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)
Any tool to optimize exits on past discretionary trades?
I am wondering about the existence of a tool meeting the below specification.
Let's suppose that I have a record of my past 100 trades with associated initial stop-loss.
I would like to optimize the exits (stop-loss and profit target), not the entries.
I would like to play scenarios like this one:
- let's suppose that I had used a stop-loss 2 ticks closer, what would have been the consequences on the results?
- let's suppose that I had fixed a R:R = 1:2, what would have been the consequences on the results?
- let's suppose that I had scaled out with such and such rules, what would have been the consequences on the results?
- etc...
To avoid any misunderstanding... I am not trying to optimize the entries. I am trying to optimize the trade management while in trade, by applying several scenarios to actually taken past trades.
Have you ever seen such tool?
Thanks in advance for any comment,
Nicolas
Can you help answer these questions from other members on NexusFi?
Tradervue is a cloud based journaling site with a slew of analytics and broker/platform imports including Ninja/Sierra. I had been keeping an excel journal and found it cumbersome to update and share with others. Tradevue's cloud implementation …
I have had a look on their Web sites and on the 2 related futures.io (formerly BMT) threads.
Apparently, they do not provide the possibility to play scenarios as I evoked here-above.
But I will have a deeper look on both. Might give me ideas.
After further thinking, perhaps the easiest way is to program the tool directly on the platform (Sierra Chart, MultiCharts, etc.).
I could code a trading strategy with:
- entries = those of actual trades taken in the past, read from a text file or whatever
- exits = whatever rules chosen by the user in input
It would allow visualizing directly on the platform (with a precise data feed, etc.) what would have been the outcome of the past trades with other trade management rules.
And the strategy could also print useful metrics (P&L, drawdown, ratios, etc.)
If you use NT, you could enter all your trades with an ATM, and apply a shadow strategy to it. This would only work for forward testing, but it is easy to implement.
The problem is the word discretionary. Is there a correlation between such a trade in the past and one in the future? It is like creating a roulette strategy from past results.
Of course you could create the optimal target for the last 100 trades. But this does mean nothing as long as there is the word discretionary.
I used something like the shadow strategy to proof me that I always took profits way too early. That was a great benefit for me. I knew it before. But after that I had the proof. Nothing more.
It is similar to backtest results vs future results. You need a large sample size, and need to understand it well. But absolutely without question such analysis is extremely fruitful.
In the spreadsheet @vvhg linked to and is now maintaining, I originally created a column similar to "Benchmark". Benchmark is the absolute best case scenario for each trade, in a perfect world this is the exact best case result.
When recording actual trades, I would measure my actual exit against the perfect benchmark exit. This would allow me to get an idea of what I would call my exit efficiency.
In a recent FT71 trader intervention video, you can also see the usefulness of analyzing your discretionary trades:
I fully agree.
My word "optimizing" was not well chosen.
The idea is not at all to optimize ProfitTarget = 2.0358 * StopLoss.
It is more to give "ideas of improvement", "food for thought", "directions to look to".
For instance, the trader could compare 2 scenarios:
- the trade as they occurred, with trader's so-called "active" trade management
- same entries but no intervention on the trade before either pre-defined S/L or pre-defined P/T is reached.
It would show if the "active" trade management is an improvement or... a problem.
Another example : some people say that S/L shall not be put at BE+1 too quick (in order to avoid stop hunting). Others say the contrary: put the S/L at BE+1 as soon as 6 ticks profit (so more BE trades and less losers). I do not want to discuss these strategies here. Exit strategies also depend on... the entries. It would be nice to have a tool to test the 2 scenarios, not on hot air, but on actual past trades. Just to have a rough idea of their impacts.
Once more, no "optimizing", but more "alternative trade management strategies". Just to confirm that everything is fine, or give ideas for improvement.
As you have self mentioned it in #4, try to use the software you have and knowing well. The entries are easy loaded from file and you can apply different exit/Trailing strategies to it. I have developed my own test suite, which will be enhanced day by day. If you have programming skills i can recommend doing this to everybody. You can debug all of your code easily and you aren't restricted by ninjatrader stuff. This falls short if you use many custom indicators from ninja. But i think that somebody (with the right amount of time) could develop a c# framework for individual programs to use ninjatrader indicators. Finally the events (OnMarketUpdate and so on) are transferd to my own toolset and a list of order actions are given back. No need to hassle with ninjatrader.
Thru testing some exit and trailing strategies fast and easily (without ninja) i could improve my initial system a lot. I hope that you also gaining some extra profits from this testing.
