Untitled Document
Brown Columbia Cornell Dartmouth Harvard Penn Princeton Yale



 Page 8 of 9 « First<6789
Username Post: UCF        (Topic#19618)
SRP 
Postdoc
Posts: 4894

Reg: 02-04-06
01-03-17 06:55 PM - Post#217280    
    In response to penn nation

Thanks for trying to gift us with an extra win, but Lehigh was one of the more frustrating losses this season.

 
penn nation 
Professor
Posts: 21081

Reg: 12-02-04
Re: UCF
01-03-17 07:00 PM - Post#217281    
    In response to penn nation

My bad. You beat Hawaii, another fairly poor team.

 
AsiaSunset 
Postdoc
Posts: 4350

Reg: 11-21-04
Re: UCF
01-03-17 07:54 PM - Post#217284    
    In response to penn nation

I think whoever mentioned the relative meaninglessness of the Pomeroy and Sagarin ratings for our league at this point of the season was right on.

It's so wide open. Even Cornell - a team I see struggling - can beat any Ivy team on any night. They aren't very good and chuck 3's all night, but there is going to be that night when Morgan hits everything and Robert Hatter has a big night to complement him.

Yale is the only team that kind of stands out and Lee is a hard gym to win in. The others kind of look a bit like us. So the 7th will be interesting. We'll see if Pomeroy means much after the Princeton game.

 
mrjames 
Professor
Posts: 6062

Loc: Montclair, NJ
Reg: 11-21-04
01-03-17 08:54 PM - Post#217285    
    In response to AsiaSunset

Empirically this is false. The ratings are VERY predictive at this point in the year.

"I think whoever mentioned the relative meaninglessness of the Pomeroy and Sagarin ratings for our league at this point of the season was right on."

 
SRP 
Postdoc
Posts: 4894

Reg: 02-04-06
01-03-17 09:34 PM - Post#217289    
    In response to mrjames

The ratings take into account the percentage margin of victory, which contains useful information even when beating bad teams: How much you beat them by helps predict future performance.

 
Jay O 
Masters Student
Posts: 547

Loc: Philadelphia
Reg: 11-16-09
Re: UCF
01-04-17 08:53 AM - Post#217303    
    In response to AsiaSunset

Asia, you've highlighted a great argument in favor of the ratings. Just because Cornell can beat any team on any night doesn't mean they can beat every team on every night, or even most teams on most nights. That's what the ratings tell us: not that they'll never beat any of the top teams, but that it's less likely they will. I don't have the data available, but I wouldn't be surprised if Cornell is expected to collectively average one or two wins in their 8 games against HYPP, even if not favored in any of them individually. Isn't this what you're saying?

 
SomeGuy 
Professor
Posts: 6391

Reg: 11-22-04
Re: UCF
01-04-17 08:56 AM - Post#217304    
    In response to AsiaSunset

I agree that Cornell is a high variance team that could knock anybody off on any given night. However, I think ratings like Pomeroy evaluate them pretty accurately. What Pomeroy tells you is where Cornell is likely to end up over 14 games. It won't correctly predict the outcome of every individual game, and it doesn't claim to be able to do that.

In regard to Penn and Princeton, the ratings are close enough that Pomeroy will give Penn a chance to win. So Penn winning the game won't tell us for certain whether the ratings are meaningful, either. I think we'll have to wait deeper into the season to see whether the current ratings hold.

For what it's worth, there is a much greater spread in the league than their appeared to be back in early November. Penn has played much better than expected, and Columbia, Cornell, Brown, and Dartmouth have all played worse. So out of conference, the top four was very clear. We'll see if it is clear in conference.





 
SomeGuy 
Professor
Posts: 6391

Reg: 11-22-04
01-04-17 09:13 AM - Post#217305    
    In response to penn nation

Sure, but Princeton has a pretty clear line. Lost to everybody 103 and above, beat everybody 120 and below. The rest of the league has a more muddled record, with at least one bad loss.

For Penn, obviously the win at UCF is better than anything Princeton has done. However, Penn managed to lose by 19 at home to a team rated lower than Princeton's worst loss (and none of the Princeton losses were by more than 11, and all but one of those were on the road). And then Penn managed to lose at Navy as well.

