The bigger question is which of these games stops the patented NU BIG losing streak.
When do we get excited?
- Thread starter SmellyCat
- Start date
I like our chances @ IA at better than 26%. Need to be really dialed in on defense. Good chance they play zone on us. Maybe more of a 40% chance.
If WI has Wahl back, no idea if there’s a timeframe on that, I feel it’s more of a 50/50 game
Nebraska I feel a 50/50
MN is getting better, feel more of a 70 on that one.
These ESPN things are based on stats and those are for losers
Last edited:
Iowa is really hot right now. The Cats better bring their best defense if they want to win.
To be fair, Maryland is complete ass on the road and Iowa would have lost to Michigan if the Michigan guy didn't gift them OT on the 4 point play. But yes, Iowa has definitely been playing better since McCaffery took his break for mental health.
They’ve won 4 B1G games in a row including Rutgers on the road since McCaffery went out. Winning there this week will be extremely difficult.
In 74 of the 84 brackets in the Jan 27 update. The consensus is a 10 seed.
Clark Kellogg just picked us to win the NBA championship.
Wow. We are in all 88 brackets in the Feb 1st update. A 9 seed. Go Cats!
In 99 of 101 brackets pre-Purdue in the Feb 11 update. Still a 9 seed.
He has 2 years left on his contract after this season.
I think it is safe to say he won't be fired in 2023.
He is coaching like his job is on the line for a reason. And the pressure is turning what we thought would be coal into a diamond of a season. 10 year contracts need to go extinct.
I’m more excited that Jared from Subway at a cheerleading competition
And Kentucky as an 11 seed... Who would've predicted us playing them first round as the (significantly) higher seed? Also, UNC projected as missing the tourney would suck for Pete.
Last edited:
They can have the spot NU won't use in the NIT.
I am a bit surprised we didn’t drop, but still a 6 seed post PSU.
Because the committee focuses on what has happened this season, not just on predictive metrics. Our strength of record and wins above bubble are much better.
Northwestern Wildcats NET - Nitty Gritty Report & Team Sheet 2023 Men's College Basketball | WarrenNolan.com
NET - Nitty Gritty Report & Team Sheet for Northwestern Wildcats 2023 Men's College Basketball. This Is a Duplication of the report used by the NCAA Tournament Selection Committee to Determine At-Large Teams in the national championship tournament with quadrants
I see Wisconsin dropped a quad for both of their games with NU. They need to up their NET rating.Because the committee focuses on what has happened this season, not just on predictive metrics. Our strength of record and wins above bubble are much better.
Northwestern Wildcats NET - Nitty Gritty Report & Team Sheet 2023 Men's College Basketball | WarrenNolan.com
NET - Nitty Gritty Report & Team Sheet for Northwestern Wildcats 2023 Men's College Basketball. This Is a Duplication of the report used by the NCAA Tournament Selection Committee to Determine At-Large Teams in the national championship tournament with quadrantswww.warrennolan.com
Because the committee focuses on what has happened this season, not just on predictive metrics. Our strength of record and wins above bubble are much better.
Northwestern Wildcats NET - Nitty Gritty Report & Team Sheet 2023 Men's College Basketball | WarrenNolan.com
NET - Nitty Gritty Report & Team Sheet for Northwestern Wildcats 2023 Men's College Basketball. This Is a Duplication of the report used by the NCAA Tournament Selection Committee to Determine At-Large Teams in the national championship tournament with quadrants
Not one of those systems is predictive, as far as I know.
They are 100% based on the games already played this season.
Not sure where you are getting the "predictive" part.
The NET puts more focus on wins vs losses. Torvik, Sagarin, Pomeroy, Miya all base their ratings on scoring differences. A 1 point road loss is a lot better than a 20 point road loss.
I'll side with the stats geniuses.
I am going to just speculate the reason the selection is not done more based on stat geniuses' models is because it would be based on something the committee does not understand.
As is, Quads, SOS, RPI, the average Joe does not understand it anyway. But the silver haired old men do.
Torvik and KenPom are predictive in that they analyze team performance on both offense and defense, project the game result and then adjust the ratings based on that game's performance. The ratings then set an expected margin of victory for the upcoming games. Torvik uses KenPom's data and amps up the 10 most recent games. These are not perfect but they do their best to predict future outcomes. And they are fun to look at.
Addendum: Ken Pomeroy himself calls his system predictive.
"The first thing you should know about this system is that it is designed to be purely predictive. If you’re looking for a system that rates teams on how “good” their season has been, you’ve come to the wrong place. There are enough systems out there that rank teams based on what is “good” by just about any definition you can think of. So I’d encourage you to google college basketball ratings or even try the opinion polls for something that is more your style.
