🧬 Creator Intel — trade the who, not the when

Launches from creators with a prior graduation (the cohort) vs an identical strategy on a 1-in-8 random sample of everyone else (the control). Same entry rule, same honest execution engine, same market minute — the only difference is who deployed the token. Validated on 148k archived launches: cohort tokens pump 38% vs 22% baseline, and the trade-level gap was positive on 13 of 14 days. Judged ONLY on the daily gap — paper P&L magnitudes are known upper bounds.

⚖ The gap (cohort − control, avg %/trade)

—

Validation benchmark: +6.3pp mean, positive 13/14 archived days.
This number staying positive is the entire thesis.

🧬 Cohort (prior-grad creators)

—

Trades

—

Win rate

—

TP exits

—

P&L (paper)

—

🎲 Control (random sample)

—

Trades

—

Win rate

—

TP exits

—

P&L (paper)

—

🏁 Arm leaderboard — every arm vs the shared control (pre-registered 07-19, judged on worst-day gap after 7 days)

Arm	Trades	WR	Avg/trade	P&L SOL	Gap vs control	Gap days +
—

💰 Running P&L (paper SOL, engine v3 reality-calibrated)

Cohort

—

Control

—

Exit prices are sampled from 869 real live trades — no lookahead fantasy. Still paper: judge the gap, not the totals.

cohort control

🚦 Decision gate — gap positive on ≥5 of the first 7 judged days

Waiting for the first judged day (needs ≥10 cohort and ≥20 control trades in a day)…

Day	Cohort n	Cohort avg	Control n	Control avg	Gap	Verdict
—

📜 Recent trades (both arms)

Time	Symbol	Arm	Creator history	Entry vsol	Exit vsol	Exit	Hold	P&L %
—