NBA DFS Ownership Accuracy
See how close FTA's NBA DraftKings and FanDuel ownership projections come to real tournament ownership, slate by slate.
15 graded days since Nov 7, 2026
9,066 players graded
This page grades FTA's own NBA ownership projections against what the field actually did once each settled DraftKings and FanDuel slate locked — real measured error, not a claim.
Graded since Nov 7, 2026, when FTA began saving the number shown at lock.
Ownership accuracy by site and slate type
-
DraftKings · Classic2.3 pts avg error · 4,527 players graded
-
DraftKings · Showdown (FLEX/UTIL)5.7 pts avg error · 1,499 players graded
-
DraftKings · Showdown (Captain/MVP)1.7 pts avg error · 950 players graded
-
FanDuel · Classic2.2 pts avg error · 4,539 players graded
-
FanDuel · Showdown (FLEX/UTIL)5.5 pts avg error · 983 players graded
-
FanDuel · Showdown (Captain/MVP)2.8 pts avg error · 546 players graded
Weekly accuracy trend (classic slates) NBA's weekly average error, on classic slates.
Day-by-day accuracy calendar (classic slates)
Shade and number show that day's average error across every graded NBA classic slate. Pick a day to see its slates.
November 1, 2026
Clear day ✕No graded slates on this day.
Pick a shaded day on the calendar above to see its slates.
How accurate are FTA's NBA DFS ownership projections?
Over the last 30 days, FTA's NBA DraftKings/FanDuel ownership projections, on classic slates, were within 2.2 points of the field's real ownership on average, across 9,066 graded player projections.
On single-game showdown slates, where ownership concentrates on a handful of players, the error was 4.3 points.
See how last week's projections landed in the NBA weekly DFS ownership report.
Methodology: "Projected" is the ownership percentage FTA showed on NBA's DFS ownership page at slate lock, never revised afterward. "Actual" is the measured share of real tournament lineups that rostered a player, from ingested contest results. FTA calls this gap the average error throughout this page — statisticians know it as MAE (mean absolute error).