What could be different about this year is the potential for more Q1-2 opportunities and a lot less risk of piling on Q3 losses. For NeT purposes, we’re far better off playing mediocre SHU and WF teams on the road win or lose. I also think Princeton will be better for our NET than Temple was. Princeton is going to have a nice looking record. They are off to a much better start than last year.
Princeton will likely be ranked…EDIT…they still aren’t
- Thread starter mikemarc1
- Start date
What could be different about this year is the potential for more Q1-2 opportunities and a lot less risk of piling on Q3 losses. For NeT purposes, we’re far better off playing mediocre SHU and WF teams on the road win or lose. I also think Princeton will be better for our NET than Temple was. Princeton is going to have a nice looking record. They are off to a much better start than last year.
Princeton 34 so massive fail by a couple of people.
Having an opinion that’s wrong on a message board is a massive fail???
Lots of massive failures by everyone if that’s the case.
I was wrong, not a big deal at all. Thought they’d be higher, oh well
yeah but Florida Atlantic lost to Bryant and are #13. That was a big time performance for Nova there
Maybe - but some have them penciled in for Q3 and that seems off too. I’m not giving up on them winning their way to a top 50 NET by blowing the snot out of bad opponents Iowa style. We’ll see what happens but I don’t think Q1 is out of the question, let alone Q3 being a foregone conclusion at this point.
FWIW we may have a big influence over if Princeton can rise to a top 100 net... by having a great season, making their win over us look like a great win for them and not a bad loss for us. Of course, in that case, we shouldn't really need to be worried about the loss.
There's a chance that Princeton could end up a Q2 game if they finish with 3 or fewer losses. They'd need help some of their non-conference opponents, who would need to have strong seasons, and they need Yale to be strong. They also have to stay healthy, as almost 85% of their minutes come from just 5 guys. Even undefeated I don't think Q1 would be possible.
I'd still say there's a better than 80% chance they end up in Q3.
I'd still say there's a better than 80% chance they end up in Q3.
Well, it makes sense the Ivy League never had an at-large bid. It would be odd if they ever did for all those years
Why? It was a neutral game. They were NET 111 with 8 losses last year and you don’t think they’d easily be in the top 100 with only 3 losses? I get that they play a weak schedule but perspective, no team outside of the NET 100 had less than 8 losses last year.
Nice! There's always a little more juice in the building when the opponent is ranked.
Rutgers has won for 4-out-of-the-last-5 vs. Illinois in Piscataway.
Last edited:
Great opportunity to steal one early
The loss to UPenn still hurts Nova despite the more recent win over UNC.
The RAC has been a house of horrors for us. In fact, neither of our teams have fared well on the road in this series. We have lost 3 in a row and 4 of our last 5 at the RAC, and Rutgers is still winless (0-7) all-time in Champaign.
It should be noted that that was after the tournament. Before they would’ve been a Q3 loss. Now this year they’ve done much better in the non-con but it’ll be close.
How do you figure that? Last year they lost to Hofstra and Navy to open the season, this year they beat RU. Change nothing else in Ws and Ls for the rest of the two seasons and this year’s team would clearly have a better NET than last year’s.
And early indications are that this year’s team is better than last year’s team that lost 8 regular season.
Half the top 25 have resumes like that this early in the year. The reason those teams are ranked and PU isn’t is just reputation and preconceived expectations. It’s not like Princeton went to the sweet 16 last year or something.
Rutgers
Hofstra
Duquesne
Monmouth
Old Dominion
Northeastern
Princeton hasn't finished in the NET 100 since it began. Last year, their SOS was 264th and their noncon SOS as 331st - this year we were the only competitive non-conference team on their slate. They do not play a schedule that lends itself to the NET algorithm, because they are in a one-bid league and NET ranking doesn't really matter to them - they really need their Ivy brethren to have good non-conference performance to elevate the entire conference's SOS.
If they finish with just 3 losses, they'll probably end up in the top 100. That will require them to stay healthy all season (they have very little depth) and not to have any off shooting nights against bad teams that get hot. I give that about a 20% chance.
It's hard for teams from one-bid conferences that schedule weak OOC to get to double-digits.
