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)
I haven't posted here in a week or 2, since there really has not been much of interest. I'll post an update at end of week, but performance since the last update has been choppy - up and down. Up a little bit since the last update, but still trading 1 contract.
The most interesting trade happened on Jan 10th. The system put a sell short limit order in at 8 AM. At 8:30:01, the short was filled. At 8:30:06, the stop was hit. 5 seconds for a loss of $450! Fun times!!!
Here are the updated charts as of Week 22 (plus 1 Sunday night trade on Week 23).
So far, the system is at a 32% annual rate of return. Acceptable, but nowhere near what is should be doing. The last few weeks have been pretty representative of this system - meandering up and down, no real progress.
I am still trading with one contract, although I am about $200-300 away from adding on a contract. Hopefully soon.
After 23 weeks, it is time to increase size (again!). If you will recall the first time I increased size, I was hit with a couple of losers, and had to drop back down to 1 contract. It took a while to get me back up above the contract adding equity point ($10,114), but I am now there. Hopefully for a while!
Here are the equity curves as generated by the Tradestation Strategy Backtest Engine. I consider these "perfect" performance - if my real money results match these results, that is good.
As it turns out, I am doing better than "perfect" - you can see that in the table in the post right above (compare "Perfect Cumu" with "Actual Cumulative").
Anyhow, my point for showing this is to show how the walkforward test results match up with the incubation/real money results. I'd say they are both performing as planned, although the nighttime strategy has been in a drawdown for quite a while.
So, after finally increasing account equity enough to jump from 1 contracts to 2 contracts (per the plan), this week I have been ready to trade 2 contracts for any trade, at long as my equity stays above the predetermined cutoff point.
Last night I took the week's first 2 contract trade. Guess what happened?
Of course, a full size loss occurred.
What makes it even worse is that I was 1 or 2 ticks from hitting my profit target, before price dropped, hitting my stop (but turning around only a few ticks lower) and then zooming up past where my profit target was! See the chart below.
Once again, I am left feeling like the market is out to get me...
Next week will probably be back down to 1 contract again.
i am assuming you have done ? the study, individual chart reads, for your losses to optimize against "market is out to get me" few or several ticks stop levels?
are the last few stop-outs simply 'outside x sigma' that you used to define your stop level?
if you were to make stops more robust, wider, how would that affect you historical back-test curve?
It is very probable that my stops right now are not be in the best location. For this strategy, the stop is 34 ticks, which originally came from the TopStep Combine I used this system for. I had to keep the max loss after slippage and commissions to $500 - hence 34 tick limit.
So, when I ran my walkforward analysis, I let the stop be anywhere from 1 tick to 34 ticks. Not surprisingly, 34 ticks was chosen for most walkforward periods as the "optimum." If I had 40 ticks max allowable, I bet the optimum would have been close to 40.
A good question for me now is "since the TopStep loss limit doesn't apply to you now, why not just change your stop?" The answer is that I could, but I decided instead to let "sleeping dogs lie" rather than continuously tinker and adjust what I felt was a decent strategy.
This may seem silly, but why not wait until your account is a bit larger before increase your position size? i.e. ~$13000 or something.
If one trade with 2 cars that hits a loss throws you back below the limit, why not wait until the account is larger and has a cushion? That way you can survive to take multiple trades with 2 cars?
sounds to me your stop level for single individual trades is not adjusted based on some price action behavior and/or price structure but perhaps some type of MAE statistics including TopStep and personal account money availability?
if your stop loss amount cannot be modeled/"explained" through price 'action' then you will have frequently have MAE cases outside your static constant x sigma???
in any case, if this has been working out in your other models then like we established in prior discussions, then it is expected to see drawdowns and flat periods based on your historical curves. your true performance then is only to be accounted for over years not months.
Not silly at all, you bring up a perfectly valid question. Earlier in the thread I explain why I chose the position sizing I did, and basically up front I wanted to add a second contract as quickly as possible, after some reasonable profits were achieved.
You have hit on the flaw with the position sizing I am using - it is a straight calculation going up and going down. So, when I am over $10,140, I go from 1 to 2. If I fall back below $10,140, I drop from 2 to 1. This means that first trade after adding a contract has to be a winner, or I'll have to drop back down.
The alternative to this is to have 2 calculations - one for going up, and 1 for going down. In this case, maybe I would add 1 contract at $10,140, but stay at that level unless equity falls below $9500 (or some other number). Obviously, I'd have to adjust both numbers to meet any risk limits.
I have done exactly this with other systems, but I tried to keep things really simple in this case. That is not looking to be a good decision!
Thanks for the comment. You brought up a great idea!