The purpose of this system is to show how strong a team would be if it played tonight, independent of injuries or emotional factors. Since nobody can see every team play all (or even most) of their games, this system is designed to give you a snapshot of a team’s current level of play."
Last edited:
Thanks for posting something meaningful, helpful and sincere.
I looked it up and found the article from which you extracted the quote - its from 2006, originally, with modifications. I read through it and refreshed my memory on what Bill James used to do in the early early days of "sabermetrics." Hard to believe that Ken Pomeroy is a James disciple, of sorts.
Anyhow, it looks quite possible that Pomeroy has tweaked his approach over the years, but maybe you know with certainty what he is doing now, calculation-wise.
While his calcs are more involved than Sagarin's basic system it still seems to me that Ken Pom is doing the following...
1. calculating offensive efficiency (points scored per possession)
2. calculating defensive efficiency (points allowed per possession)
3. adjusting those for the quality (efficiencies) of the opponent
4. adjusting for home court advantage
5. weighting recent games more heavily.
So if every team has a known number of total points scored, total points allowed and total possessions, you can take the average and get a starting point for each team's Offensive and Defensive Efficiencies and essentially run thru all the games and tweak them or solve for them, given the quality of its opponents and the scores.
But it is still based entirely on the scores of the games that have already been played. The primary difference between Ken Pom and Sagarin is that Ken Pom rates both the offense and defense for each team, while Sagarin just rates the team. So Sagarin could "predict" that NU should beat Penn State by 4, whereas Ken Pom can "predict" that NU should beat PSU 66-62.
I have built systems like Sagarin and Ken Pom. I don't think of them as predictive. I think of them as methods to rate teams, based on their past performances. Sure, you can give a somewhat informed guess as to what the final score might be, but the resulting differences between actual and predicted scores have to be relatively large.
To me, "Predictive Systems" tell you the answers to complicated things like "what's the weather going to be like in a week?" or "at what price is NFLX going to be trading in an hour?"
I think KenPom might have written "I built this system to try to set point spreads, money lines and over/unders."
I mean, you can think what you want, but the point is that Kenpom, sagarin, torvik take results of games and use them to *predict* what will happen in future games. Teams are ranked based on how likely they are to beat a team the next time they play. They are often quite good at doing so. But when you have a team like West Virginia that has been a top 25ish team in these systems all year despite having a record at one point of 15-11, 4-9, and a team like Missouri that was 19-7, 7-6 at the same time but ranked 53, at some point you have to look at what happened on the court.
The KPI and SOR ranks are only looking at who you beat, rewarding teams for *who they have beaten*. They are not predictive. West Virginia has a strength of record of 35, while Missouri is 17. These are results-based metrics that care about whether you beat good teams or not regardless of how you did it.
Yep, another way to look at it, West Virginia would probably be a 3 point favorite over Mizzou if they played on a neutral court tomorrow. But Mizzou 100% deserves to go the tourney and WV doesn't.
Hey if the NCAA Committee said something like "You have to finish in the upper half of your conference AND have a winning record in conference play to make the NCAA field," I'd agree completely. Completely!
But they don't say that.
So West Virginia is better than Mizzou and happens to play in the best conference, so they have more losses.
Not sure where they draw the lines.
On the other hand, there's a team like VCU who deserves to be in the field and won't make it if they don't win their conference tournament.
The other thing he does prior to every year is re-set the teams based on personnel changes. Not sure how first year players are dealt with, but each player's FG and FT percentages, steal/block/assist/turnover rates, etc. are used to update the aggregate team strength. This is done bottom-up and is especially important due to the portal. Then as the season unfolds the actual game performance informs predictions for the future games. I don't think players' individual perfomances impact the team rating intra-year.
This methodology has gaps - some that are impossible to fill. Team chemistry and ability to close are two notable ones. This year, as opposed to prior years, the Cats excel in these areas. Even with the Illinois and Penn State games.
I have not studied Sagarin in recent years. My biggest issue with Sagarin is the teams come into the season with a rating, presumably based on the prior year's results. This perpetuates top conferences remaining at the top and IMO, an inaccurate estimate of the team's quality because rosters change so much, even though the team name stays the same.
KenPom and BartTorvik are understandable and robust predictive tools. They are fun to track over the season, especially when the Cats are playing well. They may not be as predictive as some would like, but that is what makes the games compelling to watch, especially this year.
Go Cats!
Evan Miya does something similar with individual players. He therefore can estimate the quality of the team once a valuable player, such as Mawot Mag of Rutgers, is lost for the season.
