David Ortiz is not a clutch hitter: A primer in how to lie with statistics

August 5th, 2006 → 1:00 pm @ // No Comments

In a blog entry posted yesterday, Matt Sussman — columnist for the Toledo Free Press and editor of The Futon Report — attemtps to show that David Ortiz is not a clutch hitter. Sussman’s piece is for a site called Blog Critics, which, in what seemingly is not meant as irony, advertises itself as being written by “superior bloggers on music, politics, TV, film, books, sports, gaming, science, technology, and culture.”

You’d think by now I would have learned my lesson: it’s not nice to make fun of people who have fundamental misunderstandings about whatever it is they’re writing about (to say nothing of the complaint that the blogosphere is comprised of a bunch of guys sitting in their basements shouting at each other). But this blog is mainly about the Red Sox (and my book Feeding the Monster), and David Ortiz is, once again, pretty much carrying the team on his back. What’s more, Sussman garnered a linky link from Deadspin, one of the best sports site out there. So I feel obligated to point out the extent to which Sussman’s piece is simply moronic.

There are plenty of obvious examples in the piece. After characterizing the Red Sox front office as “a group of intelligent number crunchers” who “buy (a little)” into sabermetrics, Sussman seems to imply (although it’s hard to figure out exactly what it is he’s trying to say here) that Kevin Youkilis, “when measured under the criteria of sabermetrics,” is more valuable than Ortiz. This is absolutely asinine. Ortiz’s and Youkilis’s OBPs are virtually identical (.393 to .395), and Ortiz has an extra 182 points in slugging percentage. Sussman also writes that Baseball Prospectus‘s Nate Silver thinks “hardcore sabermetricians” think “clutch hitting is an illusion, and such an ability doesn’t exist.” Silver actually wrote the exact opposite, saying, “Clutch hitting exists, more than previous research would indicate.” Bill James has also acknowledged that there is such a thing as clutch hitting; as he told me last year, “I’m still not sure exactly how to measure clutch hitting. That doesn’t mean it doesn’t exist. Watching Ortiz, it’s hard to think it doesn’t.”

But let’s address the main point. Sussman’s thesis seems to be — and again, it can be a little hard to figure this out since the entire article is written as a discussion (or argument) that Sussman is having with himself — that there are those in baseball (Manny Ramirez, Albert Pujols, David Wright) who have higher batting averages than Ortiz does with two outs and runners in scoring position. After lauding sabermetricians, it’s odd for Sussman to rely purely on BA; OBP, slugging percentage, or OPS would all be more valuable comparison points. And it’s true: Manny and Pujols both top Ortiz in average, OBP, slugging, and OPS with runners in scoring position and with RISP and 2 outs, while David Wright tops him with RISP and 2 outs.

But if you want to try to measure something as ineffable as clutch hitting, you need to be a little more creative than just plugging in numbers from ESPN’s splits page. After all, as a certain third baseman in New York knows, there are home runs and two-out hits, and then there are home runs and two-out hits that actually mean something.

As Allan Wood recently pointed out in the excellent Joy of Sox, Ortiz has, since the end of the 2004 regular season, been at the plate in a walk-off situation 19 times. And, as Wood discovered, Ortiz reached base 16 of those times. In his 14 official at-bats, he’s had 11 hits, with seven home runs and 20 RBIs. Over the last two seasons, he’s 8-9 with five home runs and 15 RBIs.

Think about that for a second. Over the past two years, Ortiz is hitting .875 when he has a chance to end the game with one swing. His on-base percentage is .923. His slugging percentage is 2.556. And his OPS is 3.479.

To be sure, there are other ways to try and parse out Ortiz’s “clutchness.” You could look at periods during which the Sox are scuffling, like right now. You could look at stretch runs, like last September, when the Sox closed out the month going 6-2, an eight-game stretch during which Ortiz had seven home runs and nine RBIs.

With pretty much the entire world acknowledging that Ortiz has had a three-year run that will remembered for the ages, why do pieces like Sussman’s bother me so much? The main reason is that misinformation drives me nuts. And, as I learned in ninth grade when my journalism teacher made my class read Darrell Huff’s 1954 classic How To Lie With Statistics, you can use statistics to prove pretty much anything. One day, much smarter people than me will hopefully find ways to accurately measure the impact of things like baserunning ability and clutch hitting (and will come up with better ways to measure defense). Until then, we’ll need to make do with the statistics we have. And we’ll need to use them intelligently.


Post Categories: Clutch Hitting & David Ortiz & how to lie with statistics & sabermetrics

6 Comments → “David Ortiz is not a clutch hitter: A primer in how to lie with statistics”


  1. s1c

    17 years ago

    After his last two walk offs my teen age daughter over heard me say “the man is inhuman” to which her reply “you relize you are contradicting yourself”. That is what I think whenever someone wants to pooh pooh Big Papi’s late heroics. You know he will fail at some point, that is the nature of the game, but what a ride it has been.

