Best way to test strategies which rely on each tick?

Desktop SDK - Strategies
1

I have a strategy which includes a tick series. The tick series is used for intrabar entries. For example, on a 30-minute chart, I want to detect when the current price goes above the high of the previous bar, so I’m using a tick series to do that.

I’d like to know what the easiest way to backtest is. It seems like the current NinjaTrader backtester only backtests reliably for strategies that do not rely on intra-bar granularity. I am currently going through the playback connection and manually replaying the days, but this is time-consuming.

Does anyone have any ideas or suggestions? I have heard about Tick Replay and have browsed the dev forums for a solution, but I have not found something in the NinjaTrader ecosystem that matches the results I get in the playback connection.

1 Like
2

For NinjaTrader backtesting, I usually have the one tick data series added to the strategy with Tick Replay. However, you need to ensure you downloaded the tick data. The strategy triggers on the primary data series but you enter with the tick series. The most reliable is still the market replay.

I also export historical data from NinjaTrader into a database and have custom python scripts to backtest.

1 Like
3

Unfortunately Market Replay is the only way to get high accuracy in your backtesting and it is some what time-consuming, as it allows you to replay the market exactly as it occurred during real live trading, compared with using the strategy analyzer which isn’t as accurate
and this is well documented on the NT documentation.
To be able to detect if an high has been broken, there are indicators that can detect higher highs and lower lows, i suppose this is what you are looking for

i hope this helps.
Backtestpods

4

no its not exact, not even close to exact, total rubbish actually it is…ninjatrader does not cater to the requirement mentioned in this post.

The only option is running on demo.

5

In your case, you don’t need tick data… Simply set a limit order above/below the 30 min. candle! What you get: precise entries and super-fast backtesting using bar close. This is the best approach I’ve been using for more than 20 years… Of course, live trading can differ from backtesting (gaps at the close of the bar, slippage, etc.). But it is different even if you use a tick replay. Using tick data or tick replies is totally time-wasting…