This is our old blog. It hasn't been active since 2011. Please see the link above for our current blog or click the logo above to see all of the great data and content on this site.

Keeping Score: The Most Important Leadoff Skill? Not So Fast – NYTimes.com

Posted by Neil Paine on April 28, 2011

Keeping Score: The Most Important Skill for a Leadoff Hitter - NYTimes.com

Some basic sabermetric thoughts on leadoff hitters, with a little help from The Book.

50 Responses to “Keeping Score: The Most Important Leadoff Skill? Not So Fast – NYTimes.com”

  1. Doug Says:

    Neil, re: "a team’s three best hitters should bat in the Nos. 1, 2 and 4 spots", what about the #3 spot?

    Evidently, the Blue Jays should read this - they're batting arguably baseball's best hitter in the 3 hole. Presumably the 4 spot is where he really belongs.

  2. Neil L. Says:

    Neil,

    Well-written. The traditional baseball "book" is so different from the real sabermetric "book" ala Tom Tango and company.

    Managers are still mesmerized by speed at the top of the lineup. I'm not sure if its all attributible to Rickey Henderson but I agree with your premise that lineups are made out based on SB not OBP!

  3. Ben Says:

    @1, I believe the reasoning is that the 3 spot sees too many no one on, two out situations in the first inning of games throughout the season.

    Figuring a .350 OBP from both the 1st and 2nd hitters, the odds of seeing that situation in any given game is .650 * .650 = .4225 (excluding GIDPs) which equals about 68 worst case run-producing scenarios for your 3 hitter in his first at-bat of the game over the course of a season, whereas the 1 & 2 hitters will never see this scenario until at least their 2nd at-bat, and the 4 hitter will only see it if someone ahead of him scores & the other two make outs (I'd guess this happens only 5% of the time off the top of my head).

  4. Justin Bailey Says:

    I think another interesting thing about Henderson was that his power was one of the things that made him such a great leadoff man. I think it contributed to his high OBP--pitchers are more likely to walk you if you're a legit threat to put one in the seats, right?

    I've often wondered if the Giants would have scored more runs in 2001 if they had let Bonds lead off. A lot of his homers were solo shots that year anyway, and with his super-Ruthian walk rate he would have been the tablesetter to beat all tablesetters.

  5. Neil L. Says:

    @1 @3
    Ben, Bautista's AB's this year are a perfect example of what you are saying. Look at his low RBI vs HR ratio and look at his number of AB with no one on base.

    @4
    Justin, absolutely about Barry and about Jose Bautista this year also. The rub is that it takes a courageous manager to go against traditional wisdom and put a power hitter in the lead-off spot.

  6. John Autin Says:

    Rickey's power was definitely an asset. But I don't think it was much of a factor in his walk rate. In his first 4 full seasons, before his power developed, Rickey averaged 10 HRs and 116 walks per 162 games. In his next 4 seasons, his HR average soared to 26, but his walks went down a bit, to 108.

    Rickey walked because that was part of his game plan. He hit out of a crouch, and he worked the count. Odds are he would have continued to average 100 walks a year even if he'd never developed power.

  7. Johnny Twisto Says:

    Seems to me like the 2nd spot is really the most misused throughout baseball. Teams seem to try so hard to find a guy who can "handle the bat," instead of just putting one of their best hitters there. Sometimes you get lucky and Derek Jeter fills that spot, a player who has attributes of a traditional 2-hitter but is so good at those skills that he's really one of the best hitters in the lineup. Usually it's someone like Placido Polanco, a useful player but in most seasons a good team will have several more productive hitters whose ABs he's stealing.

    A few years ago I was playing a simulated baseball game and would bat my best hitter 2nd. One might think that he doesn't get enough RBI opportunities but in the AL he does -- he would end up with 120-130 RBI a season, and as many runs. In the NL there might be too few RBI opps, unless you bat the pitcher 8th as well.