    Reply

  2. cense

    17 years ago

    “Ortiz has had a three-year run that will remembered for the ages…”

    I don’t have a well-formed opinion on the clutch argument, but statements like this undermine your credibility and argument as a whole. In fact, Ortiz hasn’t even been the best player on his own team over the last three years. Since Ortiz is exclusively a hitter, let’s just look at the most comprehensive hitting metric available, OPS+ (which is simply OBP plus SLG relative to the league average). Ortiz has an average OPS+ of 149.67 over the last three years. Manny has an average OPS+ of 156 over the last three years while playing LF. Even without giving Manny any credit for playing LF, it’s clear that he’s contributed more to the Red Sox’s success over the last three years more than Ortiz. Over their careers, Manny has had four seasons (1999-2002, OPS+ between 162 and 190!) that are clearly better than Ortiz’s best season (2005, 161 OPS+). Manny’s 2003 is also better than Ortiz’s 2005 if one gives Manny credit for playing LF. Manny then has five more seasons (1995, 1996, 1998, 2004, and 2005) that are clearly as good or better than Ortiz’s second best season (2004, 145 OPS+). Manny’s 1997 is also better than Ortiz’s 2004 if one gives Manny credit for playing LF. In short, Manny’s performance over the last 11 years (1995 through 2005) has been substantially better than Ortiz’s last three years. Over 11 years, Manny’s averaged a 159.82 OPS+, compared to the 149.67 OPS+ Ortiz has posted over the last three years. Manny has also provided some (perhaps small, but some) value from playing left field. If Ortiz’s past three years aren’t the best three years on his own team and if Manny’s performance over the past 11 years blow Ortiz’s past three years out of the water, how can you claim with a straight face that Ortiz’s past three years will be “remembered for the ages”? In short, you should take your own advice and use statistics intelligently. Doing so leads to three main conclusions: (1) Ortiz’s last three seasons have been good but by no means extraordinary; (2) Manny is significantly better than Ortiz; and (3) Manny is yet another great Boston LF (obviously inferior to Williams, but better than Yaz, I think). One of the worst parts of the Ortiz love-fest is that it overshadows how great Manny is.

    That’s like my saying Mick Jaggger’s singing will be remembered for the ages and your getting all upset because somehow that devalues the guitar playing of Keith Richards. Or arguing that I can’t say Curt Schilling has been the best post-season pitcher of the last decade because somehow the attendant implication is that he’s had a better career than Roger Clemens. One has nothing to do with the other. Since you “don’t have a well-formed opinion on the clutch argument,” I’d save your arguments for some future thread about the overall value of Manny versus Ortiz. (And if you think Manny Ramirez is a great left-fielder, or if you don’t think we’ll be talking about Ortiz’s run from 2004 through 2006 for years to come, I’d worry more about your own credibility and less about other people’s.)

    — Seth

    Reply
  3. […] Jess Sussman says no. Seth Mnookin (and pretty much the rest of the world) says yes. One of the most astonishing numbers quoted in the David-Ortiz-is-a-clutch-hitter arguments is that Ortiz has come to plate 19 times since the end of the 2004 regular season, and gotten on-base in 16 of those plate appearances. In his 14 official at-bats in that time span, Ortiz has 11 hits. If this seems like statistical cherry-picking, it is, and I’ll get back to that in a second. But are these numbers statistically significant? […]

    Reply

  4. yerfatma

    17 years ago

    Thank you for this piece. It’s weird to me that people who want to claim they see players’ talents through the cold, dispassionate eye of statistics get their panties in a bunch about statistics. And I say that as someone who loves both the game and the numbers.

    Reply

  5. HFXBOB

    17 years ago

    It’s absurd to argue against Ortiz not being one of, if not the greatest clutch hitters in baseball history, whether relying on statistics or common sense. I think there is one aspect of some clutch performances that gets underplayed, and that is the fact that they couldn’t happen without key contribututions from other players. For example in the July 31 9-8 walkoff game against Cleveland, Ortiz’ big hit was made possible by Cora and Youkilis getting on base and Snyder throwing 4.1 scoreless innings. This illustrates the inherent team nature of the game and also the endless analytical possibilities. Ortiz was and is the ultimate finisher, but perhaps a different type of clutchness, and one that is measurable, could be assigned to the hit Cora got, a leadoff hit in the ninth with the team down two runs, making a rally possible.

    Reply

  6. mtalinm

    17 years ago

    Seth, that’s a somewhat self-serving reading both of Baseball Between the Numbers and Bill James’ essay “Underestimating the Fog.”

    In the former, although Silver does state “[c]lutch hitting exists, more than previous research would indicate” he goes on to say (2nd to last paragraph in that chapter) that it makes only a marginal difference–2% of his WinEx statistic, to be exact–approximately equivalent to baserunning ability.

    The most James really states in his article is that he was previously wrong to say that clutch hitting doesn’t exist; he’s pointing out that the statistics don’t allow him to say one way or the other, leaving him with little more than a gut feeling (from watching Ortiz) that it might.

    That said, you are correct that Sussman built a faulty argument on misinterpreting them. But neither has gone so far as to term clutch hitting a key component of a winning season (unless I’ve grossly misread what they’ve written).

    What no one talks about is that Ortiz also chokes in the clutch. Think of the 8-2 loss to the Angels a couple weeks ago when it was the bottom of the 9th, two out, and bases loaded. Had “Papi done it again” we could have been within a couple of runs, but he struck out (weakly, I might add). So it’s not always the happy ending. Sussman is, I think, correct to point out that we harbor the good images and forget the bad.

    Reply

Leave a Reply → yerfatma

%d bloggers like this: