What it shows
Four lines, one per lookback window — 30, 90, 180 and 365 days. Each is the share of the top 300 coins that beat Bitcoin over that window. The orange line on the right-hand axis is Bitcoin’s own price.
How to use it
Above 50%, most of the market is outrunning Bitcoin — altcoin season. Below it, Bitcoin is absorbing the capital.
The gap between the lines matters more than any one of them. When the 30-day line pulls away from the 365-day line, money is rotating now rather than continuing a year-long trend, and that turn usually shows up here before it shows up in headline prices.
Read breadth against the price line. Breadth rising while Bitcoin rallies is a broad advance; breadth rising while Bitcoin falls is money leaving Bitcoin. A single number cannot tell those apart.
What the red bands mean
They mark weeks when three quarters of the top 300 beat Bitcoin at once. That has happened three times since August 2024: four weeks around December 2024, then twice a fortnight apart in July and August 2025. Bitcoin fell more than 25% in the months after each one — 27.1%, then 27.6% and 29.2%.
Treat the 2025 pair as a single episode. The week between them dropped back to 50%, and both run down to the same low in November 2025.
Three occurrences is something to watch, not a rule to trade on. The chart marks the condition and leaves the decision to you.
The dotted section
The screener did not run between April and August 2026. Those nineteen weeks are rebuilt from daily prices and drawn dotted — the solid lines stop rather than glide across a gap.
It was an eventful stretch: Bitcoin ran to about $80,900, fell to roughly $58,600, and finished near where it began, while breadth climbed towards 70%. Breadth rising as Bitcoin falls is rotation out of Bitcoin.
Read the dotted lines as the shape of the period, not exact levels — tested against weeks we do have, the rebuild was usually within a point or two but once fifteen out. There is no 365-day line, and no Overheated band: the peak sits five points under the threshold, which is inside the margin of error.
The fine print
Each week the screener takes the top 300 coins by market value, excluding stablecoins and tokenised funds, and pulls a year of daily prices. A coin counts as outperforming if its price in Bitcoin terms is higher than at the start of the window. Bitcoin itself is excluded from the count.
Coins without a full year of history count only in the windows they can support, so the 30-day and 365-day lines have different denominators. That is deliberate: a coin too young to measure is left out rather than counted as having failed.