SomeGuy
Professor
Posts: 6412
Reg: 11-22-04
|
02-11-20 09:42 PM - Post#299598
In response to HARVARDDADGRAD
As of today, Yale would be favored even at Harvard. Of course that could change, but there is enough gap between them that I think it is more likely than not to stay that way.
Harvard would be a favorite in the rest of their games.
Of course, both teams were Pomeroy favorites in each of the four total games they’ve lost, so who is favored isn’t everything. I think Yale is a clear favorite to win the regular season, but Harvard is neck and neck with Princeton for the next best chance.
|
Bryan
Junior
Posts: 231
Loc: Philadelphia
Reg: 11-21-04
|
02-11-20 11:12 PM - Post#299604
In response to mrjames
If Yale wins out and loses at Harvard (if it loses a neutral-site game, it's toast), it would have a true resume ranking around 40. That would be an at-large resume, but with the way the committee looks at the world, that would probably be an NIT 5 or 6 seed.
Yale is rated nowhere close to 40 by NET or Pomeroy or Sagarin. What metric are you using to conclude they can get to a "true resume rating" around 40? How does that true resume rating differ from NET? If it's a better metric then shouldn't we see a similar result in at least one of Pomeroy or Sagarin as well?
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
02-11-20 11:28 PM - Post#299605
In response to Bryan
Resume and quality metrics like KenPom, Sagarin, NET are different. Resume, as measured by WAB or ESPN’s SOR are based solely on wins and losses and are decently predictable in advance because the only variable is whether teams you’ve played get better or worse.
The quality metrics are tougher to predict because it depends on how much you win by. Technically, Yale could win out and scrape by in every game and go *down* in KenPom and similar metrics. But if it finishes 13-1 in the league KenPom and similar metrics will likely rise, because it has some difficult road games remaining.
The NCAA is silent on whether “best†is best resume, best quality or both. Thus, it’s unclear whether metrics like SOR or WAB (resume), KenPom or Sagarin (quality) or both are what the NCAA is aimed at to answer their charge of selecting the “best†remaining teams.
|
PennFan10
Postdoc
Posts: 3585
Reg: 02-15-15
|
02-11-20 11:52 PM - Post#299608
In response to SomeGuy
Doesn’t this basically assume all teams are equal? Yale may have a more “difficult†path, but they will be favored in every game the rest of the way, while Harvard will not. Won’t your model almost always presume that a good team that plays its early season games on the road is in better shape than the team that played at home?
In a sense yes, at the start of the season everyone is at 0. If the home teams win the first 6 games, everyone will still be at 0. The reality is the regular season winner is almost always going to get to +4. Theoretically Cornell still has a path to do that. Yale could get to +6. Harvard’s ceiling is +4 right now.
THe model isn’t meant to rate the teams today. It’s really meant to show what kind of road they have ahead of them. Yale is currently at 0 and Harvard is +2. Harvard got to +2 by playing all their tough road games now while Yale is about to play those games. Really the model tells me Harvard is not in as bad of shape as 1-3 over the past 4 games might lead someone to believe. And Yale at 0 hasn’t really given up any ground yet either, despite the home loss to H. Negative numbers at this point aren’t that good and anyone at 0 or better is holding serve or better. The 4 likely candidates are currently 0 to +2.
|
HARVARDDADGRAD
Postdoc
Posts: 2691
Loc: New Jersey
Reg: 01-21-14
|
02-12-20 05:33 AM - Post#299619
In response to PennFan10
Having played each of the 4 other tournament contenders on the road, Harvard could be considered to be in a matter position than had it played the C’s on the road. All road games are not equal.
|
