Evaluate a trading bot
Build a useful trading bot evaluation: account for fees and slippage, separate training and test data, and examine failures before trusting a backtest.
Read the guidePut bots to the test
Trading bot research starts with costs, reproducible tests, and a clear account of what can go wrong.
Open calculatorStrategyTest sampleCosts
2.6%
Example monthly cost hurdle
Trading bots
Change one assumption
Full calculatorExplore four illustrative scenarios. These are calculations, not provider quotes or product rankings.
0 bp execution
Illustrative cost model
$30 subscription, $5,000 capital, and 0 bp combined trading fees and slippage.
5 bp execution
Illustrative cost model
$30 subscription, $5,000 capital, and 5 bp combined trading fees and slippage.
10 bp execution
Illustrative cost model
$30 subscription, $5,000 capital, and 10 bp combined trading fees and slippage.
20 bp execution
Illustrative cost model
$30 subscription, $5,000 capital, and 20 bp combined trading fees and slippage.
Start with the mechanics
All guidesBot cost worksheet
Estimate the monthly trading return needed to cover a subscription and execution costs. This is a cost model, not a return forecast.
Open calculatorCompare the approach
Use a consistent set of criteria, inspect the evidence, and understand the limits before choosing a service or method.
Compare methodsKnow the terminology
Look up the terms used in our guides. Each definition focuses on the distinction that changes how you read a cost or a claim.
Browse glossaryOne idea to keep
Trading bots, explained through the details that affect a real decision.
Evaluate a trading bot
A profitable backtest is a result to investigate. It is not evidence that a bot will produce the same return with real orders.
The guide works through the definitions, the arithmetic, and the practical checks. Read the assumptions alongside the example, then use your own inputs in the calculator to see what changes.
Read the full guideWhat we cover
Trading bot research starts with costs, reproducible tests, and a clear account of what can go wrong.
Start with the guide for context, use the calculator to work through a scenario, and check the comparison framework before evaluating a product.
- Backtesting
- Separate development data from the sample used to judge a strategy.
- Execution costs
- Include trading fees, spread, slippage, funding, and the bot subscription.
- Operations
- Consider stale prices, duplicate orders, exchange outages, and reconciliation.
- Review method
- Compare evidence and limits before assigning any score.
Evidence before a verdict
A useful comparison makes its assumptions visible.
Our starting library focuses on explainers and transparent calculations. Example amounts illustrate a method. They are not current market prices, verified provider fees, or measured product performance. Where a detail depends on a service, follow the linked primary documentation and confirm the terms that apply to your account and region.
Reviews should say whether a product was tested, which version or date was examined, and which findings came from documentation. A calculator output does not replace that evidence. We do not publish invented ratings, user counts, or profit claims.
Some future recommendations may use affiliate links. These will be labeled beside the link, and the relationship will be described in our affiliate disclosure. An ordinary source citation is a reference, not an endorsement.
How we prepare content