Is it possible to track custom stats in results? Is it possible to use tick data?

[Dwyte]

I need a way, per trade to track the maximum profit it can have before it gets stopped out or before the market closes/week ends. Then at the end results I need to tally this property as well like what's the maximum "maximum profit" from all the trades. Is this possible within the library? Can I append custom data on each trade? Thanks, great repo btw!

Edit: Follow up question, is it possible to use tick data instead, then resample it to whatever timeframe I need? thanks!

[kernc]

Thanks!

Can I append custom data on each trade?

Sure. I think something like the following might work. You can access the ran Strategy instance through stats._strategy:

class MyStrategy(Strategy):
    ...
    def next(self):
        ...
        for trade in self.trades:
            trade.max_pl = max(getattr(trade, 'max_pl', -np.inf), trade.pl)

...

stats = bt.run()

mean_max_pl = np.mean([trade.max_pl for trade in stats._strategy.closed_trades])

[kernc]

is it possible to use tick data

No, OHLC is hardcoded, but there's no lower bound: you can have 5-second OHLC bars or even 1-tick OHLC bars (O=H=L=C), limited only by your space/time resources.

[s-kust]

I try to pass some custom data with trades like you say:
trade.custom_data = something
but then I can't see it in stats._trades.tail(20) or stats._strategy.closed_trades[-1]

It shows only standard information.

How to see my custom data (aka custom column) in trades after the backtest has run?

[kernc]

It won't, for the time being, automatically show in trades DataFrame nor in Trade.__repr__. You need to access it exactly the way it was set up:

for trade in stats._strategy.closed_trades:
    this = trade.custom_data
    ...

E.g. to add it to the data frame:

stats._trades.assign(custom_data=[t.custom_data for t in stats._strategy.closed_trades])