight positive outcome on the overall performance. This adds no extra rules to my current plan. Or I can continue down the rabbit hole and see if something major happens by the reversal idea on the one point stop. Well no point in stopping now right?
----------------------------------------------------------------------------------------------------------------------------
Ok now we are getting somewhere. The average MAE for the trades that are stopped out and then reverse is over 35 points! So I just set my normal targets of 20 & 25 for the reversal trades, and we have success with a high hit percentage. Even the couple of trades that did not make target were stopped out at a profit. So unfortunately the sample size is low, but the win rate is above 90% for the trades available. So I will keep this in mind going forward for back tests, but I plan on implementing this immediately.
Here is the new equity curve compared to the previous ones. Something else that is not considered with the equity curve is the fact that trade size will increase as the account grows.