Jim Boeheim Is The New King of March Madness via FiveThirtyEight

[cuse10]

Jim Boeheim Is The New King Of March Madness

Who’s the best at coaching in the NCAA men’s tournament? When we’ve looked at the question in the past, Michigan State’s Tom Izzo always stood out from the pack. Even if you include recent seasons — which have tarnished Izzo’s reputation a little with early losses to Syracuse, Kansas and Middle Tennessee — the Spartans’ coach still gets longer tournament runs out of his teams than we’d expect based on their talent.

But that’s just one way to define the tournament’s top coaching performances. Another approach is to simply look at how often you won the games at hand, regardless of the big-picture focus on how deep you made it each year. And by that standard, the NCAA tourney’s most impressive modern coach might be the guy who outdueled Izzo last Sunday — Syracuse’s Jim Boeheim.

Boeheim is best known for his iconic zone defense and irascible demeanor(particularly with the media), plus the national championship he won with Carmelo Anthony in 2003 — the only title in his 40-plus-year head coaching career. But the numbers say that Boeheim should have more of an Izzo-like reputation for coaxing impressive performances out of the teams he’s had to work with.

Continue reading...

 

[Millhouse]

I love JB but this stuff is goofy. If you are a great coach that is supposed to win every game because your team is great because you are great, then you're not considered great because you didn't outperform the great expectations of everyone who knows how great you are

 

[Millhouse]

also, it's totally biased towards people who coached for a hundred years.

john cheney never outperformed, he just coached forever.

 

[orangenirvana]

also, it's totally biased towards people who coached for a hundred years.

john cheney never outperformed, he just coached forever.

Longevity matters.

 

[OrlandoCuse]

Great analysis. Long live King JB!

 

[Day2]

also, it's totally biased towards people who coached for a hundred years.

john cheney never outperformed, he just coached forever.

He took a six, a seven, a ten, and an eleven to the Elite Eight.

 

[upperdeck]

its also says that coaching a an avg team further than it goes is better than creating a good team to start with? there is value in both.. as much as we hate it there is value in recruiting as well.

 

[oli44]

Kevin Ollie is the 16th greatest March Madness Coach ? mmkay.

 

[ORRange]

The top ten is not surprising . Neither is Barnes, Dixon,Bennett, Thompson or Hamilton near the bottom. Self and Jay Wright are way worse than I would have predicted. But this will not work for the anti zone and JB bashers on this site.

 

[Madbiker]

When you are 7-1 as a double digit seed you are going to be high on this chart of winning when not expected to. Duh.

 

[Cusefan95]

I use statistical analysis all the time; the stuff 538 comes up with is absolute garbage. I can’t tell if it’s because they need content and cant develop quality content fast enough - or if they think things like this are valid.

I’d like to believe that they know many/most people are mathematically illiterate, so they put stuff like this out knowing that it’s trash - but has numbers and stuff so it looks legit. I think it’s more likely that Nate Silver is a dope that got wildly lucky in 2008, and has successfully convinced other dopes that he and his band of imbeciles are intelligent. Good for him...doesn't change the fact he’s a dope - just makes him a lucky, wealthy dope.

 

[Edfromsodus]

We love you Jimmy!!

 

[chugg21]

He took a six, a seven, a ten, and an eleven to the Elite Eight.

And the argument would be that if the team was good enough to get there then they shouldn't have been that bad during the regular season over the longer sample size.

 

[Millhouse]

its also says that coaching a an avg team further than it goes is better than creating a good team to start with? there is value in both.. as much as we hate it there is value in recruiting as well.

wisdom of crowds. if a lot of people think you're going to win, it probably means you're pretty good

 

[Millhouse]

I use statistical analysis all the time; the stuff 538 comes up with is absolute garbage. I can’t tell if it’s because they need content and cant develop quality content fast enough - or if they think things like this are valid.

