Appearance

Personalize your experience

Mode

Accent Color

Direction

Python: Event-Driven Strategies

Python SDK

Subclass quatick.Strategy to trade bar by bar with real orders: brackets, trailing stops, limits, risk-based sizing and a drawdown kill switch. The same file backtests locally and on Quatick.

A complete strategy

breakout.py

1from quatick import Strategy, ta, run, Bars23class Breakout(Strategy):4    params = {"n": 20, "atr_mult": 2.0, "risk_pct": 1.0}56    def init(self, ctx):                       # once, before the first bar7        b = ctx.bars8        self.hi = ta.highest(b.high, self.p.n)9        self.atr = ta.atr(b.high, b.low, b.close, 14)1011    def on_bar(self, ctx):                     # after every bar closes12        i = ctx.i13        if ctx.position or ctx.open_orders or i == 0 or ta.is_na(self.hi[i - 1]):14            return15        if ctx.close > self.hi[i - 1]:16            stop = ctx.close - self.p.atr_mult * self.atr[i]17            qty = ctx.size_by_risk(self.p.risk_pct, ctx.close - stop)18            ctx.buy(qty, sl=stop, tp_pct=6, tag="breakout")1920    def on_fill(self, ctx, fill):21        ctx.log("filled", fill.tag, fill.qty, "@", round(fill.price, 2))2223if __name__ == "__main__":24    bars = Bars.read_csv("btc_1h.csv", interval="1h")25    r = run(Breakout, bars, capital=100_000, costs={"kind": "crypto_maker_taker", "value": 0.02, "taker": 0.05},26            max_drawdown_stop_pct=25)27    print(r.summary())

Upload the file as a and the server finds the subclass, runs it with the page's costs and settings, and scores it like any other . Set = Breakout if a file defines several.

The Context (ctx)

Member	Meaning
ctx.i, ctx.time	index and time of the bar that just closed
ctx.open / high / low / close / volume	that bar
ctx.history(column, n)	the last n values up to this bar (never )
ctx.bars	the whole series (for indicators in init; index them with ctx.i)
ctx.position, ctx.avg_price, ctx.cash, ctx.equity	the account at this close
ctx.open_orders, ctx.halted	working orders; whether the limit stopped trading
ctx.buy(qty, limit=, stop=, sl=, tp=, sl_pct=, tp_pct=, trail_pct=, trail_amount=, tif=, expire_bars=, reduce_only=, oco=, tag=)	send an order (the type follows from the prices given)
ctx.sell(...)	the same, selling
ctx.order_target(units), ctx.order_target_percent(fraction)	market order for the difference to a target position
ctx.close_position(), ctx.(order), ctx.cancel_all()	exits and cancels
ctx.size_by_risk(risk_pct, risk_per_unit), ctx.size_pct(pct)	position sizing, rounded down to lot_size
ctx.log(...)	a line in the run log (shown in the Report and views)

Caution:Indicators computed in init span the whole series. Only read them at index ctx.i or earlier; the lookahead check catches strategies that do not.

The result

Field	Contents
r.stats	the engine (, , calmar, maxDrawdownPct, profitFactor, ...)
r.	Trade objects: entry / exit bar, time and price, qty, pnl, costs, entry_tag, exit_tag, mae_pct, mfe_pct
r., r.orders	every fill and every order with its final status and reason
r.equity, r.position	per bar
r.halted_at, r.logs	the bar the limit fired on; the run log

Appearance

Python: Event-Driven Strategies

A complete strategy

The Context (ctx)

The result

See also

Related articles

Appearance

A complete strategy

The Context (ctx)

The result

See also

Related articles

Related to Python: Event-Driven Strategies