Reply to thread

Message: [QUOTE="swish7, post: 4975992, member: 486"] Interesting. I don't pay much attention to this stuff, and just did what made sense to me. This is what the NCAA lists on their website as to how these rankings work. The link says the info is from 2021; not sure if it is current. The explanation of 'how it works' is a crap ton of fluff, and only 2 sentences that add any value to the explanation, and that value is minimal at best. [LIST] [*]The NET rankings are used for the NCAA tournament. The rankings are released every day in an effort to be transparent but the NET rankings in January carry little significance during the selection process in March. [*]The NET rankings do not include any preseason data, as many other popular metrics like KenPom and BPI do. [*]Today (Jan. 4) is the 40th day of the season. In a typical year, teams will have played 10 to 12 games 40 days into the season. The average number of games played is slightly lower this year, and several teams have played fewer than five games thus far. [*]Despite the unusual scheduling circumstances, the NET rankings are largely aligned with other metrics that the selection committee tracks. There are certainly outliers involving teams that have played very few games, but as more data becomes available with the addition of more games, the more stable the NET rankings will become. [*]The NCAA entertained the idea of not including teams that have yet to play five Division I opponents but ultimately decided to include everyone, recognizing that all teams will soon reach that threshold. [*]As always, the NET rankings do not include games played against non-Division I opponents. [*]After two seasons of DI men's basketball using the NET, it was tweaked in advance of this season, simplifying it to a two-component system. The first is the [B][U]Team Value Index[/U][/B], a results-based feature that [B][I]rewards teams for beating quality opponents, particularly away from home[/I][/B]. This component was an original part of the NET. The other is an [B][U]efficiency rating[/U][/B], which is [B][I]adjusted for strength of opponent and location (home/away/neutral) across all games played[/I][/B]. [*]With teams playing home games with limited or no fans, the committee and staff is studying the impact that’s having on home-court advantage compared to a typical year. [*]While both the men’s and women’s NET rankings share high-level goals and individual components, the NET algorithm used in each is different. [*]The machine learning model developed for each sport utilizes only that sport’s data. The women’s model uses only women’s game data, while the men’s model only uses men’s game data. [*]The weights for each are trained using the historical data from the respective sports. [/LIST] [/QUOTE]

Verification: What is a Syracuse fan's favorite color?