I’d like to believe that they know many/most people are mathematically illiterate, so they put stuff like this out knowing that it’s trash - but has numbers and stuff so it looks legit. I think it’s more likely that Nate Silver is a dope that got wildly lucky in 2008, and has successfully convinced other dopes that he and his band of imbeciles are intelligent. Good for him...doesn't change the fact he’s a dope - just makes him a lucky, wealthy dope.

i agree with this. they did a really goofy study of fandom using facebook likes that was embarrassing

 

[Day2]

And the argument would be that if the team was good enough to get there then they shouldn't have been that bad during the regular season over the longer sample size.

The argument was that Cheney never outperformed his seed. Clearly he did.

 

[Millhouse]

The argument was that Cheney never outperformed his seed. Clearly he did.

i forgot that he finally broke through at the end of his career

 

[Day2]

i forgot that he finally broke through at the end of his career

If you want to consider 1991 and 1993 the end of his career.

 

[Cusefan0307]

I use statistical analysis all the time; the stuff 538 comes up with is absolute garbage. I can’t tell if it’s because they need content and cant develop quality content fast enough - or if they think things like this are valid.

I’d like to believe that they know many/most people are mathematically illiterate, so they put stuff like this out knowing that it’s trash - but has numbers and stuff so it looks legit. I think it’s more likely that Nate Silver is a dope that got wildly lucky in 2008, and has successfully convinced other dopes that he and his band of imbeciles are intelligent. Good for him...doesn't change the fact he’s a dope - just makes him a lucky, wealthy dope.

You're analysis is only as good as the data available to you.

 

[Knicks411]

I think Nate Silver is pretty clearly a smart guy who knows what he's doing.

Most of the other 538 stuff, eh. Chris Herring is very good though.

 

[Waltdods]

I think Nate Silver is pretty clearly a smart guy who knows what he's doing.

He definitely knows what he's doing, but I'm not sure that's always a compliment.

Those guys know they're operating here on a tiny sample size with a ton of noise, and that the results they're spitting out are largely crap. There's probably some directional validity to it (Jim Boeheim is good, Rick Barnes is terrible), but pretending that there's much more to it should be embarrassing.

 

[OrangeDW]

also, it's totally biased towards people who coached for a hundred years.

john cheney never outperformed, he just coached forever.

Disagree. Cheney often got the most out of his Temple teams, and you don’t usually get to coach for a hundred years if you’re not good.

 

[rrlbees]

I have no problem with that analysis. All it is, is comparing the number of games you should have won based on their ranking system (elo) to the number of games you won. Pretty simple. You could do it based on kenpom rankings, sagarin, etc. They chose to use their own rankings. There was another one done using tournament seeding. As a 2 you should beat a 3 but should lose to a 1. JB faired pretty well in that one too.

Bottom line is JB is a pretty darn good NCAAT coach and probably better than given credit for. People tend to remember the Richmond’s.

 

[Knicks411]

He definitely knows what he's doing, but I'm not sure that's always a compliment.

Those guys know they're operating here on a tiny sample size with a ton of noise, and that the results they're spitting out are largely crap. There's probably some directional validity to it (Jim Boeheim is good, Rick Barnes is terrible), but pretending that there's much more to it should be embarrassing.

Well Silver didn't write this article.

I definitely agree with your point in this specific case.

 

[Waltdods]

Well Silver didn't write this article.

I definitely agree with your point in this specific case.

That's fair.

This isn't a particularly novel take, but I do think that the commercial realities of operating a million dollar operation has led Silver into endorsing (even if only implicitly) a lot of crap that is contrary to his own teachings on the limitations of data. Presidential elections and baseball are hugely data-rich environments, where high-level manipulation of the data can teach you something new and useful. Once you've got a website that has to fill 365 days, and now a lot of people doing the stuff Silver was a decade ago, you're going to end up pushing numbers where they aren't terribly useful, and sometimes are the opposite.