YouTuber ScorecardDid what they said actually happen?

Only clear bullish/bearish calls on a specific ticker count. The baseline is the close on the call's trading day; the outcome is settled at the close 60 trading days later. All returns use adjusted prices (splits are not gains or losses).

Excess return = directional return − the SPY benchmark in the same direction. For a bullish call: ticker gain − SPY gain; bearish calls are mirrored as shorts. A "hit" requires the right direction and beating the benchmark.

Calibration = hit rate of high-conviction calls minus hit rate of regular calls (a positive value means strong convictions really do land better; hidden when either side lacks samples). Clarity = share of a YouTuber's opinions that are judgeable at all.

No score below 30 samples. This is a post-hoc check on public statements — not a judgement of anyone's skill, and it says nothing about what happens next.

The list is ordered in two tiers: scored YouTubers (≥ 30 samples) first by avg excess return, then those still collecting data by sample size — unscored numbers never compete for rank. Tags under each name are derived from the data: avg excess ≥ +0.5% earns "Beats SPY", ≤ −2% "Lags SPY"; calibration ≥ +15pp earns "High-conviction calls land", ≤ −15pp "High-conviction calls miss"; clarity ≥ 40% earns "Mostly ticker-level calls", ≤ 15% "Mostly macro commentary". At most two tags per YouTuber, picked in that order; no tags before a score exists. Display tiering: feed cards show only positive/neutral tags — negative ones appear in full on channel pages and here. The data itself is never adjusted.