Whale Alpha Scoreboard
Dual-track performance: price signal accuracy (Track A) and real whale exit data (Track B). Built on on-chain truth — no survivorship bias, no cherry-picking.
Alpha tier feature · Learn about subscriptions →
How the Scoreboard Works
Two independent scoring systems, one honest picture.
🅰 Track A — Price Signal
Daily snapshot: top tokens by net bullish whale volume from the prior 48–24h window (min 5 whales). Entry price = average USD/token from buy transactions. Verdict at day 7 closing price. BIG WIN = +10%+, WIN = +1% to +10%, FLAT = ±1%, LOSS = <−1%. Peak is tracked separately as the best intra-window high — shown but not used for win rate.
🅱 Track B — Realized Alpha
For each pick, we identify the whale wallets that created the signal. We then scan for BEARISH (sell) transactions by those same wallets on the same token after the pick date. Exit price is estimated from transaction value ÷ token amount. Realized return = (exit − entry) / entry. Only DEX sells are counted — HIGH confidence. Undetected exits show as HOLDING.
📐 Conviction Score
Quality-weighted signal strength. Base formula: whale_count × buy_ratio × log(net_volume). Multiplied by a wallet quality factor (0.5 + avg_score/100) where avg_score comes from wallet reputation rankings. Higher conviction = more reliable signal. Minimum 5 unique whales required.
⚖️ No Survivorship Bias
Picks are sorted newest-first by default — not by best performance. All picks are shown regardless of outcome. Win rate denominator includes FLAT picks (±1%) but not forced-FLAT picks where no price data was ever obtained. 14d and 30d data uses historical CoinGecko prices where available, not current prices masquerading as historical.