The following chapters are adapted from "Baseball Between the Numbers: Why Everything You Know About the Game is Wrong." Copyright (c) 2006 by Baseball Prospectus. Excerpted by permission of Basic Books, a member of the Perseus Books Group. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.

From chapter 1-2: Is David Ortiz a Clutch Hitter?
By Nate Silver

Ask fans to recall the most memorable events they have seen on a baseball field, and they'll likely name some of the following: Carlton Fisk's twelfth-inning blast off the Green Monster foul pole in the 1975 World Series, a gimpy Kirk Gibson's pinch-hit homer off Dennis Eckersley in the 1988 World Series, Joe Carter's three-run homer in the bottom of the ninth to give the Blue Jays the 1993 championship, and Aaron Boone's left-field blast in the 2003 ALCS to extend the Red Sox' misery by one last season.

What these events have in common is that they were big clutch hits -- more specifically, big clutch home runs. David Ortiz slugged five home runs and hit .400 in fourteen playoff games in 2004, including key hits such as the game-winning homer in Game 4 of the ALCS. But it was the 2005 regular season that cemented Ortiz's reputation, when he totaled 148 RBI and hit .352 with runners in scoring position. Perhaps most impressive, he blasted twenty home runs that either tied the game or put the Red Sox ahead. In September 2005, Red Sox owner John Henry presented a plaque to Ortiz with the following inscription:

The Red Sox have been on both sides of a disproportionate number of clutch hits, so this is no small praise. Some readers, however, might find a bit of irony in its source. The Red Sox organization, from Henry on down, takes pride in following the tenets of sabermetric analysis, and clutch hitting is a dubious concept in the minds of most sabermetricians, contrary to baseball's popular wisdom. Sabermetricians have conducted dozens, if not hundreds, of studies on clutch hitting. All of them have come to the same conclusion: Clutch-hitting ability either doesn't exist at all or is so rare that it is hardly worth worrying about.

BUY THE BOOK
Click here to buy "Baseball Between the Numbers" by the Baseball Prospectus team of experts.

Most of these studies have applied a similar technique to evaluate the issue of clutch hitting. Situations are broken down into two groups: "clutch" situations (for instance, late in a close game) and everything else. Summary batting statistics -- whether traditional ones such as batting average or modern ones such as on-base percentage plus slugging average (OPS) or runs created -- are then compared across the two situations. Regardless of the way the situations are defined or the batting statistics are used, this technique has found little evidence of clutch-hitting ability. Hitters who rate as good clutch hitters in one season have no disposition to rate as good clutch hitters in the next one.

It would help the discussion if we could define situations along a spectrum of importance, or "clutchness." Some of these situations are obvious -- two outs in the bottom of the ninth inning with the go ahead run at home plate is a clutch situation. Others are subtler -- what we might call "hidden-clutch" situations. Leading off the inning is a hidden-clutch situation, since making the first out of the inning is more than twice as costly as making the last out of the inning. Another example is when a team hits with one out and a runner on third base. It's critical to score the runner from third base with one out in the inning, since the runner can score on a sacrifice fly, groundout, or suicide squeeze with one out, but not with two. If the batter strikes out, pops up, or hits a liner or groundout too sharply, the runner can't advance, considerably reducing his team's run expectation.

A related problem with this type of clutch-hitting evaluation follows the assumption that it's good enough to compare the same batting statistic across different situations. Even relatively sophisticated offensive metrics like runs created assume that the value of a given offensive event is static; a double, say, is worth 0.7 runs. While that might be an excellent estimate of the average value of a double over the long run, it does not account for the fact that hitting situations are highly dynamic and that the value of different plays changes dynamically with them. For example:

• With the bases empty, a walk is every bit as good as a single. This is not the case with runners on base.

• In a tie game in the bottom of the ninth with a runner on third base, a single is every bit as good as a home run.

• If the home team is batting in the bottom of the ninth down by two runs with nobody on, a walk is virtually as good as a home run.

• With a runner on first base and less than two outs, a strikeout is preferable to a groundout.

Baseball is a game that is won by exploiting small advantages over the long haul. Certainly clutch hitting may exist in the classic sense of the term, but a lot of what we think of as clutch hitting may really be situational hitting. In some sense, the answer to the question of who the best clutch hitters are is that they're usually just the best hitters, period.

Nevertheless, people are generally thinking about something different when they describe clutch hitting. When Tony Perez was the subject of some intense campaigning a few years ago on behalf of his candidacy for the Hall of Fame, a good number of his contemporaries brought up the subject of his clutch hitting. "He's the best clutch hitter I've ever seen," said his manager in Cincinnati, Sparky Anderson. The reason we heard this argument so frequently is that Perez needed a little something extra to get him into Cooperstown. Perez's most comparable players, according to his conventional batting statistics, are Harold Baines, Dave Parker, and Andre Dawson. All three are fine players but also short of Hall of Fame caliber. The argument Perez's advocates were making was that his conventional batting statistics didn't do him justice. He had a tendency to get the right hits at the right times above and beyond those statistics, they claimed. That was what should set Perez above Parker, Baines, or Gil Hodges, and that was why he should be in the Hall.

The Perez quandary raises an important question: Are there hitters who are more valuable than their regular statistics would suggest because of their tendency to hit in the clutch?

When we analyze play-by-play data, David Ortiz does rate as a clutch hitter overall, but most of the damage was limited to just two seasons, 2000 and 2005. Take those two years away, and his lifetime clutch rating is essentially zero. He didn't rate as a clutch hitter in 2004 -- at least not during the regular season -- or in 2002. It isn't a bad track record, but if clutch hitting really exists, one would expect more consistency out of the "greatest clutch hitter in the history of the Boston Red Sox."

Nevertheless, it's worth seeing what we can make out of the skill element that does exist. Is there any pattern to which sort of players do better in clutch situations? Or do you know a clutch hitter only once you've seen him succeed in those situations many times over? The table below shows the best hitters in terms of clutch performance since 1972, as ranked by lifetime clutch rating wins (wins produced beyond what his regular batting statistics would indicate):

Mark Grace rates as the best clutch hitter of the past three decades according to our metrics, producing between 13 and 14 more wins as a result of his clutch hitting. This makes a certain amount of sense. Grace had a reputation for being a very smart hitter, and there are some references to his clutch ability in the historical literature.

What we may be seeing here is the effect of smart situational hitting. As you'll recall, a player who had the ability to adjust his hitting approach in different situations -- slapping a single or blasting a home run as the situation required -- would provide some small but discernible benefit to his club. This advantage provides another incentive for teams to acquire disciplined hitters who control the strike zone and control the at-bat. Not only do those players tend to post great OBPs, but they may be better able to take advantage of clutch situations as well.

That said, apart from the bonus effects of plate discipline, it's probably folly for a club to go looking for clutch hitters -- the ability just isn't important enough in the bigger scheme of things. Producing wins at the plate is about 70 percent a matter of overall hitting ability, 28 percent dumb luck, and perhaps 2 percent clutch- or situational-hitting skill.

Clutch hitting ability exists, more than previous research would indicate. It's about on the order of something like baserunning ability. Sometimes baserunning can make the difference between success and failure. Sometimes a hitter like David Ortiz gets a bunch of big hits down the stretch, and it makes the difference in a pennant race. Usually, though, it's the big three that prevail: Pitch the ball, catch the ball, and most of all hit the ball.

Next: Why doesn't Billy Beane's stuff work in the playoffs?