Excellent commentary !!

"Without knowing anything else that happens in a particular game, would you rather have A-Rod hit a three-run home run in the first inning of a game or the ninth inning?"

OK, so what's the answer? You compared the teams' records when they scored first with when their opponents scored first. However, the answer to your question above lies in comparing the record of the first team to score with the record of the last team to score.

your argument makes sense, sure.

But if you watch the Yankees everyday, you'd understand the frustration.

A - Rod (or as my colleague calls him "backward k Rod") has just a big of penchant for early choking as late choking.

I can not count howmany times he has come up early in a game with runners in scoring position and failed.

I also read somewhere he has batted with more runners on base then any other player in the majors.

I hate to admit it but Ortiz would probably have 200 RBIs by now, put in the same situations.

And what is the winning percentage of teams who are the last to score at the end of the game?

If you're trying to compare the utility of a team scoring the first runs of a game versus a team scoring the last runs of a game, you'd have to compare teams' win-loss records in each of those scenarios. It's obvious that if a team scores first they're more likely to win the game than if their opponent scores first (as is the case for a run scored at any point during the game), but your statistics do not demonstrate the difference between an early run and a late run.

your statistical analysis is somewhat misleading. it does not follow from your score-first, score-last win-loss records that scoring early, in and of itself, actually increases the chance of victory.

let's say that a better team is playing a worse team (one could be better or worse in the sense of possessing more or less talented players, or those players could just be having a good or bad day, etc.). to the extent that one team's offense is stronger (all else held constant), that team presumably has a higher chance to score in a given inning. of course, by this token, such a team also has a higher chance to score first (indeed, at any point in the game) as well as a higher chance to win in the end by way of this superiority.

what you've done is confuse correlation for causation; scoring early may accompany victory, but this does not mean that the former significantly influences the latter. the analysis you have provided is far too simplistic to substantiate your claim.

Very interesting points but I would like to see winning percentages from teams down in the 8th and 9th innings in close games and then see what teams are "clutch" and which players hit best in those situations. Very interesting article though.

Thank you, you are the first person to agree with what I have said.

Your article is built on the faulty proposition that "clutch" means hitting late in ball games.

This could not be further from the real meaning of clutch. Clutch is coming up with a game changing play at a moment where the balance of the game is determined by the outcome of your performance.

I watched a ball game the other night where Soriano stole a base in the last inning of a blowout. Based upon your definition of clutch, that stolen base is comparable to Dave Roberts stolen base in the 2004 ALCS.

Surely you can see the difference.

with all due respect, i think you've missed the point of clutch hitting. you are absolutely right that many a game is decided in the first or second or fifth inning. but that is really irrelevant to whether there is merit to the idea of clutch hitting. a close and late situation is just that--- close and late. the game hangs in the balance and there are limited opportunities remaining. this is a unique type of pressure which is in no way comparable to a third inning at bat. what any type of athelete does when it's now or never is what separates the guys who merely have talent from the guys with guts. i think the players and fans know this. this is why stephon marbury can score more points than chauncey billups or dwayne wade without anyone thinking marbury is the better player. that's why tv ratings are higher in the post season. there is a unique pressure involved when it's now or never, it's what makes sports exciting.

"What are the records for teams that score last?" Uh, that was more of a rhetorical question and, unfortunately, I don't know of any teams that even keep track of that. Maybe this helps though: Teams ranked according to runs scored from the seventh inning and later. As you'll see from this chart, it's pretty random. You have a lot of good teams (Cardinals, A's, Mets) in the lower half and mediocre teams (Braves, D'backs) near the top.

Not sure I agree completely with the analysis that clutch hitting is overrated. True, first inning runs are very important, but just think how many times a team scores early then gives up that lead. You still need someone to drive in those runs late in order to win. This article did make me think though. Now, as for A-Rod, no, it's not a down year, but he's no mvp this year either.

Ortiz- hitting .286 w/RISP. So much for clutch.

Batting w/RISP as of Sept. 11 (MLB stats)
AB AVG RBI Xbase
OPS
Dye 131 359 77 21 [6/1/14]
1.167
Jeter 139 396 77 17 [11/1/5]
1.088
Mauer 110 382 64 14 [11/1/2]
1.068
Ortiz 147 293 75 17 [5/2/10]
.978
Manny 124 323 68 5 [5/0/10]
1.076

Sorry that doesn't quite format right in the comments box.

First or ninth inning -- clutch hitting is about taking advantage of opportunity. Dye obviously ranks tops.

