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 am new to this forum and trading in general. That being said, I have learned trading from a trading school. I learnt how to use indicators and how to scalp. I tried to trade using those principles and found myself never being consistent. So I thought of coding a strategy. I have an engineering background and learned to code in Java and C. I learned to code in Ninja Script by reading the help guide, watching webinars and YouTube videos by big mike trading.
About the strategy:
First and foremost, its a scalping strategy. Secondly, I am still developing the strategy. So far, it only trades to go long. I have set the target at 6 ticks in GC. I also have an exit logic in case the trade starts to go against me along with a hard stop loss (1.5 points). Also, its only trading 1 contract. I have attached an image of my back testing results in the attached file. I have yet to do any forward testing of the strategy.
What I wanted to ask here is whether you guys think these results are promising or not. Also, how much should I trust these results. Should I consider this a good point to start or is this information not sufficient to make any decisions yet?
Thanks for reading my post and I would highly appreciate any feedback.
Can you help answer these questions from other members on NexusFi?
based on the screenshot this is too little information to say anything about the quality of the strategy.
For example it's not clear if you optimized the strategy - which to some extend you always do of course when creating a strategy on a set of sample data. But did you let the software search for the best input settings?
Did you account for slippage and commissions in your strategy results?
This was a quick back test of what I have achieved with my strategy. I am still working on it. Yes I am optimizing my entry logic. I have not accounted for commission but looking at the numbers, that would not be a concern at all. However, I will set the commission values in ninja trader for future tests. Also, my orders are all limit order, so that ought to remove slippage, right?
what do you feel besides those points you have mentioned? For example by looking at some of the measures like sharpe ratio and max draw down etc.
All those performance metrics mean little if you curve fitted the strategy to the test period. Take a look at a period of unseen data and see how the strategy performs there with the same settings.
I see your point. However, I did not develop the strategy by keeping a set period in mind. And my optimization does not involve just one instrument on a set period. I am trying to develop my strategy so its transferable in markets and on various instruments. It has similar results over the period of one year. The current setup was 2 months.
I shall also attach the 1 year performance on ES as an attachment.
Have you run your strategy on real-time sim for a period of time , say two weeks? Or you could run it for a session each trading day, then tally up the results by the end of the week.
Estimating about $5.2 commission per trade, your strategy's gross profit made up for it well.
1. How many parameters does your strategy use?
2. How many indicators are you using?
3. What bar type are you using?
4. Are you setting a hard stop and target or something else. Looks like you are using hard target of 10 ticks for both stop and target. Don't need details but a general idea
Your test period is way too small to make any kind of judgement of potential. You need several years of data to have a better idea that it will do well going forward.
You need to get intimately familiar with back testing and optimizing strategies. Do a search on this forum for backtest and/or optimize and you will find an incredible amount of information. You will be overwhelmed by the various viewpoints that will most certainly contradict each other. But those viewpoints even though contradicting, are extremely informative and educational. In most cases the contradictions are right AND wrong but for different reasons. So, don't be distracted by that. YOU have to determine why they are right or wrong. That is the only way you will truly learn.
If you are using exotic bar types like renko, range, linebreak, for example, you better know exactly what they are doing because you will be seriously mislead by the results. Even volume and tick can be misleading. Just be extremely critical of what your strategy is calling a trade. This isn't a comment on your coding ability but a comment about how the bar type is working.
There is so much more I can say about strategies but won't.
1. Two pieces of advice. Don't give up your manual trading for the foreseeable future.
2. Regarding your strategy development, question everything your strategy is doing - make sure the trades taken are exactly what you intended - if the results seem too good to be true, they are and find out why. Quite honestly I think your strategy borders on the too good to be true category because of your basic description of the strategy and the fact that your win rate is so high. I would be very suspicious.
To answer your questions:
1. My strategy does not use any parameters that need to be input. Its entirely indicator based.
2. It uses 3 Indicators and predefined time intervals.
3. I'm using a variation of renko charts. With limit orders.
4. There is a set target for winning trades. For losing, there is an algorithm to detect trend reversal. And there is a hard stop which is placed at the Low of previous reversal or 2 point (picking the smaller one).
I will definitely read up a lot more on renko bar types and understand them better. From what I see on the charts, my strategy is doing exactly what I want it to do. My code reflects my manual trading technique.
I need to work on money management for the strategy. I am quite busy and haven't gotten the time. Will work during the winter break and I'll post my results.
@sangam, I would suggest starting with the core building blocks of algorithmic trading. We just had a webinar on nexusfi.com (formerly BMT) that did an excellent job of laying out all of this, you can watch it here: