Aike’s Model - 2/21/25 Update

[yabbadabbadoo]

How does this model and other models take injuries into consideration???

 

[Aike]

One thing tho I do think maybe metrics should factor in consistency more.

So the predictions are always based on what a team does "on average". On average we score this many points per poss. We give up this many points per possession.

I kind of wish they were factoring in the variance between games.

For example, say Team A plays 50 games. They score 1.2 ppp on average. Every game they score 1.2 for that average.
Team B plays 50 games. They also average 1.2 ppp on average. However, 25 of those games they scored 1.00 and the other 25 they scored 1.40.

I'd make the argument that Team A is more likely to win in a one and done tournament compared to Team B, despite them both avging the exact same thing. You score 1.00 in any game, you are probably going home. You score 1.20, you are probably moving on.

So I feel like consistency might be a good thing to look at. Obviously tho, you'd have to do it factoring in the schedule as well. If you are playing a lot of scrubs, it stands to reason you are going to have wild swings (put up a ton vs those teams and normal scoring when playing a decent team). Still, I feel it's something that might be getting overlooked a bit.

It’s kind of what I was talking about above when I mentioned calculating a range of scores.

But instead of the fairly simple examples that we both offered, consider that the scores are not normally distributed.

Consider a team with an average score of 3, who 60% of the time is a 4, 10% of the time ie a 3, and the other 30% of the time is a 1.

Now imagine using the distribution of each team’s scores to run 10000 simulations with the actual bracket. I think this could give you a more accurate prediction of how the actual tournament would play out (though it would still likely be only about 80% accurate at best).

 

[Aike]

How does this model and other models take injuries into consideration???

Mine does not specifically take injuries into account. That’s a great example or where your basketball knowledge would need to come in.

That being said, the ability to continue to perform at a high level despite injuries is certainly helpful if you want to advance in the tournament.

 

[Digger-Cat]

UK is probably the most difficult team to seed. We have incredible wins, but for a team with such great wins, we have some head scratching losses.
There are 4 teams on the bubble that have UK as their best win (OSU, Texas, Arkansas, UGA) and Vanderbilt is close. Hopefully we don't add Oklahoma to that list

 

[The_Answer1313]

UK is probably the most difficult team to seed. We have incredible wins, but for a team with such great wins, we have some head scratching losses.
There are 4 teams on the bubble that have UK as their best win (OSU, Texas, Arkansas, UGA) and Vanderbilt is close. Hopefully we don't add Oklahoma to that list

I feel like the teams that are most difficult to seed are the ones that haven't played anyone. The mid majors.

Because if a team has like a ton of Q1 wins, you know they are good. If a team is like 3-10 in Q1 teams you know they aren't.

It's the teams that have only played a few big games that people tend to have no idea on.

Sometimes I go on the Bracket Matrix website and I look at the teams where people disagree the most about seeding. It's ALWAYS the mid major teams. No one really has any idea where to place these teams.

It's where judging things on resume breaks down. Everyone knows a good resume, a bad resume. What if a team has no resume. What you do with those teams is where it's confusing to ppl. And I think one of the key ways computer metrics can be used.

 

[docholiday51]

We do tend to rise to the occasion.

If we just didn't stoop to the level of some lesser teams(Texas and UGA) although the UGA loss was to the officials it still counts against us. Still pissed about that one.

 

[The_Answer1313]

If we just didn't stoop to the level of some lesser teams(Texas and UGA) although the UGA loss was to the officials it still counts against us. Still pissed about that one.

I feel like some of it was at least based on injuries tho.

That Texas game was right there to be won. Obviously can't make excuses you have to play with who's available but a health Robinson/Butler we win that one.

 

[TucsonCat]

Time and again, advanced metrics do a better job of predicting results than the “eye test.”

I'm sure they're great in predicting NCA&T vs Central Michigan.
Ut I'm afraid I disagree when it comes to important games.
You mentioned a few weeks back how your model and Kenpom did in predicting the ncaa tournament.
But the chalk made by the selection committee was better than both.

 

[Aike]

I'm sure they're great in predicting NCA&T vs Central Michigan.
Ut I'm afraid I disagree when it comes to important games.
You mentioned a few weeks back how your model and Kenpom did in predicting the ncaa tournament.
But the chalk made by the selection committee was better than both.

There might have been years where it was better. The committee has actually gotten much better at seeding the tournament because they are relying more on metrics like this. It has not historically done a better job.

I’ll give you just one recent example. In 2023 I had UConn 4th overall entering the tournament. Committee had them as a 4 seed.

In that same tournament I had Purdue 15th overall. They became the second 1 seed to lose in round 1.

At any rate, my model is predicting how many games someone is likely to win. The committee is seeding according to what they think the team “deserves” based on resume. So there will always be differences.

But certainly feel free to ignore all of this. Fill out your bracket with all chalk. Sounds like a ton of fun.

 

[Aike]

If we just didn't stoop to the level of some lesser teams(Texas and UGA) although the UGA loss was to the officials it still counts against us. Still pissed about that one.

We’ve lost road games like that since the dawn of time. Almost every team does every year.

The Texas one bothered me more because we frittered it away. But I actually went into that one expecting to lose. Just hated blowing it.

Off topic, what was the name of that restaurant you like in Burkesville? And do you happen to know the name of the owners?

 

[Aike]

Tennessee looked pretty good today, btw. Although I don’t have A&M rated nearly as highly as many, so it kind of evens out.

 

[docholiday51]

We’ve lost road games like that since the dawn of time. Almost every team does every year.

The Texas one bothered me more because we frittered it away. But I actually went into that one expecting to lose. Just hated blowing it.

Off topic, what was the name of that restaurant you like in Burkesville? And do you happen to know the name of the owners?

The bait shop, Tammy and Earnest Anderson(great folks)

 

[MadaboutBlue]

Thank you for doing this. It's always one of my very favorite things on this board!

I do think you have Tennessee is too high. They just don't look consistent enough on offense. Florida looks great for the time being...I thought at the time their win over Auburn was a fluke but i'm not as sure now.

 

[Montana81]

If we just didn't stoop to the level of some lesser teams(Texas and UGA) although the UGA loss was to the officials it still counts against us. Still pissed about that one.

That was one of the worst whistles weve ever gotten. They really just let Georgia play by a different set of rules. The free throw disparity at the end was almost completely opposite of what it should have been.