There seems to be a bit of a paradox here - your sabermetric point is taken on the value of early scoring but doesn't this also serve to highlight the importance of clutch hitting when trailing in a game? The low winning percentages when trailing could also be evidence for how few people seem to be capable of hitting when it counts?

Bear with me, I'm a Twins fan.

On August 9 Justin Morneau took an 8th inning Joel Zumaya (aka 101mph) fastball deep for his 30th home run of the season. The blast scored Joe Mauer and turned a 2-1 deficeit into a 4-2 lead. That's "clutch", however, in the sixth inning he came up with the bases loaded with one out and the Twins were leading 1-0. He grounded into a double play. That's not so "clutch". In fact striking out would be more clutch than hitting into a double play. The Twins won 4-3 however they would have been leading 5-0 if he had done it in the sixth.

Tonight in the 8th Jason Tyner hustled safely to first after a wild pitch strike out with two outs and a runner on third. The run scored giving Joe Nathan a two run lead to protect instead of a one run lead. That (in my book) is also clutch.

My point is that being "clutch" is kind of like being good defensively. There will never be a number that sums it all up. You just have to be there and watch the game and make up your own mind.

I think all the bagging on A-Rod during the season has been a bit much. The fans do love him and so want(ed) him to succeed each and every at bat. For the slump of his career the numbers are pretty good. For 25 million, pretty mediocre.

One thing about A-Rod that people seem to forget, and it's actually kind of baffling, he bats better with RISP and 2-outs then man on 3rd with less than 2 outs. Not quite sure what that means exactly regarding his clutchness. I don't think his clutchness (or apparent lack thereof) is as bad as people think.

What IS bad, or hopefully WAS bad about his slump this season can't be shown in numbers but as Jerry Steinberg posted, you had to watch him everyday to understand just how bad he was. Absolutely HORRIBLE at the plate, worse so that he carried those problems into the field. Feeble hacks at really bad pitches that conjured up images of Soriano during the 2003 World Series... ok not quite that bad but definitely not good.

The strangest thing? That magically, as almost if someone flipped a switch he appears back to "normal" even when he makes an out now, he looks 100x better.

I think everyone is getting too caught up in thinking clutch is ONLY late in the game. Clutch hitting can come anytime in the game. When a team is putting pressure on another team in the first inning it could take a clutch hit to get it done and change the face of the game. It is not black and white where late in a game is clutch and early isn't or early runs are more important than late runs. A clutch situation can come at any point of a game where there is pressure to change the game. People who rip A-Rod for not being good late in the game do not watch the Yankees. If you watched them you would know he has struggled just as much early in the game as late. He is very bad during his first at-bats of the game. But his struggles are blown way out of proportion anyway. I agree with people who point out specific occurrances of clutch play, its not stats but instead plays that happen at a certain point of a game that you just know was a clutch play. Let the media have their definition of clutch they can't even watch every game, they go by box scores and "late and close" stats to decide for us what is cluth and what is not.

Nice work, Jake. Clutch hitting is a myth.

Real Yankee fans, the ones who remember Celerino Sanchez and Torey Lovullo manning third, aren't the ones hating on A-Rod. The Yanks will be fine, frontrunners and Shea transplants notwithstanding.

i agree whole-heartedly with the person who said that "clutch" is difficult to measure in numbers. clutch is like pornography according to that supreme court judge. difficult to define, but i know it when i see it. incidentally, i'm a twins fan too. is morneau clutch, or what?

I'm a diehard Yankee fan for 30 years now. I try to watch almost every game And we do support A-Rod. But he has to come through more in big situations. It seems like everytime there are men on first and second with one out he'll do the worst thing that can happen. Hit into a double play or man on third one out or no out and he'll strikeout. And that seems like it's the norm now. I know you can't be Superman all the time but for him to be the so-called best player in the game he has to come up bigger than that more often.

This is why Win Probability Added should be used instead of BA in close and late situations to determine "Clutch" players.

question: which do u think is more exciting: a blowout or a clutch homer in the bottom of the 12th? everyone loves to see their team win, but it makes it more fun to watch. i have to agree with jerry down there about ortiz n a-rod: the former thrives in those situations. thats what an MVP should be like.

