Order Fill Resolution?

[rickyblah12]

Why are Order Fill Resolution settings produce so different backtesting results? if using standard does that means it is accurate?

[NinjaTrader_ChrisL]

Hi rickyblah12, thanks for your note.

The order fill resolution is there because sometimes users want to backtest on something like 60 minute bars. The intra-price movement is not available in historical mode for non-tick-replay scripts, so the higher fill resolution is there to provide a higher resolution time frame to fill orders upon. If your bar series is 1 minute, your fill will probably be fine, if you are running on a 60 minute or 1 day chart this may not be the case. See this documentation on the order fill algorithm that NT uses:

https://ninjatrader.com/support/helpGuides/nt8/?historical_order_backfill_logic.htm

Please let me know if I can assist further.

[rickyblah12]

ok so thank you

1.if im running a strategy on a minute and using High 1 minute or standard does will the results be more accurate?

2. f I use tick replay on a 1 minute timeframe why do the results get alot worst when I do tickreplay on very small timeframes?

[NinjaTrader_ChrisL]

Hi rickyblah12, thanks for your reply.

1. No, the results would be the same if your primary time period and the high-resolution time period are identical.

2. I'm not sure specifically why, but scripts running on tick replay will not improve accuracy alone, you must submit your order upon a 1 tick series to make this effective in tick replay. See this example we have on the subject:

https://ninjatrader.com/support/helpGuides/nt8/?backtesting_ninjascript_strate.htm

Please let me know if I can assist any further.

[rickyblah12]

ok...what about limit orders Order fill resolution...if I use limit orders for a heiken ashi chart shouldnt the results from from standard be the same as 1 tick ? and the same as standard? since its a limit order so the entry price is the same

[NinjaTrader_ChrisL]

Hi rickyblah12, thanks for your reply.

I'm not sure if I understand the question. When Ninjatrader sees an active order in the backtest, it checks if there are any resting orders then does the three passes, comparing Open, High, Low, and Close prices to determine if that order should be filled.

Kind regards.

[rickyblah12]

how close is testing a strategy on playback to live market and how close is playback testing a strategy and testing on historical order fill 1 tick?

[NinjaTrader_ChrisL]

Hi rickyblah12, thanks for your reply.

There is a help guide page here on the differences you can see between live and historical performance:

https://ninjatrader.com/support/helpGuides/nt8/?discrepancies_real-time_vs_bac.htm

While backtesting through the Playback connection can be very close to real time performance, there are factors like high volume and slippage that will not show up in a historical backtest or through playback.

Please let me know if you have any questions on this material.