PRT 10.3 and IG Trailing Stop Woes

[Guest NickWallbridge]

I have been using PRT for sometime now with various strategies. I started using PRT because it is part of the IG package, but the integration between the two seems to be woeful.

 

I have been trying to get a strategy with Trailing Stops to work.  Should be a piece of cake, simply put something like

 

BUY TradeNo CONTRACT AT MARKET
SET STOP pTRAILING 13

 

Setup your entries code and off you go. Did lots of backtesting on the EUR/GBP mini market and then set it live.

 

Problem 1 - The Minimum stop distance keeps changing. Between yesterday and today it went from 5 to 15 and back to 5 again.  This meant that while it was at 15 my strategy was stopped because my code which happened to be at 13 (8:58 BST 23/5/2017).  As far as I can tell there is no way to get the minimum stop distance from IG in the PRT code so the first thing you know your AUTOMATED strategy has stopped is if you go in and look! This really isn't very good. The whole point of an automated strategy is fire and forget at least for a while.

 

So I re-coded to use 16, re-tested and off we went.

 

Problem 2 - The Step distance apparently can't be set in PRT and differs between backtest and live.  As far as I can tell and from info on this forum (https://community.ig.com/t5/Charts/PRT-Automated-trading-issues-and-suggestions/td-p/1388) when PRT backtests it uses a step of 1 (or the same as the initial value?) and when it is live, the step is the minimum for the market, which was now back at 5 (16:31 BST 23/5/2017). So all the Backtesting is invalidated. This makes creating and testing a strategy akin to shooting at a moving target, blindfolded.

 

As IG are clearly promoting PRT (it is now upfront on the dashboard), surely these things ought not to be the case? If  PRT 10.3 can find out the current minimum stop (why does it change anyway?) and if the step can be set in backtesting to match live I would be grateful to know how? 

[Caseynotes]

Hi . You are right that the PRT auto trailing stop is difficult to manage. I see (via your link)   has noted difficulties between backtest and live previously.

The minimum stop level does increase in accordance with increased volatility such as caused by a news event or data release, the increased volatility also increases the spread and the prospect of, and degree of, potential slippage. I presume IG would respond by saying all they could do to avert increases would be to set everything higher in the first place.

In my experience the live auto trailing stop seems to be stepped by 5's, so if set at 6 will step up in the next level up (10's). I agree there is a lack of flexibility and a lack of clarity which is why I only use it occasionally. 

Hopefully Stef can shed more light on the automated strategy and backtest issues.