 |
|
Oh Hi, Everybody!
I’m Matt, and I’m a Mets fan. That’s all you really need to know.
I’m new here at Hot Foot–thanks to Anthony for taking me on–but this is my 24th
season as a die-hard follower of the orange and blue. (My God, can that be right?) I was
there in 1984 when Doc won Rookie of the Year, I was there in 1986 when we won
it all, I was there when we all thought Bill Pulsipher would be the next coming
of Sandy Koufax, I was there, at Shea Stadium, when Robin hit the Grand Single,
and I was certainly there for all of 2006.
And I know it’s painful to look back, but I can’t start
writing about this year without a brief nod to last year’s team. I understand,
it didn’t turn out the way we all wanted, but last year was one of the best
years I’ve had as a fan. 1986 and 2000 have to come first and second because of
the World Series appearances, but 2006 goes right behind them. I was fortunate
enough to attend all the home playoff games last year, and beyond fortunate
enough to catch a foul ball off the bat of Jose Valentin in Game Six of the
NLCS (I think it was his hardest hit ball of the postseason). I can’t ever
remember a team as simultaneously likeable and talented as the one we had last
year, and which, by and large, we still have now.
wins? Ah, this is where my sentimental side disappears and my analytical side
takes over. The good people at FanGraphs tell us, to no one’s surprise, that
based on Win Probability Added (WPA), Carlos Beltran and David Wright were our two
biggest contributors. It may surprise some people to learn that Billy Wagner
rated our third biggest contributor. Really, Billy Wagner? Ahead of Jose Reyes,
Carlos Delgado, and Tom Glavine? WPA is supposed to directly assign credit for
wins and blame for losses. It’s somewhat results-oriented, as a guy who hits
four homers after his team has already taken a ten-run lead will get less
credit than the guy who gets hit by a pitch to drive-in the winning run, but I
still find WPA to be a fascinating statistic. I followed our players’ WPA
throughout the season last year, and became more and more puzzled as Wagner’s
number grew. Wasn’t he the same guy who blew a 4-0 lead to the Yankees? And
wasn’t he terrible in the playoffs? Yes and yes, but the Yankee game was just
one out of 70 he appeared in, a fact the stats can accept far better than I
can, and the WPA numbers are for the regular season only. (Wagner’s playoff WPA
was -.221.) Still, those regular season numbers had to be inflated somehow,
right?
I’m no expert; I’m just a guy. But I wanted to find out what was going on with these numbers, and I hope some of you reading this find the explanation as interesting as I did.
As I’m sure many
astute readers know, the WPA stats at FanGraphs are based on the calculations done
in The Book. The important word there
is “calculations.” For any given situation, The
Book has determined the theoretical percentage
of the time a team of average major leaguers would win against another batch of
average major leaguers, with no consideration for home-field advantage. For
example, take Game 1 of our NLDS. Wagner entered the game in the ninth
inning with the Mets up by two. According to The Book, the Dodgers still had an 8.2 percent chance of winning
the game. (And yes, I bought The Book
mostly just to look up numbers such as this one.) But we have data from actual
major league games over the past twenty-five years, and we can see that the
visiting team, in this case the Dodgers, actually wins only about 5.6 percent
of the time. What accounts for this difference?
faced average hitters at a neutral site, the team down by two heading into the
ninth inning would expect to win 8.2 percent of the time. But in real baseball,
you don’t get an average pitcher in the ninth inning—you get the closer. We
damn well expect the closer to do better than an average pitcher, or else a lot
of teams have wasted a lot of money. In addition, ninth-inning lineups do not
typically feature average hitters. They can’t, as the best players on the team
usually start the game, and some of them are often removed by the time last licks
rolls around. The ninth inning features an above-average pitcher against
below-average hitters. We won’t punish Wagner for being an above-average pitcher, but we can punish him, statistically speaking, for facing weaker opposition than the gentlemen who start games.
his WPA is largely a product of the situations he’s placed in. That fact may be
obvious to some of you, but I for one was pretty taken aback at just how
drastic the difference is between getting three outs in the ninth and getting
three outs at some earlier point in the game. Say you’re Pedro Feliciano, and
you enter the game in the top of the sixth, up by two, and you retire the side
in order. You get a WPA of .036. If you’re Billy Wagner in game one of the NLDS,
you can come on in the top of the ninth up by two, give up two hits and an
earned run, and still get credit for a WPA of .082. In the ninth inning, all
sins are forgiven as long as you close out the game. When you also consider
that the ninth inning is worth a lot more to begin with, you see what I finally
realized: WPA is not the best statistic for determining a closer’s value to his
team. It might, however, be a perfect statistic for comparing values of closers
against other closers.
statistic, provides a better answer to the question, “which Met contributed the
most in 2006?” VORP for pitchers is merely “the number of runs a pitcher
surrenders below what a replacement level pitcher would have given up in the
same number of innings.” In other words, there is no inherent advantage to
being the closer in VORP-land. Using VORP, Wagner was only the seventh-best Met
in 2006, falling behind not only Reyes, Delgado, and Glavine, but even Paul Lo
Duca.
was excellent for us last year. Wagner was fifth in VORP in 2006 among all NL
relievers. His 94/21 k/bb ratio was fantastic, and his .308 BABIP means he was
very unlucky. While PECOTA sees a decline in Wagner’s strikeouts in 2007, it
remains clear that Billy Wagner is not one of the guys Mets fans need to worry
about.
now. First preseason game is a week away, and from this point forward I will
only write about 2007. Let’s go Mets!
Leave a reply
You must be logged in to post a comment.