Coins sorted by how much they trade over 30 days: the 10 most traded (large caps), ranks 11 to 50 (mid caps), ranks 51 to 200 (small caps) and ranks 201 to 500 (micro caps). Over 1,634 days in bull markets and 1,402 in bear markets (Aug 17, 2017 to Oct 7, 2026), each lower tier moved with the large caps the same day: a correlation of +0.76 to +0.92. A day later it was −0.09 to +0.04: next to nothing. The catch-up trade, buying the tier that lagged after the large caps ran ahead: 67 of its 324 versions had enough cases, 0 passed the first cut and 0 also earned on the newer 40% of the history. Before fees. Past patterns, not financial advice.
as of · our daily closes, Aug 17, 2017 to Oct 7, 2026
Each bar is how closely a lower tier's daily move followed the large caps' move: on the same day, then 1 to 14 days later. 1 would be moving in lockstep, 0 no link at all.
Market
Tier
Days
Same day
+1
+2
+3
+5
+7
+14
Back, +1
Bull
Mid caps
1,634
+0.86
−0.04
+0.02
+0.00
−0.00
−0.02
−0.01
−0.06
Bull
Small caps
1,457
+0.81
−0.01
+0.02
+0.01
−0.02
−0.02
−0.02
−0.06
Bull
Micro caps
980
+0.76
+0.04
−0.03
+0.02
+0.00
−0.06
+0.06
−0.06
Bear
Mid caps
1,402
+0.92
−0.09
+0.07
−0.01
−0.07
+0.00
−0.01
−0.13
Bear
Small caps
1,110
+0.90
−0.08
+0.04
−0.01
−0.11
−0.01
−0.02
−0.15
Bear
Micro caps
846
+0.84
−0.01
+0.06
+0.05
−0.00
−0.02
−0.07
−0.07
Bull markets: Bitcoin above its 200-day average; bear markets: below it. "Back, +1" is the other way round: the tier's day against the large caps' next day.
§ 02 · The catch-up trade0 held
The catch-up trade
as of · the strategy search's gate
The idea: after the large caps ran ahead of a lower tier, buy that tier and short the whole market, betting it catches up. Or, inside the lagging tier, buy the coins that move most with the large caps and short the ones that move least. 324 versions: how far the large caps ran, over how many days, how far the tier lagged, and how long to hold.
67 versions had enough cases to test; with that many at once, a luck test of about 3.4 was needed to pass the first cut. 0 passed it, and 0 also earned on the newer part.
Not one of the 67 versions passed both cuts.
The closest misses, not past the cut
Version
Older t
Newer t
Average
Middle
Cases
After the 10 most traded coins rose 5% or more in 14 days while the mid caps trailed them by 5 points or more, the mid caps trailed the whole market over the next 14 days
1.7
2.4
+1.97%
+1.75%
64
After the 10 most traded coins rose 10% or more in 14 days while the mid caps trailed them by 5 points or more, the mid caps trailed the whole market over the next 14 days
1.9
1.8
+1.57%
+1.65%
51
After the 10 most traded coins rose 5% or more in 30 days while the micro caps trailed them by 10 points or more, the micro caps beat the whole market over the next 7 days
1.4
2.2
+0.40%
+0.19%
58
After the 10 most traded coins rose 20% or more in 30 days while the mid caps trailed them by 5 points or more, the mid caps trailed the whole market over the next 7 days
0.4
3.0
+0.82%
+0.60%
70
After the 10 most traded coins rose 10% or more in 30 days while the micro caps trailed them by 10 points or more, the micro caps beat the whole market over the next 7 days
1.3
2.0
+0.39%
+0.19%
52
§ 03 · How it is tested
How it is tested
as of · methodology
A luck test (t) says how far a result sits from zero, in units of its own noise: about 2 is rarely luck for one test, but many tests at once need more, which is what the first cut is for.
The tiers are re-drawn at every close by each coin's median dollar volume over 30 days (large 1–10, mid 11–50, small 51–200, micro 201–500), pegged and wrapped coins left out, each tier an equal-weight basket. The same-day and later links are read off the whole history and choose no trade.
A catch-up version is a trade: it passes a 5% false-discovery cut over every test at once on the older 60% of the history, then a luck test of 1.645 or more on the same side on the newer 40% it was not chosen on, every case counted. No two cases of a version overlap. A trade has four sides (the tier bought, the market sold, and back), so "room for fees" is its average move split four ways.
Our price history holds today's coins: coins that died before we began collecting are missing, which flatters the older years.
§ 04 · Revisions1 dated
Revisions
as of · every change to its headline numbers, dated
This study is re-run every night on the whole history. It was first written on Oct 6, 2026; its numbers last moved on Oct 6, 2026. Each night its headline numbers moved is listed here, newest first.