This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| Ken Pomeroy's ratings | |
|---|---|
| Name | Ken Pomeroy |
| Occupation | Statistician, Analyst, Writer |
| Known for | College basketball analytics |
Ken Pomeroy's ratings Ken Pomeroy's ratings are a suite of college basketball power and efficiency rankings developed by statistician Ken Pomeroy that influenced analytical practices across National Collegiate Athletic Association Division I basketball, sports media, and team analytics. The ratings combine play-by-play-derived efficiencies with tempo adjustments to compare teams and players, becoming a reference for sportswriters, broadcasters, coaches, and data scientists working on March Madness forecasting, bracketology, and performance evaluation. Pomeroy's work intersected with developments in sabermetrics popularized by figures such as Bill James, the analytics approaches of Basketball Reference, and predictive models used by organizations including ESPN and FiveThirtyEight.
Pomeroy produced numeric ratings that quantify offensive and defensive performance, tempo, and luck-adjusted outcomes for college basketball teams, integrating box score information with play-by-play-derived possessions to yield per-possession metrics. The ratings were published on an independent website and cited by outlets such as The New York Times, The Washington Post, USA Today, CBS Sports, and The Athletic, and were used by analysts including KenPom users, Joe Posnanski, and bracketologists following Selection Sunday. The system aimed to reduce biases present in polls like the Associated Press Poll and the Coaches Poll by relying on statistical computations rather than human voting.
Pomeroy's methodology centers on estimating possessions and per-possession efficiencies through play-by-play reconstruction and adjusted regressions; it employs pace estimation similar to approaches used by Dean Oliver and incorporates schedule strength adjustments akin to methods in Elo rating system adaptations. The calculations adjust raw offensive and defensive efficiencies for opponent quality, game location (home/away/neutral), and recent form, using iterative algorithms related to iterative proportional fitting and weighted least squares as practiced in sports rating literature. The approach also models variance and uncertainty in single-game results comparable to probabilistic forecasting used by Nate Silver and FiveThirtyEight.
Key metrics include Adjusted Offensive Efficiency, Adjusted Defensive Efficiency, and Adjusted Tempo, each expressed in points per 100 possessions and adjusted for opponent strength, venue, and schedule. The ratings incorporate outcome prediction via logistic or normal-based models to estimate win probabilities and expected point margins, paralleling methodologies in Pythagorean expectation adaptations and Ken Pomeroy-style predictive frameworks employed by analysts at ESPN Analytics and independent bloggers. Advanced derivatives include luck and variance metrics, lineup-level estimations comparable to approaches used by Synergy Sports Technology and plus-minus analyses akin to Bayesian adjusted plus-minus models used in NBA Advanced Stats.
Teams and coaches at Division I programs, as well as front offices at professional organizations like NBA franchises, referenced Pomeroy-style efficiencies for scouting, scheduling, and player evaluation, while media coverage during NCAA Division I Men's Basketball Tournament selection and seeding quoted ratings alongside metrics from RPI and NET. Bracketologists and predictive modelers integrated Pomeroy ratings into ensemble forecasts with contributions from FiveThirtyEight, The Ringer, and independent data scientists to improve tournament upset predictions. The ratings also influenced analytics curricula at institutions such as Harvard University, Massachusetts Institute of Technology, and Stanford University where sports analytics courses incorporated per-possession analysis.
Critics argued that the ratings, like other efficiency models, can underrepresent contextual factors such as game strategy shifts, lineup minutiae, and matchup-specific scouting exploited in Final Four gameplans; commentators from Sports Illustrated and Bleacher Report noted limitations when comparing teams across nonoverlapping schedules. Methodological critiques referenced potential overfitting, the sensitivity of the schedule-adjustment process to outliers, and the challenge of modeling clutch performance, similar to debates around Adjusted Plus-Minus and small-sample Bayesian estimates. Some coaches and analysts emphasized qualitative scouting from programs such as Duke University and University of Kentucky over sole reliance on metrics for recruiting and in-game decisions.
The ratings emerged in the early 2000s as part of a broader analytics movement that followed breakthroughs by Bill James in baseball and Dean Oliver in basketball, coinciding with the rise of data-driven sports coverage at outlets like ESPN and niche sites such as Basketball-Reference.com and Football Outsiders. Over time, Pomeroy iteratively refined tempo and efficiency calculations, incorporated play-by-play data as its availability expanded through providers like STATS LLC and Synergy Sports Technology, and adapted the model to changing styles of play influenced by teams such as Gonzaga University and Villanova University. The ratings' prominence grew in parallel with advances in computing, open-source statistics in communities around R (programming language) and Python (programming language), and the popularization of predictive bracket competitions.
Mainstream sports media, data journalists, and analytics practitioners widely cited Pomeroy's ratings during March Madness coverage, selection analysis, and season previews, with commentators from ESPN, CBS Sports, The New York Times, and The Athletic leveraging the metrics for narratives and forecasts. The ratings influenced fantasy sports commentary, betting markets monitored by platforms such as DraftKings and FanDuel, and academic research published by scholars at University of Michigan and Northwestern University exploring sports analytics methodologies. While embraced by many in the analytics community, the ratings also sparked debate among traditionalists at institutions like Kansas Jayhawks fanbases and veteran journalists skeptical of quantitative primacy.
Category:College basketball analytics