A Visualization of NFL betting lines. Cool viz, but it makes me wonder how many people out there are, say, betting on the Cowboys because they're so good at beating the spread this year. I'd think a betting system that just counter-bets stupid analysis would be a winner. But then again, there are probably already thousands of sharks who do that.
How have defenses adapted as the fullback disappears from NFL offenses?
Little technicalities like this can drive guys like me crazy when the numbers don't add up. Neil has the answer.
Peyton Manning is obviously on steroids.*
Did the Toronto Maple Leafs prove advanced hockey metrics wrong? Phil's take.
Never miss an installment of Bill's Thank You for Not Coaching series.
Here's one for you snooker fans out there.
I was so happy to see the Jaguars get so aggressive vs. the Broncos last week. Davids need high variance to beat Goliaths. (By the way, check out Dean Oliver's original and ancient post on the subject.) This post illustrates the principle nicely in terms of the Jags-Broncs game.
Did you know that in England they call a buttfumble an arsedrop?
You can buy futures in Arian Foster and other athletes! Can you short them too? If only I could have shorted Chris Johnson a couple years back...
Greatest sports rivalries of all time. I didn't realize that about the Chargers and Royals.
Jason does an excellent job using the WP calculator to empirically show that teams that went for it on 4th down have won more often than expected. Jason also did a great analysis of the Patriots big comeback.
A viz of the height and weight of every NFL player. Pretty cool color-coded clusters by position.
Doug tells us that Manning's 2013 is ahead of Brady's 2007 through 6 weeks.
There's a lot of reason why betting on huge favorites is not smart. Here are a few.
ESPN has a new draft database. PFR's is also excellent, and keeps getting better.
*just kidding
Roundup 10/19/13
published on 10/19/2013
Subscribe to:
Post Comments (Atom)
I think it's time to start initializing Jacksonville WP at about .2 at the start of games.
True for several teams. But then the winner only gets +0.20 WP for the win?
Brian, the only way to solve that is to give players/teams an expected baseline of WP, and if you do it correctly it'll add up the same.
For example, Rivers had +0.37 WPA against the Jags this week. If you started the WP chart at 0.8 for the Chargers, maybe Rivers only gets +0.17 in the game, but then you add in his expected +0.2 WP to get 0.37 WP. In the end you should get to the same place (if your expected values are good) so it's fine to keep it the way it is for individuals. The only reason to change it would be to determine the in-game team probabilities.