It's Time, Unfortunately...

[Knight78]

Espn's BPI does most of that. I actually prefer it to kenpom, because while kenpom measures performance it doesn't care if you win. And unless you're playing stratomatic you play to win the games.

Kansas: BPI 4
Lamar: BPI 349

So theoretically, you think it should be the same weight to lose to a team like Kansas by 2 or lose to them by 20? Likewise, it should be the same weight to beat a team like Lamar by 2 or beat them by 20? And do you really think a 2 point loss to Kansas should score less than a 20 point win over Lamar?

Yes, you play to win. But there are other factors involved.

 

[wundergrape]

I've always wondered how the top team in a conference always seems to have the toughest schedule. Kansas is by far the best team in the big 12 and every other team has to play them twice but obviously Kansas doesn't have to play themselves. That should be a huge factor in strength of schedule but it never seems to matter.

I know in this case Kansas played a very difficult non conference schedule but it seems to always work out that way for some reason.

My completely unscientific analysis is that the best teams in the conference are typically blue bloods (Syracuse, Kansas, Michigan State, Arizona) who schedule for April instead of March.

 

[PCHawk]

My completely unscientific analysis is that the best teams in the conference are typically blue bloods (Syracuse, Kansas, Michigan State, Arizona) who schedule for April instead of March.

Yea I completely get that, but how much does Syracuse have to compensate for not having to play the number 1 team in the nation twice?

 

[BryceC]

Kansas: BPI 4
Lamar: BPI 349

So theoretically, you think it should be the same weight to lose to a team like Kansas by 2 or lose to them by 20? Likewise, it should be the same weight to beat a team like Lamar by 2 or beat them by 20? And do you really think a 2 point loss to Kansas should score less than a 20 point win over Lamar?

Yes, you play to win. But there are other factors involved.

And BPI measures all of those as well.

 

[Benddontbreak]

Kansas Jayhawks: Have you taken the time to really evaluate the Kansas resume, if not I would. It is very, very impressive. I have no problem with them getting a 2 if selection was today.

Iowa State: If they seeded the NCAA Tourny today Iowa State would get a higher seed than Iowa. Not tough..


Minnesota: If tourny selected today they would be a LOCK. How they hell are they not?


RPI is a joke but IMO it wouldn't change the above.

They played a tough schedule no doubt, but lost to pretty much every respectable team they played in the non-con, excluding the November 12th game vs a struggling Duke.

 

[Skeets77]

Pick two teams to compare record, RPI and SOS details head to head. By default, the top two teams in RPI are shown.

NCAA College Basketball RPI Team Comparison - CBSSports.com

 

[DDThompson]

Oh, and you could actually build a model that includes weightings for when a game is played and include factors for momentum and adjustments for individual injuries.

They do.

It's called BPI.



Nevermind....ARRGHGGH

 

[libradawg]

I really do think ESPN is trying to weigh their sports influence by constantly bleating on about their BPI. Obviously I won't argue with it since it's fair to us, but the way they talk about it they want to see how much weight it gets from the committee. Because they talk about it a LOT.

 

[atxhawk]

RPI doesn’t take into account where games are played. The fact that its used by selection committee is a freaking joke.

Or margin of victory.

 

[DDThompson]

I really do think ESPN is trying to weigh their sports influence by constantly bleating on about their BPI. Obviously I won't argue with it since it's fair to us, but the way they talk about it they want to see how much weight it gets from the committee. Because they talk about it a LOT.

ESPN talks about ESPN a LOT. ESPN breaks news a LOT that has already been broken by other NOT-ESPNs.

The forumula as explained makes more sense than the RPI. There's some breaking news....

 

[atxhawk]

Someone explain this one to me. Team A has a non-conference RPI of 13 and a conference RPI of 12, but their overall RPI is 9.

Iowa is 60 for non con, 13 conference, and 29 overall.

 

[MisterLucky]

So....We go to Penn St. and win by a dozen and drop two spots. Makes sense to me.

 

[H8IAST8]

Seeds Of 4+ With Best Final Four Odds
Team Final Four %
Louisville 33.7%
Iowa 19.6%
Ohio State 15.1%
Kentucky 14.3%
UCLA 10.1%


And 11% chance to make Championship. Shhhh, dont tell Woody

 

[volhawk50]

So....We go to Penn St. and win by a dozen and drop two spots. Makes sense to me.

Of course. Because our winning percentage went up 0.01 and Penn State's winning percentage went down .02, and opponent's win percentage counts twice as much as your win percentage (as it should, right?), so all else equal we should have lost 0.03 percent with that win.

 

[jacknicholson]

I prefer Sagarin's rankings, probably has no meaning to the NCAA committee however.....

He has the Hawks ranked #8
Wisconsin #7
Michigan State #10
Michigan # 11
Ohio State # 14
Iowa State # 16

 

[jacknicholson]

I love the Kenpom rankings.....

2014 Pomeroy College Basketball Ratings

 

[RocknRollface]

RPI doesn’t take into account where games are played. The fact that its used by selection committee is a freaking joke.

Its the metric that is least accurate in determining the best teams and therefore gives the midmajors their highest rank.

The commite wants as many mm's as they can get away with so they use the RPI. They know exactly what they're doing.

Im still ****** Iowa wasnt in last year when they were clearly better than a ton of teams picked ahead of them.

 

[ssckelley]

Someone explain this one to me. Team A has a non-conference RPI of 13 and a conference RPI of 12, but their overall RPI is 9.

Iowa is 60 for non con, 13 conference, and 29 overall.

The teams weighted win percentage is 25% of the score. Really SOS ranks do not factor into the RPI at all.

 

[RocknRollface]

I should say that I am a big, big proponent of using mathematical models as an input in determining relative rankings of NCAA teams. It's just that as mathematical models go, RPI is frajacking embarrassment. (I'm mean 75% of the prediction weight does not even depend on a team's performance.... are you fracking serious, errrmmmmmaggggrrrrddddd!!) As a guy who does statistical analysis for a living, I find its use by the NCAA to be offensive to all mathematical modelers out there.

You are correct, way too much stock put into RPI. What it can't account for is time of year, injuries, who is hot when, ect. Why we say somebody should or shouldn't be in a tournament based on a mathematical formula is one of the most absurd things in sports. There are several other formulas that make just as much sense, if not more than RPI. Why so much love for RPI?

It fits their agenda.

As an Iowa fan I hate it but I completely understand why they want the mm's in rather than midle of the road power conference teams.

It makes for much better tv when the little guys win the first week or two.

 

[HawksMN]

It fits their agenda.

As an Iowa fan I hate it but I completely understand why they want the mm's in rather than midle of the road power conference teams.

It makes for much better tv when the little guys win the first week or two.

Ha last year's bracket shot that logic all to heck. Mid majors bubble teams did NOTHING in last year's dance. What a bust. We all know Iowa belonged in the bracket last year, no use rehashing painful memories.