Discussion re Pure Renko strategy
- This topic has 345 replies, 24 voices, and was last updated 4 years ago by
.
-
The backtest looks amazing! And I also love the 10 sec shift, it reduces the number of trades and that’s good.
The problem is still that we can’t test it on live demo, spreads only go up and up…For the forex yes, it couldn’t be suitable because there are no “fixed spreads” like Wall Street, but having a look on some pairs, I noticed that in the 8-22 range the EURUSD is pretty solid between 0.6 and 1.2, so I think it’s the only pair where could be worth develope the strategy in the future without having to worry about spread changes.
As always, thanks for your contribution Paul 🙂
1 user thanked author for this post.
no itf, only a pic to compare backengine to live demo.
the big loss of 0.5% was unfortunate, right at the top before going down.
backtest will normally be much better then demo (no spread i.e. for the exit, and live demo will be better then real.)
example old scenario, a profit target at 50 points. stoploss 50 points
long in the back-tester (spread at 2.8) at 19000 and goes to 19051 and goes down again. Backtest result 50 points gain.
in demo/live however, the price paid is 19003. It never reaches the 50 points gain and you go out with 50 points loss.
In my thinking, the backtest is better to be optimised on the entryprice(=orderprice with spread/slippage) I’ve set and calculated from that point for the stoploss/pt/ts. Preferably a worse price than in reality. In my backtest I never see that trade as a win, but as a loss. So that means, if I have a good backtest, it’s more likely to hold up in live trading. Does that make sense?
As result of my approach and the spread set in the backengine to 0, it does include spread/slippage (set in the code) in the orderprice and is used to calculate the exits, but not used when the exit actually happens in the backtest (because spread is set to 0)
Anyway, I don’t know, it’s just a different approach.
4 users thanked author for this post.
Bear with me, I’am poor with explaining 🙂 Not easy to put into words. Maybe wrong too!
The spreadbox is too resemble as close as possible the price paid as in live trading. But the big downside is that it ignores time and spread differences & slippage. It falls short. In live/demo you are much likely to pay more which effects everything.
Normally, if spread in the backtestengine is set to 0, you see the paid price is the open of the bar for the market order.
In above backtest long scenario, (spread 2.8, which adds +1.4 to the open price)
if criteria is reached and bought at the next bar (open=18998.6) you see in the backtest a paid price of 19000 & profit target is reached.
In demo if criteria is reached and bought at the next bar (open=18998.6) and you get 19003 as fill, no profit target is reached and get out on stoploss.
Difficult to put pro’s & cons to words. Nothing is ideal. It’s comparing what’s most comfortable/reliable.
The price you get from IG influences everything. It’s better to use your own terms.
Find exclusive trading pro-tools on 
