Fri Mar 14 23:50:00 UTC 2025: ## College Basketball’s March Madness Embraces Analytics, But Debate Rages On

**INDIANAPOLIS** – The NCAA men’s basketball tournament selection committee is using advanced metrics more than ever, but the process remains a source of contention among statisticians, coaches, and analysts. For the first time, seven different ranking systems, including the newly introduced Wins Above Bubble (WAB), inform the committee’s decisions. While the use of analytics is lauded as an improvement over outdated methods, the question of how – and even *which* – metrics should be used remains fiercely debated.

Bart Torvik, creator of the influential barttorvik.com rankings, now officially part of the selection process, admits to feeling the pressure. His website, along with KenPom, ESPN’s BPI, and the NCAA’s own NET rating, are considered “predictive” rankings, focusing on team efficiency. In contrast, results-based metrics like ESPN’s Strength of Record, the Kevin Pauga Index (KPI), and WAB assess a team’s accomplishments relative to its schedule.

Interestingly, the creators of these influential metrics largely agree that *results-based* metrics are the most appropriate for selecting tournament teams. Ken Pomeroy, creator of KenPom, even stated his system shouldn’t be used for selection, a sentiment echoed by Torvik. Even the creator of the NET rating, Alok Pattani, conceded its limitations.

Despite this consensus, the public, and even the selection committee itself, appear heavily reliant on the NET. While the NCAA emphasizes the NET is a “sorting tool,” not the final say, its prominence overshadows other metrics. This raises concerns, as some argue the NET is susceptible to manipulation and doesn’t fully capture the nuances of a team’s performance throughout the season. Former Syracuse coach Jim Boeheim highlighted the flaw of prioritizing early-season success over late-season form.

The inclusion of WAB, created by a professional gambler, adds another layer to the complexity. This metric, designed to quantify a team’s performance relative to expected results based on its schedule, is viewed by some as a potential solution to the inherent biases in other systems.

While the selection committee remains tight-lipped about its precise methodology, the increasing reliance on data has transformed bracketology into a year-round phenomenon. However, the debate over the best approach to utilizing this data, and whether the current system adequately addresses the complexities of college basketball, persists. The upcoming tournament will be a crucial test of the efficacy of the current multi-metric approach, particularly regarding the committee’s ability to reconcile conflicting signals from different ranking systems.

Read More