My server bot was liquidated in $TIA on Oct 29, on day 12. A 10-day exit would have closed it on Oct 27 at about −$36 instead of −$49. So this time the cut would have won. I’m still not switching — here’s why.

Idle time has a price. A DCA bot has a small take profit and a heavy loss. And while a position is stuck, the slot earns nothing. So the obvious idea: don’t wait, cut the stuck position after a few days and move on. It would be the only exception to my rule: take profit or liquidation, nothing in between.

What two years of history say. I tested it on two years of data, in dollars, floating losses included (table above):
— exit on day 3–5: the worst of all. You lock in the loss right before the bounce;
— exit after 10 days: the most profitable, +$834 vs +$626 for my live rule;
— but the worst half-year doubles: −$125 vs −$62. Max drawdown −$234 vs −$177.

And the peak at 10 days looks like luck of the dates, not a rule: 7 and 14 days already give less. A real edge changes smoothly.

My live rule. So my live bot exits by a stop after 30 days — if it hasn’t taken profit or been stopped by liquidation by then. In two years of history that never happened once. It’s a safety net, not a strategy.

The bot runs on a dedicated sub-account and can size the grid from whatever is on it — one switch in the settings. That turns reinvesting into a simple choice: how much of the profit to leave in.

I chose a calmer worst case over the maximum profit.

What would you pick: +33% profit, or a worst half-year twice as calm?

Follow along: I test what I believe — and drop what the numbers don’t support.

#AlgoTrading #DCA #DCABot #TradingBot

No guessing here about where price goes or which levels to watch. Here it’s statistics and math against an unknown future.

Not financial advice. Leveraged futures can wipe out your deposit.