> For the complete documentation index, see [llms.txt](https://www.degenbot.wiki/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://www.degenbot.wiki/performance-and-methodology.md).

# Performance and methodology

Understand the performance report, accuracy, the null baseline, and the limits of automated analysis.

A performance figure is useful only when you know what was measured.

Start with [Use /performance](/performance-and-methodology/use-performance.md). Then read the explanations below before comparing screenshots or judging a stream by a headline percentage.

| Page                                                                                   | What it explains                                                    |
| -------------------------------------------------------------------------------------- | ------------------------------------------------------------------- |
| [Use /performance](/performance-and-methodology/use-performance.md)                    | How to request the report and identify the sample behind it.        |
| [Accuracy, null, and edge](/performance-and-methodology/accuracy-null-and-edge.md)     | The meaning of the labels, with fictional calculations.             |
| [Evaluating results fairly](/performance-and-methodology/evaluating-results-fairly.md) | Versions, forward observation, backtests, missing data, and timing. |
| [How the analysis works](/performance-and-methodology/how-the-analysis-works.md)       | Data, evaluation, delivery, outcome tracking, and AI limitations.   |

**Alert accuracy is not account return.** A correctly graded observation does not establish that a subscriber obtained its reference price or made money after costs.

Keep each stream, method version, evaluation period, and outcome rule separate. A historical rate does not tell you what the next alert will do.