Methodology

scoring.formula

score = 0.5 * breadth_scaled + 0.5 * velocity_scaled

Each input is rank-scaled across active narratives to 0..1 (best = 1, ties share a value).

A narrative is listed while it has at least one coin first seen in the last 24h. The 'unassigned' bucket is never ranked.

breadth 0.5velocity 0.5holders 0 · phase 2quality 0 · phase 2recency 0 · phase 2

version phase1-breadth-velocity

excluded_inputs

Market data is never a ranking input. These are excluded:

pricemarket capvolumeliquidityfdv

coverage.stats

Coins tracked

120

With description

70.8%

With website

71.7%

With X

95%

With Telegram

18.3%

Deployer resolved

Validation (not used for ranking)

Validation only. Market data below is logged to test whether the score works. It is never used to rank narratives. Observed until ….

by listing

Peak market cap from launch to end of window, coins with any website/X/Telegram vs coins with none. Coins younger than the window are excluded.

1h window

Group	n	Mean	Median	≥10k	≥50k	≥100k
with social or website	39	$45,429	$29,434	79%	31%	10%
without	0	$0	$0	0%	0%	0%

6h window

Group	n	Mean	Median	≥10k	≥50k	≥100k
with social or website	2	$45,080	$45,080	100%	50%	0%
without	0	$0	$0	0%	0%	0%

24h window

Group	n	Mean	Median	≥10k	≥50k	≥100k
with social or website	2	$84,493	$84,493	50%	50%	50%
without	0	$0	$0	0%	0%	0%

by rank bucketproposed

Peak market cap (24h after launch) of coins whose narrative was in the top 3 / ranks 4-10 / below 10 when the coin was first seen. Needs a small exporter query over score_snapshots + market_log; not in market_report.py yet.

24h window

Group	n	Mean	Median	≥10k	≥50k	≥100k
top 3	0	$0	$0	0%	0%	0%
rank 4 10	0	$0	$0	0%	0%	0%
rank 11 plus	2	$84,493	$84,493	50%	50%	50%