Penn has a legit chance to finish top 3, I think (and a legit chance to finish out of the top 4, as well). But the rankings strike me as an accurate depiction of how everyone has played thus far.

 
pennhoops 
Postdoc
Posts: 2470

Reg: 11-21-04
01-04-17 10:27 AM - Post#217308    
    In response to T.P.F.K.A.D.W.

  • T.P.F.K.A.D.W. Said:
  • TheLine Said:
The SOS at CBB Reference is explained here :
http://www.sports-reference.com/blog/2015/03/srs-c...

It's a straight-forward calculation for college basketball.



Got lost in the black hole that is sports-reference.com and came across the name Maurice Watson, Jr. Now playing for Creighton and putting up some impressive numbers. Averaging 13 and 9 at a Big East school.

Wasn't Jerome Allen recruiting this kid?

http://www.sports-reference.com/cbb/players/mo-wat...



Nearly six years on, I stand by this post in full:

pennhoops
Postdoc
Posts: 2368

Reg: 11-21-04
Online 03-30-11 03:17 PM - Post#102784
In response to Ben Franklin

If he is not playing for Penn in 2012 something is very, very wrong.


 
penn nation 
Professor
Posts: 21081

Reg: 12-02-04
01-04-17 10:35 AM - Post#217309    
    In response to SomeGuy

Most of Princeton's wins are against teams far, far below 120. Sometimes below 200 and sometimes even below 300.

  • SomeGuy Said:
Sure, but Princeton has a pretty clear line. Lost to everybody 103 and above, beat everybody 120 and below. The rest of the league has a more muddled record, with at least one bad loss.




 
SomeGuy 
Professor
Posts: 6391

Reg: 11-22-04
01-04-17 11:06 AM - Post#217310    
    In response to penn nation

Yes -- same as everyone else in the league. The difference is the lack of any bad losses. We know that everyone else in the league is susceptible to losing to sub 200 level teams. Princeton is the only team that hasn't lost any of those. Perhaps they are prone to losing them just like everyone else, but we haven't seen it yet. The fact that they've taken care of business in those games is a good fact for them, not a bad one.

Note that, while I am using Pomeroy to look at who each team has beaten and lost to, ultimately this type of rating is more a measure of how well you play. As Brown shows, simply winning a bunch of games against sub 200 teams isn't going to give you a good rating. Princeton has a high rating because they've consistently played tight games with top 100 teams, and they've consistently beaten sub 200 teams (by increasingly high margins).

 
mrjames 
Professor
Posts: 6062

Loc: Montclair, NJ
Reg: 11-21-04
01-04-17 12:01 PM - Post#217312    
    In response to SomeGuy

I am a little more pessimistic on Princeton than Pomeroy is (primarily because I take a less preseason-weighted view on teams that have changed from the preseason). I have Yale as marginally outperforming Princeton thus far and both of those two marginally ahead of Penn and Harvard (though Harvard's December run is better than any stretch any Ivy has put together thus far).

While each of the current Top 4 teams has a 90 percent or higher chance of making the Ivy tourney, it's important to note that the odds that all four make it are closer to 75%. The most likely to bust the party is Columbia, though Brown and Cornell are at nearly 10% and Dartmouth (playing like it has recently) should be in that neighborhood as well.

It is also very unlikely that all four of the Top 4 will get a clean break from the bottom four. There should be some semblance of a race deep into the season even if those Top 4 are all prohibitive favorites to be the qualifiers for the tourney. Yale has 5-of-6 on the road to start, so it's either going to be in great position for the 1-seed OR playing significant catch up at home down the stretch. Harvard has a similar, but less acute problem, with 5-of-8 on the road to start and the toughest pairing in the league (Penn/Princeton) as its only home weekend.

 
whitakk 
Masters Student
Posts: 523

Age: 32
Reg: 11-11-14
01-04-17 08:29 PM - Post#217332    
    In response to mrjames

Do you know how much preseason weight is still in KenPom? IIRC they fully phase out by late January, so it should be pretty small by now.

