How to back test a manual strategy? | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

Manual Grid trading vs Martingale Strategy's 6 replies

Need an EA for this Manual Trading Strategy 2 replies

Why EA test is forward test better than back test?? 1 reply

Manual Entries with Strategy Tester 2 replies

how can i quickly test out manual systems? 1 reply

How to back test a manual strategy?

  • #1
  • First Post: Feb 20, 2020 3:55am Feb 20, 2020 3:55am
Everywhere I read I'm being told to back test my strategy to see if it works but how the hell do you back test a strategy where every trade is based on several different factors. I take into account the trend, the COT report, stochastics, 200day ma, retail sentiment, support and resistance levels etc.

Obviously if your strategy was something like "buy when rsi is over sold and sell when rsi is over bought", well this is simple trading logic and easy to create an EA and back test it automatically.

But for those of who is realise these EA's don't work, how do we go about back testing a manual strategy?
  • #2
  • Feb 20, 2020 4:09am Feb 20, 2020 4:09am
Hi,
When you have multiple setup conditions as described you don't have much choice except to go back through the charts manually and see what happens. Alternatively, you can forward test for a while on demo to see how it goes on paper.

You could also test out each setup condition individually, should be a lot easier and it will also give you some possibly valuable insights into which of the setups are adding the most value and which are possibly redundant. In this way you might be able to usefully simplify your method without any loss of performance.
  • #3
  • Feb 21, 2020 8:25am Feb 21, 2020 8:25am
Quoting Oldtraderman
Hi, When you have multiple setup conditions as described you don't have much choice except to go back through the charts manually and see what happens. Alternatively, you can forward test for a while on demo to see how it goes on paper. You could also test out each setup condition individually, should be a lot easier and it will also give you some possibly valuable insights into which of the setups are adding the most value and which are possibly redundant. In this way you might be able to usefully simplify your method without any loss of performance....
Yep, agree, forward testing on demo then micro account. Although this takes longer, it gives a trader more confidence in their strategy imho.
  • #4
  • Feb 21, 2020 9:23am Feb 21, 2020 9:23am
Quoting Hype
Everywhere I read I'm being told to back test my strategy to see if it works but how the hell do you back test a strategy where every trade is based on several different factors. I take into account the trend, the COT report, stochastics, 200day ma, retail sentiment, support and resistance levels etc. Obviously if your strategy was something like "buy when rsi is over sold and sell when rsi is over bought", well this is simple trading logic and easy to create an EA and back test it automatically. But for those of who is realise these EA's don't...
Well you need to have a complete system, of every step you take in the markets, entries, trade management exits etc... a strategy is just part of your system. Then you test through charts manually with that system, also test it on a demo account too for 3-6 months recommended but also test everything, every process of your system through data and find the logic behind it, so you can always improve it in the future, and understand why you did certain things, and also it's important to keep it logical, so you know what to do in certain situations, since the markets are dynamic. The market never repeats itself, but it rhymes.
  • #5
  • Feb 23, 2020 9:39am Feb 23, 2020 9:39am
Quoting Oldtraderman
Hi, When you have multiple setup conditions as described you don't have much choice except to go back through the charts manually and see what happens. Alternatively, you can forward test for a while on demo to see how it goes on paper. You could also test out each setup condition individually, should be a lot easier and it will also give you some possibly valuable insights into which of the setups are adding the most value and which are possibly redundant. In this way you might be able to usefully simplify your method without any loss of performance....
Micro Futures ( Dow etc) might be the way to get your feet wet ?

https://www.cmegroup.com/trading/equ...i-futures.html

https://www.ampfutures.com/
Europe: https://www.ampglobal.com/
https://ninjatrader.com/Futures

Brokers like AMP and Ninja trader are just two brokers one could use.
every Saint has a past. Every Sinner has a Future
  • #6
  • Last Post: Feb 23, 2020 10:18am Feb 23, 2020 10:18am
It's called grunt work--it's boring and tedious and time consuming yet by manually backtesting you can prove (or disapprove) whatever stategy you have developed. Whatsmore is that you can tweak or adjust it ,refine it if you will, as you go along if neccesary and perhaps wind up with an ammended strategy that is superior to that which you started with. In my opinion strstegies that lend themselves to easy EA backtesting will eventually if not immediately fail, The risk here is that with manual backtesting we all bave a tendency to see the outcomes we hopefully expect rather than those that actually obtain. Any strategy worth anything at all must be manually backtested
0 traders viewing now