> For the complete documentation index, see [llms.txt](https://crushtrading.gitbook.io/documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://crushtrading.gitbook.io/documentation/readme/why-ai-quant-trading.md).

# Why Crush — Signals With Receipts

Why every signal in Crush carries measured history — sample size, win rate, real outcomes

Anyone can show you a signal. Social feeds are full of them: someone is always certain, and the losses quietly disappear from the story. The question that actually matters never changes: **should I take this trade?**

Most products answer that question with confidence. Crush answers it with receipts.

## What a receipt looks like

* **A measured history.** Data-driven setups are backtested over years of market history before they ever reach your feed. Each one shows its sample size ("52% hit rate over 21 signals"), typical resolution time, and how it has performed live since publication.
* **Honest labeling.** Simulated history is always marked as simulated. Live performance is marked live. Your own results are shown net — after fees — because that's the only number that's real to you.
* **Freshness you can see.** Signals expire. A card tells you when it appeared and when it stops being valid, so you're never trading yesterday's idea.
* **Outcomes, not amnesia.** When a setup resolves — win or lose — the result is shown. The losses stay on the record. That's the point.
* **Provenance for disclosure signals.** When a card is built from a public filing — an insider purchase, a fund's disclosed position — the receipt is the filing itself: who filed, when, and how fresh it is by the time you see it.

## The house rule

If it can't carry a receipt, it doesn't ship. That standard costs us speed and flash. It's also the entire reason to use Crush instead of following another confident stranger.