Confusing Extended Backtest Results

Questions about MultiCharts and user contributed studies.
wismc848
Posts: 6
Joined: 02 Nov 2015
Has thanked: 2 times
Been thanked: 1 time

Confusing Extended Backtest Results

Postby wismc848 » 13 Nov 2015

I ran a classic method backtest on a strategy using TZA 60 second Ask as data1 and TZA 60 second Bid as data2. Not that it should make any difference here, but TZA runs a Bid/Ask spread typically in the 2 to 3 cent range. The signals I used produced about 20 trades over 4 days. The signals all use "next bar at market" type buy/sell code.

I rerun the backtest in extended mode with data 1 for Ask and Data2 for Bid. I expected to get the same number of trades with greater losses. But the extended mode produced 4 trades. How can this be? Both tests should be generating signals on data1 ask data. Am I misunderstanding the concept or is this a MC programming error?

As a test I ran the extended backtest with data1 selected for both the ask and bid. As I anticipated, the result was identical to classic backtest. But change the Bid (on the backtest setting screen) to data2 Bid data and the results change from 20 trades to 4 trades.

AM I missing something here or or is this an erroneous result?

User avatar
Henry MultiСharts
Posts: 9165
Joined: 25 Aug 2011
Has thanked: 1264 times
Been thanked: 2957 times

Re: Confusing Extended Backtest Results

Postby Henry MultiСharts » 23 Nov 2015

Hello wismc848,

You need to isolate the issue. Check the first trade - if it is not correct then add traces to your code to understand which conditions have triggered it. If it is correct then move to the next trade. Please refer to the following page for debugging code example: Why an Order Was or Was Not Executed


Return to “MultiCharts”