I love this stuff yet it seems weird that if we are the #3 team in the best league that we have a 64% chance to make the tournament. Seems like it should be at least 80% but I am an idiot when it comes to this stuff.
2023-2024 MBB computer projections thread
- Thread starter Sigmapolis
- Start date
-
- Tags
- bracketology cyclones net quad rankings
That is odd, but most likely it happens like that early in the season and becomes more "logical" as more games are played -- it wouldn't surprise me if some other teams right now are projected to have a lower seed but higher chance for at-large. Not enough "data," maybe.
@Sigmapolis probably can explain how the entire algorithm is structured (and I think has done so multiple times since starting to compile/post these in recent years.)
With how many games we have left in that tough league I'm guessing that brings down the percentage since a lot of those games are probably close to 50/50
This is true, but it is impossible to predict and optimize given the way the tournament is thrown together by some dudes in a room the weekend before the tournament. All you can do is build the best resume that you can to (1.) ensure you get in and (2.) avoid the higher seeds as long as possible.
It's impossible to try and optimize your exact seeding line (say 3-4-5 depending on where the committee "needs" you) or your tipoff time or your exact opponent or where you're going to travel to play. All those factors are essentially random when the bracket is built in such a random, ad hoc manner.
Agreed -- sinking no lower than a #7 has a big impact on your survivability.
#8 and #9 always seem to have one upset over a #1 but 25% isn't great odds.
I would explain it like this...
The model has Iowa State ranked highly (#3 in the Big 12) but is being very "cautious" right now about how much better teams are than one another. We just don't have much data this early in the season to declare firmly that Iowa State is better than the rest of the non-KU/UH Big 12 slate of challengers.
So, Iowa State is ranked highly in a VERY tightly packed distribution of team quality. As that distribution spreads out as more games are played, especially between P6 foes, and more data comes in, then the projection for the NCAA tournament will either head up (if we really are the #3 team in the Big 12 then that should be 90%+ in a month) or down (if we are more like the #10 team, which we might if we play badly in Orlando and give up a bad "oopsie" game to DePaul or Iowa, they we'd be at <50% or so to make the tournament).
So, our ranking is high, but it's not robust. A loss would drop us like a rock right now.
If we were ranked that highly in late January, then it would be robust. A loss wouldn't change our outlook all that much. The swing of one W/L isn't much at that point in the face of more sample size.
Does that help?
Iowa State moved up to 19th in KenPom after today’s game.
AdjO is 59th — Up from 65th
AdjD is 3rd
Iowa State is projected to finish 20-9 (10-8)
**Currently 2nd in TO%
AdjO is 59th — Up from 65th
AdjD is 3rd
Iowa State is projected to finish 20-9 (10-8)
**Currently 2nd in TO%
Our offense is just skyrocketing. We obviously have a lot more scoring this year and the staff seems to be using it properly. Didn't even have a good performance from 3 today and still moved up
As an add on to my previous comment it's nice that when the 3 isn't falling this team can score other ways so they don't just have to keep chucking. With Gabe and Grill last year that was kind of the offense
Started the year at 80, I believe.
Also, there is a ton a bias towards last year in these ratings, still. That tells me our offense has been very good
Last edited:
Trending in the right direction but of course it is still very early...
#3 in the Big 12 and #10 in the country jumps out at you.
Long way to go to prove those ratings are remotely accurate, though.
#3 in the Big 12 and #10 in the country jumps out at you.
Long way to go to prove those ratings are remotely accurate, though.
Things will get tougher and they'll have some weird games or stretches of games like every team but hopefully being strong on both ends can help balance things out to come out a tourney bid.
I really want to see how it goes against even a 'bad' P6 team.
They went 4/13 against Idaho St. from three.
Two observations after the typical throat clearing about Idaho St. being terrible...
-- not a great hit rate (30.8%) but at a low volume... they backed off when they weren't falling
-- found other ways to score... last year they would have kept on chucking for lack of other options
Hopefully that carries over into games against more rugged opponents.
Torrvik allows you to remove preseason bias.
*Disclaimer: 2 to 3 game sample sizes are extremely volatile. For example UMass Lowell is the #3 team in the country when you do this.*
Iowa State's week 1 numbers...
Overall: #6
Adj Offense: #49
Adj Defense: #7
I seem to remember from Idaho State's stats that they had defended pretty well from three in their first two games. Going in I thought that was a good sign because we don't shoot it a ton out there and that would spread them out a bit for our drivers.
These computer numbers are great so far. Love to see it! I was worried because we have the weakest nonconference schedule I can remember in some time that the computer's would not be a fan of us and bury us toward the bottom before starting conference play. Will be interested to see the NET rankings as well. Have those been released yet or do they wait to get more data / games?
Hope they can go undefeated in the noncon this year. With our weak schedule (except for the possibility of good games at the ESPN tournament), we could use it. Hope they win the first one at the ESPN tournament and get in the winner's side of the bracket for some quality games on a neutral court.
Hope they can go undefeated in the noncon this year. With our weak schedule (except for the possibility of good games at the ESPN tournament), we could use it. Hope they win the first one at the ESPN tournament and get in the winner's side of the bracket for some quality games on a neutral court.
I think our NET is going to be a bit weak from the non-conference, but we'll have tons of opportunities to improve that dramatically
Texas A&M and FAU are the most significant impediments to an undefeated non-con.
TAMU beat Ohio St. a few nights ago and won convincingly.
FAU has only played one game (defeated Loyola-Chicago).
Very little data on them so far (or any team really) but that would be a "proved it" game.
Some bonkers numbers in there, mostly due to the competition, but still
the >2.0 Steal to Turnover ratio AS A TEAM is insane. They've taken care of the ball really well for how fast they've played - but I do expect that to change - Gilbert and Jones are a bit loose with the ball at times. While the steal rate won't stay > 20%, I expect they'll stay in the top 10.
The other crazy thing is that opponents are taking a huge # of 3s and we are taking very few. I expect we'll take more as competition heats up - but opponents will continue to take a ton
I was thinking that too - but looking at the numbers we really didn't shoot many 3s last year - averaging < 20 attempts per game for the first time since pre-Fred.
Iowa State is doing a GREAT job of beating the crap out of these bad teams. Your computer numbers can really tank if you don't best these teams with ease.. If they can keep this up against the 5 teams left in the non-con that are rated 250 or worse by Torvik/KenPon/pick your system.. They have a chance make the bad scheduling not hurt seed line come March.
I was thinking about this this weekend. As hard as some of the offensive performances have been the last couple years I think it will be a huge long term boost to the program. It forced TJ to really lean into the defense and effort part and learn new coaching skills there. And it has now laid a foundation of defense and effort that has proven to win games and now he can bring the offense up to speed. Which is probably the easier of the 2 to do.
