If my aunt had balls she’d be my uncle. That’s so silly… if this team did this and if that team did that and if the shot that was missed was made…
One game can be an outlier. Three teams from the “weakest” conference making the elite 8 and 2 making the final four is not an outlier. The model was bad.
You didn’t mention ND beating Alabama either… but if the players woke up on the other side of the bed then maybe the ACC wouldn’t have won every matchup in the tournament against the SEC…
But maybe you’re right. Maybe kenpom had us pegged as a sub-50 team and maybe we were just lucky to paste Auburn before another double digit win in the sweet 16.
Or maybe kenpom is a model that has some predictive value but is a poor ranking system and fails to account for important parameters/variables including actual wins and losses.
Yes, and here's the thing.
I usually don't criticize predictive MODELS based solely on the model. There is usually a problem with DATA as well, which is why I have pointed out the circular self-fulfilling prophecy of being labeled a "good team" right out of the box. So I won't belabor THAT issue, for now.
Then you have the voters and the polls. And they do NOT have to be "the same". Voters can see things with their own eyes and vote accordingly. BUT, if they walk into their voting booths with biases, like "UNC and Creighton and Gonzaga are really really good this year", then it is possible to taint the eyeball test too. This is where I have tended to rip the voters (and some of our pollster-ball-licking porsters) who come here to tell me how Miami's ONE LOSS to a #13 ranked Maryland (yes, by 18 points, but YES, on the second game of a neutral site tournament) should be held against us SO MUCH, while I can hear people spinning reasons why "Good Team X" was just sooooo unfortunate to lose 2 or 3 games to "other good teams".
I would also add something new. We REEEEEEALLY need to consider getting rid of "polls" the first month to six weeks of a season. So, for football - first poll comes out in October. For hoops, January. Why? Because the game has changed, we have SUCH a huge impact of "Portal" that everything has become a crap shoot. An "unranked" team can be a contender simply because of the impact of transfers. A "historically good" team can completely MISS in the Portal (or even recrruiting, cough cough aTm Football) and lose way more games than expected. So how about we give it a rest for a stretch. Will we really suffer if we do NOT have rankings for a few weeks? Particularly when I read this thread and see so many people trying to tell me that rankings don't matter, and only the post-season matters?
So if that is so, then get rid of the early rankings. Why should we keep svcking teams' d!cks for beating North Carolina or Creighton when it now looks that they aren't as good as we thought they were?
Bottom line, setting aside all of the circular and reductive arguments - there is an argument that says that a prior year Elite 8 team that snagged two of the most highly sought-after Portal transfers...is going to be a pretty good team this year, particularly when they have multiple core players and/or leaders returning. That's "eye-test". No stats or computers models. Just common sense.
I don't need AP and kenpom to agree on everything, but I'm tired of the concept where we are not looking at THIS YEAR and maybe, just maybe, LAST YEAR to determine how good we expect a school to be. I'm not trying to ****e on North Carolina's decades of success, but that doesn't mean they will always be good. At a certain point, wins count. AND I DON'T CARE ABOUT THE MARGIN. Short benches (as Miami played with last year and has only started lengthening in the last couple of games) tend to create more close wins and more "big" blowouts, if you keep trying to play the same 6 or 7 guys for 30 or 35 minutes per game.
