Top 3 NBA Playoff Surprise Candidates (CURSD Score)
| Rank | Team | Seed | Record | Pythagorean | CURSD Score | Why |
|---|---|---|---|---|---|---|
| 1 | New York Knicks | #3 | 53-29 | 49W | -43 | Played like a #1-2 seed all year. Shot quality, scoring differential, expected win rate all point higher. |
| 2 | LA Clippers | #9 (play-in) | 42-40 | 42W | -32 | Finishing variance and shot quality worked against them for 82 games. |
| 3 | Miami Heat | #10 (play-in) | 43-39 | 43W | -11 | Underlying metrics say they belong higher. Low-seed upset is not random. |
What the CURSD Score Measures
The CURSD Score is a composite metric that combines Pythagorean expectation, shot quality differentials (offensive and defensive eFG%), close-game variance, and schedule strength into a single number. Negative means the team's results were worse than their underlying process deserved. Positive means the opposite.
Across 82 games, these signals are statistically meaningful. They reveal which teams won the games they were supposed to win, and which teams spent the winter cashing coin flips.
The Most Cursed Playoff Teams: Surprise Potential
The New York Knicks finished 53-29 as the three-seed. Their Pythagorean record is 49 wins. Their CURSD Score is -43, the most negative mark of any playoff team in the league.
What that means practically: the Knicks played like a top-two seed all season. Shot quality, scoring differentials, expected win rates - all of it points to a team that was better than its 53-win record. They absorbed the variance the wrong way for four months.
Playoff basketball tends to compress the luck factor. The underlying process reasserts itself when the sample is a seven-game series against one opponent. The data suggests the Knicks enter the bracket performing significantly better than their seed implies.
The LA Clippers sit in the play-in with a -32 CURSD Score. Their finishing variance and shot quality signals worked against them across the full season. The data says they belong in the bracket proper, not fighting for the last spot. They had to earn their way in anyway.
The Miami Heat (-11) and Portland Trail Blazers (-8) show more modest values. Both teams have underlying metrics better than their records show. History says low-seed upsets built on this kind of foundation are not random events.
The Most Blessed Teams: Regression Risk
| Rank | Team | Seed | Record | Pythagorean | CURSD Score | Gap |
|---|---|---|---|---|---|---|
| 1 | Denver Nuggets | #3 | 54-28 | 50W | +47 | +4 wins of pure variance |
| 2 | Detroit Pistons | #1 East | 60-22 | 54W | +39 | +6 wins |
| 3 | LA Lakers | #4 | 53-29 | 46W | +35 | +7 wins (biggest gap) |
The Denver Nuggets won 54 games and hold a three-seed. Their Pythagorean expected wins: 50. CURSD Score: +47. That is the largest positive value of any playoff team, and it is a significant flag.
Variance of that magnitude typically regresses toward process. The Nuggets won four more games than their underlying play justifies. In a playoff series, the market is pricing in a team that played better than it actually did all season.
Detroit finished 60-22 as the East's top seed. Pythagorean says 54 wins - a six-win gap. Oklahoma City (64-18, Pythagorean 58, +37) and Cleveland (52-30, Pythagorean 48, +37) sit in similar territory.
The LA Lakers' gap is the starkest. They finished 53-29 with a Pythagorean record of 46. Seven wins of luck, expressed in games, is roughly a full month of basketball. The data suggests the Lakers' record is flattering them more than any other contender.
What This Means for the 2026 NBA Bracket
The CURSD Score does not predict outcomes. Playoff basketball has its own variance layer, and a blessed team can keep running hot through June.
But if you want to understand which contenders have a foundation of legitimate process, and which ones are walking into the postseason carrying luck debt, the CURSD Score is the most honest accounting available.
The Knicks are undervalued by their seed. The Nuggets, Pistons, Lakers, and Thunder are all carrying positive variance that, historically, tends not to survive contact with a seven-game series.
The data says what the seedings don't.
Methodology
The CURSD Score combines seven signals for NBA teams: Pythagorean gap (point differential vs actual wins), offensive shot quality (eFG% vs expected from shot distribution), defensive shot quality, close-game record, opponent free-throw luck, 2-point conversion variance, and 3-point conversion variance. Each signal is z-scored independently, weighted, and scaled to a -100 to +100 range. Full methodology at cursd.com/methodology.