Backtest
A backtest replays a trading rule on historical data to see what it would have done. It tells you how a rule behaved in the past, not how it will behave.
How Backtest is calculated
A backtest needs a written rule, a clean price history, fill assumptions (when and at what price orders execute), a cost model, and a way to count results. A fair one only uses information available at each moment, and charges costs on every trade. In Tickfloor's Lab a signal on a bar's close is filled at the next open, and costs are charged per side.
How it is read
Read the number of trades, the costs charged, the data window, and what the result was compared against before reading the return.
Common mistakes
- Trying many rules and showing the best one.
- Leaving out costs or using unrealistic fills.
- Treating a good backtest as a forecast.
Test it yourself
The Lab is a backtest builder with receipts: write the rule, choose costs, and read the result with the number of attempts shown. The Lab charges trading costs on every trade, fills on the next day's open, and shows how many attempts you have made.
What Tickfloor tested
Tickfloor's research desk has backtested 678 strategies, net of modelled trading costs. After correcting for the 761 tests run, 0 passed. That does not prove no strategy works. These are historical diagnostics, not validation under Testing Standard v2: the backtest harness predates that standard and has not been re-run to meet it.
Lessons that cover it
- A popular day-trading pattern, tested
- Going live and checking it still works
- Costs inside a backtest
- Preregistration: write the test down before you look
- Data that lies: frozen prices, missing values and unadjusted splits
Related concepts
General information only. It doesn't consider your objectives, finances or needs. Tickfloor holds no financial services licence and never places trades.