Bitcoin Market Temperature — Chainmeter Composite Score

A single 0–100 score aggregating all eight on-chain and statistical indicators, weighted by methodological rigor. One number for how hot the market is running.

Data: PriceUSD, CapMrktCurUSD, CapMVRVCur, IssTotUSD, FeeTotNtv · Coin Metrics community API · computed live in your browser, nothing uploaded

What this score shows

The composite score takes a weighted average of eight indicator readings, each normalized to a 0–100 scale where 0 represents historical accumulation extremes and 100 represents historical distribution/top extremes. Indicators are weighted by their methodological tier:

Score = Σ (normalized_indicator × weight) across all 8 indicators

Higher-tier indicators (on-chain cost basis) receive more weight than lower-tier ones (fitted models). The weights are fixed and transparent — see the breakdown cards above.

Score zones

0–20 · COLD: Most indicators near historical accumulation extremes. Rare. Every major cycle bottom has registered in this zone.
20–40 · COOL: Below-average market temperature. Characteristic of early recovery phases.
40–60 · NEUTRAL: Where Bitcoin spends most of its life. No strong signal in either direction.
60–80 · WARM: Above-average temperature. Consistent with mid-to-late bull market conditions.
80–100 · HOT: Most indicators approaching historical top extremes. Every cycle peak has entered this zone.

Limitations

This score inherits all the limitations of its component indicators. It cannot tell you when a top or bottom will occur, only where current conditions sit relative to a small sample of historical cycles. The weights are a design choice, not a mathematically derived optimum — different weightings would produce different scores.

Aggregation also conceals divergences. When on-chain indicators are cold but price-based indicators are warm (or vice versa), the composite smooths over a meaningful disagreement. Always check the individual indicator pages for the full picture.

FAQ

Why not just use a simple average?

The eight indicators are not equally trustworthy. MVRV Z-Score and NUPL are derived from directly measured on-chain quantities with clear economic interpretations. Pi Cycle is data-mined folklore. Weighting equally would treat them as equivalent. The tier-based weighting reflects the methodological hierarchy documented on the Methodology page.

Why doesn't the score perfectly predict tops and bottoms?

No aggregate of descriptive indicators predicts anything. The score tells you where conditions currently sit relative to historical episodes; it makes no claim about what comes next. Bitcoin has had four measurable cycles — that is not a sample large enough to build a reliable forecasting model.

How is the historical score computed?

The same formula is applied to each day's historical data — each indicator is normalized using fixed historical ranges, then combined with fixed weights. Because the normalization ranges are fixed (not rolling), past scores do not repaint when new data arrives.