Clutch hitting is as important as early hitting when the big games are on the line, especially in the playoffs. When Boston was beating the yankees, in what was a major disaster for the Yankee organization, A-Rod could not come up with freeking sacrifice fly to score a run. I am sorry if you don't appreciate the importance of clutch hitting, which in definition means hitting when it really counts. Just look as the way as Derek Jeter or Paul O'Neill in his heydays approach each at bat, they are REAL YANKEES....Unlike A-Rod who flops when it really counts, see his stats for the last two years on playoff....THAT WHAT CLUTCH IS ALL ABOUT.

Check the last two playoff years, and see how Jeter approaches every at bat.....A-Rod is not a real YANKEE

If you want to say David ortiz is clutch thats fine, he has proven time and again that you don't want to face him in the bottm of the ninth when he is the winning run, but he doesn't quite stack up to certain yankees in other "clutch" situations like bases loaded, RISP, and RISP with 2 out. In those situations in that order Ortiz bats .333/.293/.279, A-Rod bats .471/.291/.315, and Jeter bats .455/.396/.368.

Now if A-Rod or Jeter has a hit in the 5th inning that puts the yankees ahead and they win the game is that any less "clutch" then an Ortiz walk-off homerun in the bottom of the ninth? To me they aren't, they both win the game and that's the bottom line.

Santana MVP, No body compairs.

You people just need to lay off A-Rod. Assuming he returns to form after having this "off" year - he will be well on his way to being probably the best player to ever play the game. (At least before Pujols gets done). Why aren't you Yankee fans picking on Giambi - that batting average in the .250s isn't too becoming, but we don't hear anything about that. Leave it alone...

Does everyone seem to forget that he single handedly help push the yanks past the twins in the playoffs a couple years ago. Or the fact that barry bonds was "unclutch" in the playoffs until he hit 40yrs old and everyone then forgot about it. Or John Elway used to be labeled as couldnt win the big one until he finally did at the VERY end of his career. Cut A-Rod slack, he will eventually come up big I bet much sooner than these other guys did. Just because AJ Pierzynski came up big with the Sox last year in the playoffs doesnt mean hes a better "cluth" hitter or that I would rather have him up than AROD in the 9th of a close game.

A-Rod is a very good player who at times is great. However to say he's "arguably the best player in the game" is total nonsense. Which I guess is based on having a few great years on last place teams. In actuality he is the most overrated player in the history of baseball.

Pepe, please shut up with your "Real Yankee" BS. That's one of the worst media-driven catch phrases of all time.

No such thing as clutch...

As clutch as Ortiz is, he can't pitch and he can't play defense so who cares if he can hit a grand slam in the bottom of the ninth if you are aleady down 10 runs. He is a great hitter, not a great player, there is a difference.

this was a great commentary. people just run of at the jibs when they hear something. i'll take a-rod on the cubs no matter what inning or situation. ALEX RODERIGUEZ HATERs will never go away. so what he makes a lot of money, you would have taken it too.
sincerely,
sick cubs fan

If someone can step-it-up for clutch situations, then why does Ortiz choose not to hit monstrous home runs all the time. It's ridiculus to say that clutch hitting exists more than a tiny amount which is changing the TYPE of successes. e.g. A walk is as good as a single with nobody on, so try to be more patient. This effect is tiny however, less than 5%. I suggest reading Baseball Prospectus' new book 'Baseball Between the Numbers,' that is where I got this from. I didn't read it for stats, just as a baseball fan, and it is very good.

Interesting article, but your use of statistics is questionable! It reminds me of my dentist who quotes statistics showing that people who floss are less likely to get cancer. Of course, because people who floss generally are more health-conscious people and are thus less likely to get cancer. (How many smokers do you know that floss???) This is correlation, not causation.

Your argument suffers from the same logical flaw, albeit in a less conspicuous manner. Of course the team who scores first is more likely to win, because the team to score first is more likely to be the better team!

Using your own statistics, the best team in baseball (the Mets) have scored first in 57.7% of their games (82 out of 142 games). The worst team in baseball (the Royals) have scored first in 41.7% of their games (60 out of 144 games).

Make sense? Even history majors should have to take an intro statistics course!!!

I am willing to bet if you ran the same statistics for the team who scored last, you'd find the same correlation - that the team who scores last wins more often. This will be true because the team who scores last is more likely to be the better team.

Extraordinarily interesting!!!
This is the FIRST time I have seen someone come up with this simple formula, and VOILÀ, it shows us some interesting development.

Now, as far as being a History Major and going far with the MATH. You not only piqued my interest, you seem to have made MATH fun again. thank-you Jacob

Maybe we should have MORE History Majors teaching MATH! Keep preserving that sports history...We need it!