  8. John Autin Says:

    I think the 2001 Giants would have scored more if they'd had a good leadoff man, since they got a .315 OBP and .707 OPS out of that spot, plus 12 CS against 17 steals. They actually gave 18 leadoff starts to Shawon Dunston and his .280 OBP.

    But I don't know that hitting Bonds leadoff would have been better than using any other good hitter. While it's often noted that Barry averaged less than 2 RBI per HR that year, he still had 137 RBI, 4th in the league, despite 3 major drags on his RBI:
    -- weak leadoff men;
    -- frequent walks with a base open; and
    -- 37 HRs by SF's #2 hitters, mostly Rich Aurilia.

    He did hit 27 HRs with at least one runner on. If you think of that as a percentage of his total HRs, it's easy to forget that 27 HRs with runners on is a LOT. It's just 1 less than Sammy (64 HRs, 160 RBI) Sosa hit with men on that year. Their walk rates aside, one big reason that Sosa had 23 more RBI than Bonds is that he had 60 more PAs with any runner(s) on, and 54 more PAs with RISP.

    The Giants would have been better off having any of their good hitters leading off -- which was a much bigger issue than figuring out exactly which one of their good hitters would have been the optimal leadoff man.

  9. John Q Says:

    @7 Twisto excellent point, I agree with you 100%. It's starting to change a little but teams are still batting a guy who can "handle the bat" rather than one of their best hitter in the #2 spot.

    One of the famous examples of this and it's something that might have cost the Expos two trips to the WS in '79 & '80 was when they batted Rodney Scott #2 with a lineup. Scott batted .233/.319/.294 in 1979 and .224/.307/.293 in 1980. The Expos only finished two games out in '79 and one game out in '80.

    You could look at the collapse of the 2008 Mets and blame Randolph/Manuel on how they used some of their #2 spot when Castillo was either out or batting in another part of the lineup.

    Argenis Reyes hit .218/.259/.245
    Endy Chavez hit .267/.308/.330
    Nick Evans hit .257/.303/.404

    In 2007 Paul Loduca hit .272/.311/.378 and spent most of the first half of the season in the 2 spot. And Endy Chavez his .287/.325/.380 overall and spent some time in the #2 spot.

    Even going back to 2005, Willie Randolph really underutilized the #2 spot.

    Kaz Matsui: .255/.300/.352
    Miguel Cairo: .251/.296/.324

    And to make matters worse he batted his best offensive player (David Wright) in the 6th & 7th spot for half of the season. And it didn't help matters that Reyes had a .300 on base percentage.

    I remember having the '87 Mets on Strat O Matic and I always batted Daryl Strawberry lead-off. Really in retrospect that's where he should have batted. Not only did he have the best on-base percentage on that team, he was also the fastest player and led the team in stolen bases.

  10. Dave Says:

    Doug Says:
    April 28th, 2011 at 6:48 pm
    Neil, re: "a team’s three best hitters should bat in the Nos. 1, 2 and 4 spots", what about the #3 spot?
    ****
    I always thought your best hitter (Pujols, Bonds, CarGo, Bautista, etc) should bat 3rd and it's pretty easy to see why. The 3rd position, over the course of a season, will get more plate appearances than the 4th spot. I've been able to find different numbers but lets say that it's 30 more. Do you want those 30 appearances to go to your best hitter or not?

  11. Johnny Twisto Says:

    Dave, if more plate appearances is the only consideration, then why wouldn't you bat your best hitter first? Obviously other things matter. That said, while I read The Book and thought its findings made sense, I would never fault a team for batting its best hitter 3rd. If it's the "wrong" decision, it's not nearly as wrong as putting a .280 OBP speedster leadoff.

  12. Dave Says:

    @11, Johnny, batting 1st would seem to be obvious but (IMHO) batting your best hitter 3rd, usually a power hitter, means he is coming up with runners on base...unless the manager likes the "speedy" lead off guy and "pesky" 2 hole guy who have a combined OBP below .600.