So I don’t know when exactly NET started but your basis of Princeton never being top 100 doesn’t really make sense as a predictor for where their NET would be this year with 3 losses because they had never had close to that few losses while NET was around.
In 2022 Princeton had 6 losses (double the 3 losses your suggesting) and was very close to top 100. Princeton also finished with only 6 losses in 2016-17 and were Ken Pom 58 that year so I’m pretty sure they would have been NET top 100 if it had existed. I don’t know how to look up historical NET data but Id guess very few, if any, teams with only 3 losses end up outside of the top 100.
Kenpom and NET are very different from one another, teams are frequently 30-50 ranks different between the two. This is also assuming Princeton wins every game they're supposed to win from here on out, which is threading a needle.
Kenpom is meant as a predictor of future performance, and is designed to rank a team's strength. NET is meant as an evaluation tool for a team's past resume.
If Princeton only plays four games against teams in the NET top 110, and loses three of them, that works against them in the algorithm. Rutgers is also their signature win, so if we don't perform and slip down to the high double digits, that would also work against them.
A lot has to go right for a team with that weak of a schedule to get into the top 100 - including staying healthy and actually playing the games, rather than just penciling them in for only three losses (barttorvik projects six losses, fwiw). I'm not assuming every ball bounces their way.
But again, Princeton really shouldn't care what their net is, because their only path into the NCAA tournament is by winning their conference tournament. That number matters much more to us than to them, because our most likely path to the dance is on the strength of our resume.
I only used Ken Pom because I couldn’t find historical NET data. But in the year I’m talking about Princeton clogged in 47 on RPI too. I highly doubt that team wouldn’t have been top 100 in NET and that was with 6 losses, not 3. The NET formula isn’t transparent, but I doubt very many teams with less than 5 losses ever finish outside the top 100 unless they win a ton of really close games inefficiently. IMO Lee looked too good for that to be the outcome with Princeton.
FWIW - ESPN’s BPI has Princeton favored to win all of their remaining games except the one at Yale.
That's before NET was a thing, but there's a decent chance that season would have been better than 100. They played a slightly tougher slate than usual that year, which would have increased their SOS with the algorithm. They got a little help from Monmouth and Bucknell having exceptional seasons.
Looking at barttorvik, these are the top 5 OOC teams they played in 2016-17 (they finished at #58, including a 2-pt tournament loss to #21 Notre Dame):
61 - @VCU (26-9) (2nd place A-10)
67 - (N) California (21-13)
77 - @Bucknell (26-9) (1st place Patriot)
82 - @BYU (22-12)
87 - @Monmouth (27-7) (1st place MAAC)
And this year (early rankings, subject to change once conference play starts)
56 - (N) Rutgers
60 - @Duquesne
112 - @St. Joe's
114 - @Hofstra
118 - Drexel
Last edited:
Net's been around since the 2018-19 season. These were Princeton's rankings on Selection Sunday each of the last five seasons.
2022-23: 111
2021-22: 107
2020-21: n/a
2019-20: 166
2018-19: 180
As a comparison, here's the bart ranks for recent years heading into the postseason:
2022-23: 112
2021-22: 112
2020-21: n/a
2019-20: 139
2018-19: 170
2017-18: 177
2016-17: 59 (5 games against Top 100 opponents)
2015-16: 62 (4 games against Top 50 opponents)
2014-15: 149
Last edited:
NET was changed a couple years back and now tracks much closer to something like Kenpom. It's not identical but if they actually finished at #53 in Kenpom I would bet a lot of money they would be in the top 100 in NET.
Last year when Princeton finished #93 in the NET (post NCAAs) they finished #91 in Kenpom.
The maximum difference between Kenpom and NET for a team last season was 45 spots (Cal was ranked #315 in NET, #270 in Kenpom). The average difference was only 6.9 spots.
Furthermore, strength of schedule is actually positively correlated with <NET Ranking - Kenpom Ranking>. In other words, teams with worse SOS actually tend to outperform their Kenpom ranking while teams with better SOS underperform it.
Last year when Princeton finished #93 in the NET (post NCAAs) they finished #91 in Kenpom.
