What the hell is up with the BPI?
Michigan basketball is currently ranked 4th in Bart Torvik's rankings, and 6th in kenpom. So what the hell is up with BPI?
http://www.espn.com/mens-college-basketball/bpi
We're ranked 15th, one spot behind (wait for it...) Villanova. (To be fair, we're second in ESPN's "resume" computer ranking. But I think maybe they're overvaluing preseason expectations if they've got us ranked behind a team we beat by 20+ on the road who also lost to Furman).
November 30th, 2018 at 9:32 AM ^
We're also well behind UNC, who currently sits at #2.
jenniferlawrenceokaydotgif.
November 30th, 2018 at 11:06 AM ^
Here you go...
.gif)
November 30th, 2018 at 9:33 AM ^
And staee at 10, yeah I call bullshit!
November 30th, 2018 at 9:39 AM ^
It's an ESPN analytic, which by definition means it's barely thought out.
November 30th, 2018 at 10:20 AM ^
BPI = Barely Processed Information
November 30th, 2018 at 11:58 AM ^
Barely Provides Information
November 30th, 2018 at 9:53 AM ^
Yeah, all of ESPN's fancystats are terrible. I'm convinced they do it on purpose so the Mike Golics of the world can rail against them and how nerds are ruining sports.
November 30th, 2018 at 10:04 AM ^
You can't ever trust a computer to rank anything. You need committees of highly impressionable middle-aged white guys.
November 30th, 2018 at 10:36 AM ^
And Condoleeza Rice
November 30th, 2018 at 11:23 AM ^
Everything is better with Condoleezza Rice. Without exception. The same rule holds true for bacon.
November 30th, 2018 at 11:51 AM ^
Then why am I not on the committee?
November 30th, 2018 at 10:14 AM ^
I heard this on the radio the other day. Judge Judy is the top earner among day time TV hosts. She made a cool $147M this year. How is that even possible?
November 30th, 2018 at 10:22 AM ^
I read about this recently; last year, she sold past and future rights to the show for $100 million. So her usual measly salary is just about $47 million per year.
It’s rumored that Byrd makes $1 million per year. Not bad for working about 50 days.
November 30th, 2018 at 10:08 AM ^
At the end of the day, it doesn’t matter.
November 30th, 2018 at 10:35 AM ^
But what about the beginning of the day?
November 30th, 2018 at 11:35 AM ^
Morning wood!!!
November 30th, 2018 at 10:09 AM ^
November 30th, 2018 at 10:37 AM ^
Yeah, there is one dude that is really pissed off about this.
Just one though.
November 30th, 2018 at 10:43 AM ^
Because the GOAT racehorse (Secretariat) is no longer around to prop up Belmont's ranking?
November 30th, 2018 at 10:11 AM ^
They probably are overrating preseason expectations. However, SOR is based on BPI, so we wouldn't be as high in that if BPI ranked the other teams where they belong. This early in the season, BPI isn't every going to be very accurate, but SOR can be, so maybe that's what they are going for.
November 30th, 2018 at 1:10 PM ^
I don't think that's the case. It looks like it is essentially just calculating points per possession (for and against), taking into account the strength of the team they played (among other factors), and then adding together a score based on how well their points per possession look offensively and defensively. It does not take into account the actual wins and losses that take place.
So a team like UNC still has ridiculously high points per possession and looks good in this type of ranking.
November 30th, 2018 at 10:12 AM ^
Mickey Mouse rankings system
November 30th, 2018 at 10:14 AM ^
I wish college basketball had more ratings formulas
November 30th, 2018 at 10:18 AM ^
small sample size. torvick and kenpom use data from last year. i'm not sure if BPI does. by the end of the year, BPI, torvick and kenpom will all be about the same. this is nothing to get worked up over.
November 30th, 2018 at 10:32 AM ^
BPIs failing is that it relies too heavily on ESPN recruiting rankings for its preseason projections. I think overall it’s okay when it catches up, but for a team that lost a lot of production from the prior year, and was replacing it with primarily unranked 3 and 4-stars that had little or no college performance to project from, it is going to doubt us until we have even more of the season behind us.
November 30th, 2018 at 10:36 AM ^
The one thing the NET does is it takes away the preseason and last seasons rankings and only looks at this years body of work. That is why you have a Belmont and Loyola Marymount toward the top. I like that it takes out any of the bias.
November 30th, 2018 at 10:40 AM ^
Even Mark Dantonio thinks this is BS.
November 30th, 2018 at 12:11 PM ^
"Next ranking. You heard me."
November 30th, 2018 at 10:54 AM ^
This isn't a problem until ESPN starts using it to issue weekly mock NCAA tournament brackets, complete with play-in games, starting the week before Christmas.
November 30th, 2018 at 11:06 AM ^
What's up with it? Nothing. It's always been pretty terrible.
November 30th, 2018 at 11:08 AM ^
This doesn't bother me as much as the fact that the selection committee team sheets still include some bullshit metric called KPI, developed by a former MSU student manager, that has consistently done a terrible job of seeding teams in its history. And this thing is specifically highlighted (along with RPI, KenPom, BPI and others) on each school's official team sheet.
(KPI had Michigan a 5 and was the only entry in the Bracket Matrix that had them anything other than a 3 or a 4.)
November 30th, 2018 at 11:39 AM ^
BPI equally good as the FPI, if you catch my drift.
November 30th, 2018 at 11:47 AM ^
It heavily takes pre-season ranking into account so we have an anchor on us from that still.
November 30th, 2018 at 11:55 AM ^
It's trash.
November 30th, 2018 at 10:26 PM ^
Description from ESPN site...the BPI is a ranking “that is meant to be the best predictor of performance going forward”
meant to be, but clearly NOT, a predictor of fuck all