More on advanced stats for golf.
Here is a great application of the Expected Points values I released into the wild last season. This is exactly the kind of stuff I hoped would start happening. It's a look at the Cowboys running game through the lens of EPA. I'd love to see more people take advantage of the EP model.
Here's another article in the same vein. It implores Broncos coach Josh McDaniels to use the EP model to become more aggressive on 4th down.
Sean McCormick at Football Outsiders has a great write-up on the much-hyped Draft class of 2004.
The geometry of free-throw shooting. (Hat tip: M/R)
A strange quirk in how seeding in the NCAA basketball tournament can allow lower seeds to advance more easily. #10 seeds are actually more likely to advance than #9 or #8 seeds. #12 seeds are twice as likely to make the Sweet 16 than #8 seeds.
Turnovers are the key to predicting upsets.
One reason why understanding statistics is important in business and public policy.
An NFL study finds that ACL injuries occur more often on FieldTurf, a brand of artificial grass. Most of these kinds of studies are total junk, but from the article it sounds like solid research. I'd love to see the actual paper. (Hat tip: PFT)
Jason Lisk at PFR is running the Woulda'-Coulda'-Shoulda' Tournament, a playoff among the best teams to fail to win a championship.
More from Jason--his thoughts on completion percentage. Specifically, when does a very high completion percentage not matter. Guys like David Carr and Brian Griese can post extremely high completion percentage without really helping their teams. I ran some numbers, and if you're not getting more than 10 yards per completion, a QB is not going to be a net positive for your team no matter if his completion percentage is 75%.
How much is a statistical analyst worth financially to a professional sports team?
What are Bill Parcell's 4 rules for drafting a QB? (Hat tip: Smart Football) Basically Parcells likes to see sustained performance. Too many one-year-wonders have become quick busts. Every year there's going to be a dazzling passer who puts up great stats behind a dominant line against a run of weak defenses. Parcells' rules essentially rules these guys out.
Some critiques of the Plus/Minus stat in the NBA.
At Fifth Down, Chase Stuart looks at what proportion of each team's player value was obtained in the draft. In other words, which teams are built on the strength of their draft picks? At PFR, Chase also looks at the QBs taken in the 2007 Draft. JaMarcus Russel and Brady Quinn are both huge busts, but how does the pair from 2007 compare to other "bust-classes?" If the Browns had only listened to the Onion back then...or if the Broncos are listening now...
Chase has been busy. He looks at how formerly-elite players have fared after changing teams. And here is a great 2-part look at the best undrafted players.
Mike Florio thinks teams should gladly poach restricted free agent wideouts who are proven performers. His point is that there are lots of top round busts at WR, so the best thing to do is grab a proven guy off another roster. But it's not so simple. WR is no different than other positions in that the best WRs really do come the top rounds of the draft. Plus, Florio is only looking at the benefit side of the equation. The costs are much higher for a poaching team. The costs to the team trying to pry away the restricted free agent include both the salary and draft picks, while the team with restricted rights to the player only needs to match the salary offer.
Do you hate your favorite team's rival as much as I hate my favorite team's? Have you wondered why you fixate on a single opponent like that? This might be the reason.
I need a new term for hat tip. Tango has "glove slap." What about "helmet knock?"
Roundup 3/20
By
Brian Burke
Subscribe to:
Post Comments (Atom)
Butt slap?
How can the geometry of free throws article not have pretty graphs?
While the current team only has to match the salary offer, their costs in a sense are the same, b/c if you choose to match you do not get the draft picks from the other team. So if another team decides to make an offer, the cost of the player is the same for both teams. The salary offer and the picks involved. One team will "give up" the available picks and pay the money to keep the player, the team who restricted the player just decides which team is which.
hmm...maybe we should reseed the ncaa tournament? but that would never happen because people love filling out brackets. and even the NCAA isn't so ardently anti-gambling as to take on the brackets that spark so much interest in their tourney.
Also, in keeping with football tradition, I think butt slap (or slap on the rump?) would be more appropriate than helmet knock. :)
Definitely the helmet knock
Authentic Sports Memorabilia
need the new NFL overtime post
I just read the first link - advanced stats for golf - and I have to say that I'm impressed as heck with the PGA. Usually, the role of sports' governing bodies in stat management ends at "we will officially track this stat or we won't". But here we have an example of a major sports organization actively reaching out to, and partnering with, the highly respected scientific and academic bastion of MIT to improve their putting metrics.
The PGA recognized a core problem - the fact that their current putting stats are, at best, highly subjective - and they reached out to those with some statistical muscle to fix the problem. Wow. I can't imagine the NFL or NBA doing this.
I think that the general consensus amongst most leagues is that there is an expected body of stats that surrounds their sport. And if those stats are meaningless or inaccurate, then that is simply fodder for the folks on sports radio and it really isn't an issue to be addressed by the league. I'm not really a golf fan, but major kudos to the PGA.
There was already a post on the proposed NFL rule, so I'm not sure what more there is to say. Before the team that won the initial overtime kickoff won 60% of the time. The new rule will lower that to about 56%.
Chest Bump