Macfax Cinderella Index
Upset danger relative to seed and expectation
The Cinderella Index is not a ranking of the best teams. It is a ranking of upset danger relative to expectation — how dangerous a team is when their seed, public perception, or resume understates their actual quality or stylistic threat in a single-elimination context. It combines five structural dimensions into a 0–100 profile score, with higher scores indicating greater misalignment between perceived and actual threat.
What It Measures
The Cinderella Index scores teams on five dimensions: underseeded strength (how much better the team is than their seed implies), defensive disruption capability, possession control and ball security, performance variance (reliance on three-point shooting and pace), and resume legitimacy relative to schedule. Teams with a large gap between perceived strength and actual measured quality score highest. When the bracket is set, a separate P(Sweet 16) probability is also computed using bracket simulation.
Why It Matters
Tournament seeding is partially based on perception, selection committee criteria, and conference record — not purely on adjusted efficiency. A team that is measurably better than their seed suggests, plays high-variance basketball, and has a defensively disruptive style represents structural upset danger regardless of name recognition. The Cinderella Index quantifies that gap, making it easier to identify dangerous matchups before they happen.
How to Interpret
Scores run from 0 to 100. Higher means more structural upset danger relative to expectation. Four tiers: Elite Threat (65–100) — strong across multiple dimensions, genuine seed/perception gap present; Notable Threat (50–64) — real upset potential, at least one or two significant factors working in their favor; Moderate Risk (35–49) — profile broadly matches expectation, limited structural edge; Low Threat (0–34) — profile does not suggest meaningful upset risk above seeding. This is not a quality ranking — a top seed can score high if their true strength far exceeds even a top seed. The metric is most meaningful for mid-major and mid-seeded teams. When tournament seeds are loaded, a P(Sweet 16) probability is shown separately as a bracket-path-aware supplement to the profile score.
Component Weights
| Factor | Weight |
|---|---|
| Underseeded Strength | 28% |
| Defensive Disruption | 27% |
| Possession Control | 21% |
| Performance Variance | 14% |
| Resume Legitimacy | 10% |
Technical Notes
- All five component scores are computed as percentiles against the full Division I population for that season — not just seeded or tournament-qualifying teams. This keeps scores stable and comparable regardless of how teams are filtered in the UI.
- Underseeded Strength is anchored by adjusted efficiency margin. When the bracket is set, a seed residual — the gap between a team's actual seed and the seed their efficiency rank would imply — is incorporated as a secondary signal.
- Defensive Disruption combines adjusted defensive efficiency, shooting suppression, and turnover-forcing ability into a single defensive sub-score.
- Possession Control combines ball security (own turnover rate), turnover-forcing, and offensive rebounding into a possession-leverage sub-score.
- Performance Variance reflects the structural volatility of a team's style — teams with higher three-point shot volume and slower pace tend to produce more variable game-to-game outcomes, which works both ways in a single-elimination game.
- P(Sweet 16) probability, when shown, uses bracket simulation based on adjusted efficiency ratings and the team's actual seed and bracket region. It is a bracket-path estimate, not a general upset indicator.
- Component weights and sub-weights may be recalibrated between seasons as additional tournament validation data accumulates.
- •Cinderella Index is most meaningful during tournament time, when seeds are assigned. In-season use is directional only.
- •Tournament randomness is real — even a low-scoring team can pull an upset through variance alone. The index identifies structural threat, not guaranteed outcomes.
- •The metric works best for mid-major and bubble teams. Extreme seeds (1, 2, 15, 16) have less meaningful scores because the seed/perception gap logic is most informative in the middle of the bracket.
- •Injuries, late lineup changes, and mid-season role shifts are not directly modeled — the index responds only to what the underlying ratings reflect.
- •Early-season scores carry more uncertainty because adjusted ratings are noisier before 15+ games are played.
- •Component weights are still evolving with each additional tournament season of data.
Illustrative: a 12-seed with an adjusted efficiency margin that would typically imply a 5-seed, combined with an elite defense that suppresses shooting and forces turnovers, a high three-point attempt rate, and a legitimate road-game resume — that profile scores in the Elite Threat range. The seed gap is large, the style is high-variance and defensively disruptive, and the resume is real. That is the structural Cinderella profile. A 12-seed with a profile that genuinely matches a 12-seed — average efficiency, average defense, low variance — scores low regardless of seed.