> For the complete documentation index, see [llms.txt](https://docs.tidelight.app/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.tidelight.app/see-the-signal-between-sessions-4.md).

# See the signal between sessions

## Test the idea.

## Respect the evidence.

Test inspectable rules across Bitget Reality stock tokens and US stock perpetual markets. Compare trend, mean reversion, breakout, and after-hours ideas against completed candles with costs, rolling checks, and a chronological holdout. Eligible runs require at least 60 days of history and 30 days out of sample.

* Public Bitget candles
* Deterministic signals
* Private saved runs

**MINIMUM HOLDOUT**\
**30+ days**

EVENT *→* EVIDENCE *→* EXPOSURE *→* SCENARIO *→* **DECISION**

### Test the rule, then test its limits.

**PAPER RESEARCH ONLY**

Reality rTokens · spotUS stocks · perpetuals

Bitget Reality stock tokens. This market can feed the Nightwatch spot paper account.

### Start with a market idea.

Describe the behavior you want to test. Tidelight maps it to an inspectable rule, then replays it against a Reality rToken with costs and a separate holdout window.

**WHAT WOULD YOU LIKE TO TEST?**

0 /800 · Suggestions are hypotheses, not performance claims.\
Draft a testable idea ↗

* ALPHA RULE
* Trend · SMA 20/50
* Mean reversion · RSI 14
* Momentum · 20/10 channel
* After-hours · weekend drift
* Trend pullback · EMA / RSI
* Semiconductor · volume breakout
* 20-period trend above the 50-period trend.
* Reality rToken symbol
* Candle interval
  * 1 hour
  * 4 hours
  * 1 day
* Run holdout test ↗

**REPRESENTATIVE VALIDATION SET**

* NVIDIA · rToken
* AMD · rToken
* Tesla · rToken
* Apple · rToken
* Microsoft · rToken

Run full matrix ↗

### Rules you can inspect and replay.

Choose among six fixed rule families: trend, mean reversion, price channel, after-hours drift, trend pullback, and a volume-confirmed semiconductor breakout. Each replays one verified Bitget market at a time; the basket comparison selects a training leader but does not model continuous rotation. Signals use completed candles, simulate at the next eligible open, and include estimated fees and slippage. Generated hypotheses select only implemented rules.

|             |                            |
| ----------- | -------------------------- |
| **20 / 50** | Trend reference            |
| **66 / 34** | Training and holdout split |
| **0.15%**   | Assumed cost per side      |

{% hint style="info" %}
Historical replay is not a forecast. Results omit dividends, issuer actions and order-book liquidity; stock perpetual results also omit funding, margin and liquidation. No orders are placed.
{% endhint %}

### Sign in to keep a private run history.

You can inspect the method without signing in; saving results is account scoped.

[Sign in ↗](/see-the-signal-between-sessions-11.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.tidelight.app/see-the-signal-between-sessions-4.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.