I wrote, "Obviously, odds ratios aren't the best choice for predicting NFL results. With such a short schedule, it's silly to pin teams down to their records." I used football as an instructive example and to contrast their one-game playoffs with MLB's, and then the NBA as a larger-sample example. Read more

The rest of the sentence is "the one-game first round and short second round are designed to embrace baseball's inherent randomness." The NFL's one-game elimination is a better predictor of quality than MLB's one-game elimination. Read more

Just take your 100th 1-run win and stop whining, Scocca. Read more

PHEW Read more

+2-14 Read more

I've always felt that the Jets and Knicks were my punishment for the Yankees. Read more

I have been a Jets fan since I was 8 years old. Today, I went to my first-ever NFL game. It was Jets-49ers. This team is misery. Read more

Not to mention that Dickey's sample is the smallest part of my case against using FIP on him. Go check the career BABIPs of Tim Wakefield, Steve Sparks, Tom Candiotti, Phil Niekro, Joe Niekro, Charlie Hough, and Wilbur Wood. Read more

Gio Gonzalez's career BABIP heading into this year was .296. That is league average in a park with foul territory the size of a small country. His minor league BABIP was league average. But yes, I am the one who's trying to fit a narrative. Read more

And in response to your second point, I completely agree that the margin of error makes it essentially a tie. I'm criticizing the approach that you're accusing me of — finding a pitcher who wants to be the Cy Young and using tiny differences to make the case. Read more

Gonzalez's BABIP is far more likely to be the result of random variance. The results of his balls in play have less to do with his value than Dickey's. FIP doesn't penalize the two equally just because they both have low BABIP. There is an unequal amount of information in that component, which makes a FIP comparison… Read more

This isn't a post against sabermetricians. It's a warning for people who use sabermetric stats but might not be as familiar with them as sabermetricians are. It's great to see how much WAR has spread, but to keep the movement going, discussions about which stats are best have to include beginners and intermediates too. Read more

And that's your prerogative! But if you *are* going to use stats, as many people do, you should know how to apply them well. Read more

I was trying to say that by including only BB, K, and HR and scaling the figure to ERA, it implicitly fills in the gap. Read more

Damn. Fixed, thanks! Read more

I think that if you went season-by-season over the last 20 years in the AL, you'd find that the Rangers would be on the low end of the #1 teams in the league (they definitely are by winning and expected winning percentage). But after reading your comments I looked at the more recent (5 or so) seasons, and they're… Read more

By "default" I meant that they're weak by the standards of the typical team that has the best record in the league. If you put a team of that relative strength in another year, they probably wouldn't be the best. Read more

No it isn't. Read more

"The mean age of baseball players in my sample was 64.19 while the mean age of football players was 60.91. (Barnwell tweeted that the difference in ages between his two groups, which were defined slightly differently, was about 24 months.) The mean ages of my two groups is significantly different with a p-value of… Read more

Dodgers better not land on Boardwalk. Read more