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)
That is a great question. When I test with @ES and then @MES, I always make sure the Net P/L of @MES is 10% of @ES, if the slippage and commission for @MES is set at 10% of @ES.
If it is, then I adjust @MES slip and comm to actual values.
If there is a mismatch, it is almost certainly a code issue, since most of the @MES history is really scaled @ES history. Either that or your strategy is REALLY sensitive to 1 or 2 tick differences in OHLC prices...
Honestly I wish I new. I have not explored it deeply enough, but I can run one of my systems on RTY and M2K, for example, and RTY will pass and M2K will fail. It may be that I am over-estimating slippage or the micro trading costs are just high enough to bump it out. I have not had time to look into this, as my workload right now is at about 125% my capacity, but given your earlier comments, I really want to build a micro portfolio from my system inventory and see what that looks like. I probably won't get to it until late August, maybe September. Hopefully by the time of the Beyond Factory event.
The 10% rule is useful, but the problem with it is that it underestimates commission. This post in my journal is a bit dated (last year), but I looked at costs for ES versus MES last year, when I was trading MES and MNQ (the presentation is a little confusing, but the point is there):
This is a supplemental entry, but one that reflects how much I do not know. I did a quick search last night for micro e-minis, mainly because I wanted the contract specs for MES (micro e-mini S&P). I ended on the CME micro e-mini page and was stunned: …
I know the micro costs have gone up since I wrote that, but at the time, commission and fees for micro were 150% of the mini. I am guessing, at least in my testing, that this is where my micros are failing.
I understand how the data is built for the micros before inception. As for the code, the code is always identical. The only variables are commission and slippage (and of course, minor price differences between micro and mini). These difference can echo into optimizations for the systems I optimize. I could, in theory, optimize ES and use the same parameters for MES, or vice versa. But that is a topic for another day.
In any case, I don't have much interest in trading them at this point. I will explore the idea @SMCJB inspired, by building a micro portfolio, hopefully sooner rather than later.
The 10% is just to compare MES to ES directly. MES results SHOULD be 10% of ES.
If that is true, then you know that the strategy is set up correctly.
After that check, then you can adjust MES slippage and commission as appropriate.
I didn't explain it well, but doing this 10% exercise identifies/eliminates code issues (like maybe having SetStopLoss(X) instead of a contract size adjusted stop SetStopLoss(Y*BigPointValue).
Thanks @kevinkdog, as always. In a few systems, I use a limit order based on a calculation from a prior bar's high or low. For the period in which the micro data is real (i.e. from inception), there are differences in bar highs and lows which would explain some of the differences in results. In the end it may all wash out, but it will require some more analysis on my end to see what is happening.
If I remember correctly, @SMCJB and I discussed using ES on data2 last year, trading MES on data1 based on the data2 signals. This would give us the perfect 1:10 ratio, commissions/fees aside.
Trading: Primarily Energy but also a little Equities, Fixed Income, Metals, U308 and Crypto.
Frequency: Many times daily
Duration: Never
Posts: 5,059 since Dec 2013
Thanks Given: 4,410
Thanks Received: 10,226
Big event this last weekend in Cleveland. Seventeen people (including multiple nexusfi.com members) from 5 different counties, 3.5 days of advanced level algo designing, collaboration & discussion. Very unique experience. Thanks Kevin.
Thanks for sharing @SMCJB. Now I can put names to a few faces (Kevin shared as well). I hope you were able to build some useful algos and have some fun at the same time.
Awesome event indeed! Really enjoyed meeting Kevin and so many others who will now play a critical roll in my development as a trader. I feel like a turtle after hanging with those guys. Super smart group of active traders…all with their own little EL script robots of various kinds.
I came away with more ideas to test than I have time to test. The CPU is the limiting factor now.