I can’t even fathom how this is a real question, but it seems to come up all the time in this sport. If we aren’t going to judge teams based on actual accomplishments and instead just try to guess their “potential”, then I don’t know what the point of any of this is.
I don’t think Ken Pomeroy would appreciate your classifying his work as “guessing their potential”.
lol probably not, but I’ve been a paid subscriber to his site since the first day he put the pay-wall up so he can deal with my poor phrasing.
Obviously I think his work is great and useful and interesting, I’m just not sure what I think its role should be in the selection process.
Here’s another version of the same question.
Who should get the 36 at-large berths in the NCAA tournament? The NCAA awards conference champions (usually tournament) the guaranteed berths but what’s the rationale behind the other 36? The simple answer is the “best” 36 teams who haven’t gotten an automatic bid.
But that begs the question of how do you define “the best”? If by “the best”, you mean the ones who have had “the best” seasons to date, then the natural thing would be to start the whole process off by giving the first at-large bids to the regular season conference champions who didn’t win their conference tournament. But that’s not what the NCAA does. They are using some other criteria and many people would say they are giving the bids to the 36 teams who have “the best” chance at actually winning the tournament, or at least winning a game or two. So that drops you into the predictive measurement arena.
How would you determine the at-large bids?
Personally, I hate using the word “best” here, and I don’t know if that’s the official term they use for describing the at-large bids or not but I wish they would change it if it is. I think that term is where much of the ambiguity stems from regarding the goal of the selection process.
A better term would be the “most deserving”, or “most accomplished” even. Winning your regular season conference championship is an accomplishment for sure, something to be considered by the committee in my opinion, although it’s obviously not the same level of accomplishment across the board - winning the Northeast conference is generally not going to be as impressive as winning the Big East.
So how do we determine how impressive all these accomplishments are in order to sort this all out and rank these resumes? I actually don’t think the RPI was incredibly far off the mark. We should look at who all you beat and where you beat each of them (and possibly by how much you beat them). I personally wouldn’t focus so much on losses (but they probably can’t be ignored either), and I would put a lot more weight on road wins than home ones. Sounds a lot like the RPI I guess.
If you could devise a way to focus on these things without making the system so easily gamed, that would be ideal to me. Obviously it’s going to always have a lean in favor of power conferences just because they get more swings at the plate against good teams than smaller conferences do, so that’s another consideration that something like strength of schedule has to address.
The quandrants were/are a good start… I don’t know if a single metric can capture all of this in a meaningful way, but I want to know how many games you won, where you won them, how good were the opponents, and then how many similarly challenging situations did you face where you lost?
The NCAA website that defines the process that the selection committee is to follow states as its primary principle.
• The committee selects the 36 best teams not otherwise automatic qualifiers for their conference to fill the at-large berths.
So yes, they specifically say “best”.
Another article worth reading (from FiveThirtyEight)
Down near the bottom of the article, they (via Ken Pomeroy quotes) point out the inherent dilemma in this process.
In the past, the NCAA has sent conflicting messages about whether its selection and seeding process is fundamentally a forward-looking endeavor or a backward-looking one. “In terms of the how the committee should select teams, it actually says in [the NCAA guidelines] that they need to select the best 36 at-large teams — based on results,” Pomeroy said. “Best” and “based on results” don’t always line up, though. “It’s like a contradiction right in that sentence.”
Not necessarily a contradiction when the statement is viewed as a whole. “Best” is qualified “based on results.”
This suggests that among all teams that may be considered based on results (presumably more than 36), the committee must select what it considers to be the top 36.
Still leaves room for a wide variety of criteria but does not imply that anticipated performance is necessarily one of the criteria.
I’m not sure replacing “best” with “top” clears anything up.
Pomeroy and others, myself included, seem to feel that best (or top) implies a predictive measure. If you are going to decide who is best among a group it seems to go without saying that you need to establish a process that decides, when comparing two teams, which team is better. In my mind, which team is better implies which team would most likely beat the other in a head to head contest on a neutral court. That requires some sort of predictive analysis.
I’m with Craftsy in that “most deserving” or “most qualified” would be a better term if you really wanted to primarily use a results based approach.
I would agree with that.
NET settling in as maybe not so bad afterall.
Thought this was interesting, from the same Stadium article that Cappy linked to about the most disappointing hoops teams of the year so far:
NC STATE MANIPULATING THE NET?
I understand no one knew the formula for the NCAA’s new NET rankings prior to this season, but NC State has certainly taken advantage of the new metric by pounding the hell out of bad teams, and thus sending their efficiency numbers through the roof.
How else could you possibly explain the Wolfpack checking in at No. 30 with a resume that includes seven of their 14 wins coming at home against teams ranked in the bottom 50 of the entire country? UMES and UNC Asheville actually check in at 352 and 353 — the worst two teams in the entire nation — while NC State has also fattened up on Loyola, Md. (312), St. Peter’s (319), USC Upstate (325), Maine (331) and Mount St. Mary’s (342). I mean, Western Carolina (271) looks like a virtual powerhouse compared to some of the others.
The other six wins have come against Auburn (24), Pittsburgh (55) and Mercer (201) at home, Penn State (85) and Vanderbilt (78) on a neutral court and Miami (95) on the road. Hardly an overwhelming cast of victims.
Remember, the scoring margin is capped in the NET, but efficiency ratings are not. So this still gives teams an incentive to pound the you-know-what out of opponents. NC State’s average margin of victory against the eight “cupcakes” is 35.9 points per game, and the degree in which the Wolfpack blew out those teams is helping their efficiency numbers to a huge degree.
I’d be far more apt to buy into the KPI’s ranking of NC State (65) than the NET or even KenPom (29).