That said, KenPom (and other ratings) have some Caruso/Brase performance baked in, so you do need to discount Princeton's ranking some.

 
mrjames 
Professor
Posts: 6062

Loc: Montclair, NJ
Reg: 11-21-04
01-04-17 10:09 PM - Post#217336    
    In response to whitakk

It's a lot. KenPom has always said that technically preseason weights are still valuable at the end of the regular season (i.e. more predictive than no preseason weights). In that same post he explained that he was extending the influence of those weights versus previous iterations of his model.

 
penn nation 
Professor
Posts: 21081

Reg: 12-02-04
01-04-17 11:02 PM - Post#217339    
    In response to mrjames

Thank you for confirming my hunch. The Princeton ranking just did not pass the sniff test based on actual season play alone.

  • mrjames Said:
It's a lot. KenPom has always said that technically preseason weights are still valuable at the end of the regular season (i.e. more predictive than no preseason weights). In that same post he explained that he was extending the influence of those weights versus previous iterations of his model.




 
TigerFan 
PhD Student
Posts: 1871

Reg: 11-21-04
01-05-17 12:18 AM - Post#217340    
    In response to whitakk

whitakk said:

  • Quote:
That said, KenPom (and other ratings) have some Caruso/Brase performance baked in, so you do need to discount Princeton's ranking some.




I've been very disappointed by the Tigers' play this year but actually think they are playing better on both ends of court since Brase and Caruso were lost for the year. I don't know what Ken Pomeroy's computer says but the cutting, passing, shot selection, 3-point shooting, FT shooting and team defense have all looked better to me since they went down.


 
SomeGuy 
Professor
Posts: 6391

Reg: 11-22-04
01-05-17 12:27 AM - Post#217341    
    In response to TigerFan

I do think that losing Brase kind of simplified things for them. They were almost a little too deep with him back, and it seemed to cause them some confusion as to who they were. I agree that they've played even better since Caruso went down, which surprises me a little more. Cook seems like he's come into his own as a reliable scorer, which helps fill in where they sometimes needed Caruso. My guess is that ultimately they will be worse off without him (and at this point another major injury could be devastating), but it really hasn't manifested itself yet.


 
whitakk 
Masters Student
Posts: 523

Age: 32
Reg: 11-11-14
01-05-17 12:38 AM - Post#217342    
    In response to mrjames

Do you remember when he made that change? Last I can find (and still linked from his FAQ) is from 2012, when it was still a January phase-out (though he suggests it should probably be longer)

http://kenpom.com/blog/preseason-rating s-why-weigh...

 
Penndemonium 
PhD Student
Posts: 1877

Reg: 11-29-04
UCF
01-05-17 06:16 AM - Post#217343    
    In response to pennhoops

I looked this up and was sorry I did. Think we could have used this kid?

http://www.gocreighton.com/roster.aspx?rp_id=904

There is some video of him on youtube. I think if he went to Penn, perhaps Allen would still be coaching, we'd have won a few championships and NCAA berths, and we'd have a pipeline to the public league.

I'm dreaming, of course.

Edited by Penndemonium on 01-05-17 06:26 AM. Reason for edit: No reason given.

 
JadwinGeorge 
Senior
Posts: 357

Age: 75
Reg: 12-04-15
01-05-17 08:57 AM - Post#217345    
    In response to TigerFan

The "disappointing" aspect of the Tigers' record is their inability to close out winnable games on the road. They actually played well enough to compete for more than 35 minutes in most games. If they had finished the job in 1 or 2 of those games the story line would be much different. They sleepwalked through St. Joe's at home, obviously reeling from the news that Caruso was lost. Stephens gives us everything Caruso offered. And the "5 spot by committee" is working so far. BTW the Penn game is available to Comcast Sports subscribers in Phila.

 
 Page 8 of 9 « First<6789
Icon Legend Permissions Topic Options
Report Post

Quote Post

Quick Reply

Print Topic

Email Topic

19104 Views





Copyright © 2004-2012 Basketball U. Terms of Use for our Site and Privacy Policy are applicable to you. All rights reserved.
Basketball U. and its subsidiaries are not affiliated in any way with any NCAA athletic conference or member institution.
FusionBB™ Version 2.1 | ©2003-2007 InteractivePHP, Inc.
Execution time: 0.229 seconds.   Total Queries: 16   Zlib Compression is on.
All times are (GMT -0500) Eastern. Current time is 11:43 AM
Top