Hockey InsightsDraft Value
Draft Economics · 8,747 picks, 1990–2026 · realized value, not projections

What a pick
is actually worth.

Every draft slot priced by what the players taken there actually did (expected career games, NHLer rates, star rates), on a Schuckers-style curve fit to this dataset. The cliff is steeper than draft night makes it feel.

Pick 100 trades at 13 cents on the dollar.
965 GP
Pick 1 · expected career
vs
126 GP
Pick 100 · expected career
—
Expected GP · Pick 1
—
Expected GP · Pick 100
—
Steepest cliff
—
Star rate · Rounds 4–7
CURVE

The value curve.

expected career GP by slot
Realized production, never a forecast Modern 7-round era · 2005–2025 Mature classes · full careers only Smoothed = 3-slot rolling median

The draft is a cliff, then a beach.

Schuckers-style expected career games played at each overall slot, from the classes old enough to judge. Square-root pick axis so the top of the draft doesn't hide its own drop.

"By pick 100, the expected career is about an eighth of pick 1's. Everything after round three is a lottery ticket with good odds of nothing."

SLOTS

Slot reality.

mature stage · sample-weighted bands
Pick bandExpected GPNHLer rateImpact rateStar rate10-yr stayerSample
RECEIPTS

The receipts.

all-time · 1990–2026 · mature classes only
Value = career GP vs full-history slot expectation Classes 12+ years old: careers, not headlines
All-time steals · outplayed the slot

Found money.

Picks outside round one that returned first-overall careers. Delta is games played above the slot's expected career.

All-time busts · top-10 slots

Sunk cost.

Top-ten selections whose careers fell furthest short of what the slot historically returns.

TEAMS

Who drafts best.

kept-pick production per pick · 1990–2026

"Games played per kept pick is the bluntest honest measure: it can't be gamed by one lottery win, and it punishes teams that draft well and develop badly."