Baseball has actually taught me a lot about MATH! Jacob mentioned the fact that he is/was a History Major and that he had served up the best math he could give us!

Well, you know what? It worked for me (until I started reading further down the column here and found out some OTHER very interesting and intriquing commentary and further questions) and for my purposes. I was simply AMAZED at those statistics of the teams winning % while scoring first, and the difference of their % for scoring last.

Sports make learning MATH exceedlingly FUN! NOW GO OUT and PRESERVE MORE SPORTS HISTORY!

How do you sign out of the blogger.com site? Hey, it's my first day on this thing...

It's not like yahoo answers, but it certainly comes close.

I find this all extraordinarily interesting!

Ciao for now...Until next time, or unless I cannot get out of here by simply signing out. I thought this would be easy, but it is NOT. And, I am not a Math Major either, but a Communication Major. I would like to communicate my way out of here now...I need to go to bed. It is 00:26 here in Europe. Somebody help me stop this...I must do this with more prudence I can tell. I am going to have to train myself, and then PACE MYSELF. God bless all the rest of you guys making all those fascinating comments and questions! Hats of to Jacob for getting it going (but then again, it is his JOB and he gets paid to do that). Good for him indeed!

Yankee fans who trash A-Rod, please, take your head, put it in the toilet, and flush. You. Are. Idiots. You do NOT deserve to even watch Yankees games. I think I speak for all real Yankees fans and all generally smart baseball fans when I repeat: You Are Idiots.

The A-Rod haters, who carp about his lack of 'big hits' are the off-spring of the very same kvechers who used to carp about the Mick before Roger Maris had the 'nerve' to hit 60 HRs to the Mick's 54, and Mantle was 'home grown.'
Prior to 61, all I heard, and read, was how Berra was more clutch, and how Mantle was no Joe 'D', yada yada yada, etc, etc. After 61, the Mick was as a god to the very same fans who used to deride him.

Unfortunately for A-Rod, he is not 'home-grown'; so, he likely never will gain full acceptence no matter how good his Yankee numbers are. That is the price one pays for playing in NY, for the Yankees, and for making all that $.

I remembered one thing that I wanted to tell you, something which even made me laugh out loud.
Some guy from PETA objected to the terms 'cat' and 'dog'. He wants dogs referred to as 'canine-americans'. What can you say. I read it in the Daily Woof.



I have come to the unfortunate contusion that stats suck. Let me qualify this. I read Leo Durocher’s autobiography and he talked about the way he put together a team. He sought out players like Eddie Stanky who was famously described as : He can’t hit, he can’t field and he can’t run. All he can do is beat your brains out. Durocher was a constant winner specializing in taking morbid franchises and making them contenders. He did it by knowing his players and what they were capable of and making a team of them. I know that there are some very successful general managers who have some very obscurely structured stats all influenced by Bill James and his ilk. I do know that Durocher would love to have an A-Rod on his team if for no other reason than he plays a beautiful game of baseball. On the other hand he would not like an Ortiz because he hated big fat slow guys. This was in the day before the designated fat guy.
I saw somewhere where the highest rated clutch hitter of all time was Eddie Murray and his stats are spectacular. He’s in the Hall of Fame. Yet I remember him playing and he reminded me of the ‘54 Orioles when every game ended with the winning runs left on base by a mighty strike out or a game ending double play. I saw him do stupendous things in lopsided games where it didn’t matter and the magic words “mathematically eliminated” seemed to drive him into hitting streaks and while everyone else was playing out the string Murray seemed to be getting his second wind. The phrase ‘late season salary drive’ seemed coined for him. I saw Murray play and I saw Gil Hodges play and there’s no way I would take Murray over Hodges (whose exclusion from the Hall is one of baseball’s a grave injustices. I guess the stats tell a different story but one day the last man who saw Hodges play will have died and Murray will have the honors and Hodges will be trivia (4 home runs in a game). Long live stats.

Re: Anonymous 6:11 p.m.
I addressed the "scoring last" take in an earlier post.

OK, so the Royals have scored in 41.7 percent of their games, yet they have an overall winning percentage of 37.9. Similarly, the Devil Rays have scored first in more than half of their games yet have a winning percentage of 39.6. That's why I pointed out the improvement of the bad teams when they score first.

