Contact number in UK
- This topic has 14 replies, 4 voices, and was last updated 3 months ago by
.
-
It is my understanding that Algos that work with IG dont work with IB and vice versa so you could be starting a lot of heartache for yourself by opening an Account with IB.
There is lots on here by PeterSt as to why above is the case.
1 user thanked author for this post.
PeterSt posts may be in here …
https://www.prorealcode.com/topic/automated-trading-interactive-broker-prorealtime/
1 user thanked author for this post.
Maybe I can reassure you a bit…
I have two identical systems running on IB and IG and they run 95% parallel to each other, the (small) differences in execution is due to the difference in price/volume between the CFD (IG) and the future (IB)…
There is lots on here by PeterSt as to why above is the case.
It depends largely on how close to the facts you operate;
If you have a daily timeframe nothing will be different much (better read : nothing at all). But would you mere scalp and use a “seconds” timeframe in order to do your thing, nothing will work at all.I would like to attest that this is not related to CFDs as such, but merely to the way IG operate their business. Take Forex for example; both IB and IG use CFDs for that. IB is hard to differentiate from Spot Fx (at least for EUR/USD which I sorted out at the moment IB offered both spot and CFD). Now compare that to IG for a smaller average (say 100 bars of 1 second) and be shocked. Just nothing compares and this starts with the 10 times or so more “volatility (ahem) of IG. So it is movement (doing their own business) all over the place, and no scalping can deal with that.
Using a time frame of 1 minute already bears very different results, but depending on what you do it could be portable. It could even be so that the results are only 5% off, but that 5% could just be the matter of day and night of winning vs losing all. Coming from IB you could re-optimize in IG so all looks good again. But then IG will change their internal strategies and you will later blame changed markets for your algo not wanting to work any more.
This latter really did it for me :
Over and over I saw big changes with backtesting at certain dates without me recognizing the why of it (like no new Putins or Trumps or Covids). Then I started to work with IB for the same instruments and nothing of such changes ever occurred (unless with Putin etc. reasons). So Yes, CFDs may matter, but it is about what’s done with them;I always claimed or saw that the “spread” visible with what IG makes of it vs reality, is exactly the playground IG could use to take your money. As long as they stay within their own offset bounds … (prove me wrong but please mind the italics in could).
I have running 12 systems in PRT-IG for 18 months without ever looking at them. That were 25 systems at first, which I reviewed last January (2024). I then ditched 13 mere rotten apples who could not stand the times and upscaled the remaining. If you look at the overall result in that instrument in IG now ( 🙂 🙂 )… But this can only be done by empirical finding. Say machine learning by my own eyes without intelligence. – blurp
1 user thanked author for this post.
Find exclusive trading pro-tools on 