The maximum difference between Kenpom and NET for a team last season was 45 spots (Cal was ranked #315 in NET, #270 in Kenpom). The average difference was only 6.9 spots.
Furthermore, strength of schedule is actually positively correlated with <NET Ranking - Kenpom Ranking>. In other words, teams with worse SOS actually tend to outperform their Kenpom ranking while teams with better SOS underperform it.
Great post, @fluoxetine - we still see some weird outliers, but there is a much stronger kenpom/NET relationship than there had been before. The outperformance vs. SOS is a more moderate relationship, but does show slightly negative - very interesting.
A couple of observations:
First, kenpom's 2023 data includes all the postseason games - so a school like UConn ended up at #1 in kenpom and #3 in NET, but they were #8 in NET on Selection Sunday. I'm assuming this comparison was post-tournament for both ranking systems - which gives a little more accuracy, as post-season teams will have more games in the data set. I wonder whether the relationship would be at all weaker if the pre-tournament rankings were used.
Second, looking at the SOS rankings on kenpom's site, 17 of the 24 schools with greater than +10 SOS were also in the Kenpom Top 50 - which also makes it harder to outperform (fewer spaces above you) than underperform (for example, UConn was the kenpom #1... making it impossible for them to outperform in NET). Similarly, 6 of the 9 schools with below -10 SOS were ranked 300+, which makes it harder to underperform.
A couple of observations:
First, kenpom's 2023 data includes all the postseason games - so a school like UConn ended up at #1 in kenpom and #3 in NET, but they were #8 in NET on Selection Sunday. I'm assuming this comparison was post-tournament for both ranking systems - which gives a little more accuracy, as post-season teams will have more games in the data set. I wonder whether the relationship would be at all weaker if the pre-tournament rankings were used.
Second, looking at the SOS rankings on kenpom's site, 17 of the 24 schools with greater than +10 SOS were also in the Kenpom Top 50 - which also makes it harder to outperform (fewer spaces above you) than underperform (for example, UConn was the kenpom #1... making it impossible for them to outperform in NET). Similarly, 6 of the 9 schools with below -10 SOS were ranked 300+, which makes it harder to underperform.
Yes, I used the post tourney numbers for both as that is all I have access to for Kenpom. Both of your points I think are valid. I wish the NET released their actual rating number in addition to the ranking as it would make it easier to directly compare; 20 ranking spots means a lot less in the middle of the distribution than it does towards the top or bottom.
Sure okay. You’d clearly have a point if you were saying there’s a good chance Princeton would slip to Q3 if they end up dropping 6+ losses (like in every one of these other years). But how can you correlate this on a predictive basis to what Princeton’s 2024 NET would look like with only 3 losses without also at least considering the gap in the loss count in these other seasons?
I’m not saying SOS isn’t a factor too (of course it is), but my point is the 2019-20 NET of 166 (in a year Princeton was 14-13) means nothing on a relative basis to 2024. Very few teams historically (if any) have ever finished with 3 or less losses outside the top 100 in the metrics - at least in my memory.
None of this is to say Princeton can’t suffer from the injury bug or go on an unexpected losing streak but that wasn’t the point raised. Princeton’s NET wouldn’t need “help from us and other sources” to clear 100 at 26-3. It would be the other way around. They’d need it to be the case where, even moreso than in 2021-22, they pile up single digit wins over Q4s and probably take 2 double digit losses too. Even still, they’d probably be top 100 with a record like that.
Last edited:
I'm not saying that I think there's a better than 80% chance that they are Q3 with only three losses, I'm just saying I think there's a better than 80% chance that they are Q3 in general. Running the table of every game you are "supposed to win" isn't easy, especially when you have a six-man rotation.
If they manage to lose only three games, then there's a very good chance they will be Q2. But there are a lot of games left to play.
I'd love for them to have a miracle undefeated season and make our loss look better, but I wouldn't put any money on that.
this thread will be 10 pages by March 1
Latest posts
-
TKR LIVE SHOW -- Was this the worst weekend in RU Athletics History?- Latest: BIGRUBIGDBIGredmachine
-
-
-