Any discussion of the NL MVP race that fails to mention Jose Reyes is incomplete. While the world is fascinated by Ryan Howard's HRs, Albert Pujols' consistent excellence, and even the overall dominant performance by teammate Carlos Beltran, the runaway top team in the NL would be significantly hamstrung without their leadoff man. Reyes' impact on his team is the single most compelling of all the candidates on their teams--simply, when Reyes gets on Base, the Mets score--and score a lot. And win. As the banter exploded around the AL's alleged superiority-even over the Mets-all perspective was lost. The Mets have an "AL-like" lineup. But unlike these AL teams, the Mets have Reyes-the most dominant leadoff man since Rickey Henderson in his prime. When the Mets traveled to Boston earlier this summer for three against the RedSox, they were without Reyes. The result-a 3 game sweep--which helped to foster the AL dominance theory. Value like this is obvious, but when seen on an everyday basis, breathtaking.

Assuming that I did this right, here are records for teams broken up by whether they scored last (for 2006 season, through games of September 13th):

Overall record of teams that score last: 1443-785

Team: record when scoring last (when opp. scores last)
Arizona: 42-31 (31-47)
Atlanta: 53-35 (20-44)
Baltimore: 40-33 (24-51)
Boston: 50-32 (28-43)
ChiCubs: 33-31 (26-58)
ChiWhiteSox: 56-18 (29-45)
Cincinnati: 49-21 (24-52)
Cleveland: 50-30 (18-48)
Colorado: 40-27 (29-50)
Detroit: 54-15 (36-44)
Florida: 45-23 (31-51)
Houston: 42-22 (33-52)
KansasCity: 41-30 (17-61)
LAAngels: 52-27 (28-42)
LADodgers: 44-22 (35-47)
Milwaukee: 51-29 (16-52)
Minnesota: 55-19 (31-41)
NYMets: 65-25 (26-34)
NYYankees: 55-16 (35-43)
Oakland: 57-23 (27-40)
Philadelphia: 50-29 (30-47)
Pittsburgh: 43-34 (17-56)
SanDiego: 56-25 (19-46)
SanFrancisco: 48-28 (27-47)
Seattle: 47-30 (22-46)
St.Louis: 50-24 (27-45)
TampaBay: 32-27 (25-62)
Texas: 48-28 (30-44)
Toronto: 52-21 (24-49)
Washington: 43-30 (20-56)

And to quickly followup:

Teams, Win Pct when scoring first, Win Pct when scoring last

AL Teams, 0.647, 0.664
NL Teams, 0.671, 0.634
All teams, 0.660, 0.648

So scoring first is huge, and scoring last is huge. I'd guess that scoring in the middle is pretty huge, too.

This is the most intelligent baseball article I have read in a very long time. You made an excellent point and actually supported it with research (about as rare as hen's teeth in baseball writing). The only disagreement I have is that you actually didn't go far enough. "Clutchness" doesn't exist at all. It is a myth built on isolated events and statistics on very small samples (mostly World Series and other big-publicity games). In other words, the reputation of "clutch" hitters are statistical anomalies from very small samples.
Gary Graul
Irvine, CA

First of all, let me just say I enjoyed reading just about every comment here. Many good points have been made regarding this article.

But, I was thinking, if the article main thesis deals with the myth of "clutch", wouldn't the fact that scoring last is just as important (as scoring first) support the notion that "clutch" is a myth? Everyone always focuses on late-inning heroics when "clutch" is the topic. Taking that into consideration, I'd conclude that Jacob Luft's statistics (presented here) are sufficient for the article's primary subject.

Although, the numbers are not necessarily appropriate when contemplating the question "Given the choice, would you rather have your team score first in a game or last?". However, when teams score first they are at 0.660; 0.648 when scoring last. And, based on that, it is still slightly better to score first.

Also, another thing to think about is sometimes a team will score first and last. Furthermore, how often does the team that scores first also score last? I would think that the team that scores first usually scores last.
Additionally, it likely that a team's run production is higher when they score first. Although, I doubt the same can be said for teams which score last.

Anyhow, it would be very hard to prefer scoring last over scoring first.

Hmmmmm - doesn't it really work that the team who scores the most often during the 1400 plus innings in a season (in addition to winning more) will generally be scoring more during all innings, including the early innings. This means that they are more likely to score first than less run-producing teams. ALso, in a sport where teams more or less randomly scatter runs over all of their innings at a given rate and where scoring 0 for 9 innings is a possibility, isn't this like saying I'll give 1 team a run or two headstart for this 9 inning stretch and see if that matters?