chainmeter
EN | 中文 | ES

Article

Bitcoin Composite Cycle Score: Combining 8 Indicators Into One Number

Chainmeter · ~2 min read

Eight indicators can easily disagree with each other — MVRV Z-Score might read hot while Puell Multiple sits neutral. Composite exists to collapse that disagreement into one number, but not by averaging all eight opinions as if they carried equal weight: an indicator built from measured on-chain, cost-basis data gets more say than one built from fitting a curve to price history. The output is a single 0–100 score, deliberately weighted, not just averaged.

How the weighting works

Rather than treating every indicator as equally informative, the Composite score weights each of its eight inputs according to this site's own tier structure: Tier 1 on-chain, cost-basis indicators (MVRV Z-Score, NUPL, realized price) carry the most weight; Tier 2 production-cost indicators (Puell Multiple) carry meaningfully less; and fitted statistical models carry the least. The logic is explicit and disclosed on this site's methodology page, not a black-box weighting scheme.

How this differs from the Market Dashboard

This site's Market Dashboard takes a deliberately different approach: it converts five core indicators into percentile ranks against their own individual history and averages them equally, specifically to sidestep the risk of hardcoded, easily-miscalibrated absolute thresholds. Composite instead uses each indicator's own native scale and explicit tier-based weights. The two pages are complementary lenses on similar underlying data, not competing claims about the "true" market state — reading them together, and noting where they agree or disagree, is more informative than trusting either alone.

What a single 0-100 number can't capture

Compressing eight distinct indicators, built from different data types and measuring different aspects of market conditions, into one number necessarily discards information about which specific indicators are driving the current reading. A Composite score of 70 reached because every input agrees looks identical, at the headline level, to a 70 reached because a few extreme readings dragged the average up while others disagreed — which is why this site's Composite page shows the full breakdown by indicator, not just the final number.

Today's composite score, broken down by every underlying indicator, updates live at the Composite Cycle Score page.

FAQ

Is Composite score more reliable than the Market Dashboard?
They're different, complementary methodologies rather than one being strictly better. Composite uses tier-based weighting on native indicator scales; the Dashboard uses equal-weighted percentile ranks. Reading both, and their component breakdowns, gives a fuller picture than either alone.
Can the Composite score be gamed by one extreme indicator?
A single indicator's tier weight limits, but doesn't eliminate, its influence on the total. That's exactly why the full per-indicator breakdown is shown alongside the headline number rather than hiding the composition.
Why not just average all eight indicators equally?
Because this site treats on-chain, cost-basis indicators as more rigorously measured than fitted statistical models (as explained on the methodology page), an equal-weighting scheme would give a curve-fit model like Power Law the same influence as a directly-measured metric like realized price — a choice this site's tier structure is specifically designed to avoid.
This site has more than eight indicators — why does Composite only use eight?
The eight were chosen for having a long enough, cleanly measurable history across every past cycle to normalize against. Newer or flow-based additions like exchange netflow and reserves, or shorter-history tools like hash ribbons, aren't in the blend, though they're worth checking individually alongside the composite score.