NET and other ranking systems question

[jleves]

Do they take into account overtime games and somehow discount the final margin?

For example, both UConn and Purdue won by 9 points last night, but Purdue needed OT to get that 9 point advantage. Do they treat OT games like a 1 point win or loss or do they get full credit for the difference?

 

[mbt0512]

Glad I’m not the only poster who’s wondered this and last night was a prime example of why..

NET no longer looks at scoring margin as an individual factor, but apparently did for the first two seasons 2018-19 & 2019-2020–during which it capped the margin at 10 for all games and treated OT games as if they were 1-point margins regardless:

The NCAA NET Rankings Explained

Diving into the new and updated version of the NCAA's primary rankings metric.

sonsofsaturday.com

The above link explains changes that took place beginning in 2020-21.. the efficiency factor (which remains one of the two underpinnings) would presumably still take into account OT data, if that makes sense.

KenPom does incorporate overtime margin in his algorithm, and below is an article on the rationale for doing and the logic seems quite sound, so I would assume most other sophisticated ratings systems do so:

https://kenpom.com/blog/the-simplest-proof-that-overtime-data-is-useful-and-that-garbage-time-exists/

Also some interesting graphs on how garbage minutes become statistically irrelevant, for the nerd crew. Not 100% certain on all of this, so anyone should feel free to chime in if inaccurate.

 

[BGesus4]

I haven’t studied the NETs methodology but the eye test says it doesn’t hold a candle to kenpom and torvik