  13. John Autin Says:

    @10, Dave -- In the current era, average PAs per season decline by about 17 PAs for each spot lower in the order. So the difference in PAs for any two adjacent spots is pretty minor.

    I've long agreed with Johnny Twisto's point @7; the choice of #2 hitters is one of the last bastions of old-school strategy. Number 2 hitters, as a group, generally score fewer runs than #3 hitters, which makes no sense at all.

    What bugs me almost as low-OBP guys in the #2 spot is the general insistence that he be a singles hitter. So far this year, 7 teams have a SLG of .300 or lower from the #2 spot, and the median is .385. Ten teams have no more than 1 HR from the #2 spot; just 9 teams have 3 or more HRs there.

    And while I wouldn't want to make too much of this, 12 of the bottom 13 teams in OPS from the #2 spot are at or below .500.

  14. John Autin Says:

    Correction in next-to-last paragraph @13: "What bugs me almost as much as low-OBP guys in the #2 spot...."

  15. BSK Says:

    SI did a piece a few years back looking at the "ideal" current player for each spot. They looked at who would be the best leadoff guy or the best two-hole hitter, all based on the perception of what those hitters are *supposed* to be like. They had Ichiro leading off, Jeter in the two-hole, and on down the line. Now, if they were choosing the best hitter who hit predominantly in a certain spot, most of their choices would have made sense. But if they were constructing an ideal lineup, wouldn't you simply slot in the best 8 or 9 guys? It just seemed to confirm the nonsense that certain batting positiosn were supposed to behave in certain ways. Booooo...

  16. BSK Says:

    "Along the way, his feats on the base paths fueled the stereotype that speed is the leadoff man’s defining tool."

    I'm curious if this is actually true. Did the speedy leadoff man stereotype originate with Henderson? What did the prototypical leadoff man look like before him?

  17. Lawrence Azrin Says:

    @16/BSK Says: "...I'm curious if this is actually true. Did the speedy leadoff man stereotype originate with Henderson? What did the prototypical leadoff man look like before him?

    It probably originated with Luis Aparicio in the NL in the late 50's, and Maury Wills in the NL a little later. Stolen bases had gone down steadily in both leagues since the start of the live-ball era, then those two came along, stealing 30/40+ bases every year, and the perception was that they were leading the "stolen base revolution".

    Thus, by extension, the stereotype of the small speedy middle infielder as the ideal leadoff hitter was developed. Never mind that a good OBA far outweighs in value 20/25 SB a year...

  18. John Autin Says:

    @16, BSK -- I think it could be argued that Henderson's career did fuel the stereotype, though obviously it didn't start with him.

    The first national attention Henderson garnered was for base-stealing. When he had 100 SB in 1980 (his first full year), it was just the 3rd 100-steal season in modern times. Two years later, he shattered Brock's season record. And in 1991, he became the career SB leader.

    Along the way, thoughtful observers noted that, while the steals were nice, what really made Rickey the best leadoff man (ever) was his high OBP. Arguably, then, by concurring with the speed crowd that Henderson was the best, the saberists may have helped perpetuate the stereotype.

    When someone shouts "yes, but for a different reason!", most people just hear "yes."

  19. BSK Says:

    JA and LA-

    Thanks. Just to clarify, I did not intend to dispute the assertion. Rather, as someone born in 1983, all I've ever really know was the speedster stereotype, so I didn't have the historical perspective necessary to know what preceeded it.

    JA, I think your last point is particularly salient and one too often ignored in such "debates". Whether or not Henderson was the best leadoff hitter ever isn't really much a conversation. What made him such is. Unfortunately, the "stolen bases" folks have had a major head start in shaping public perception of the debate and the public perception is more than just that, since it also has had a major impact on actual team construction and game management.

  20. Cyril Morong Says:

    Bill James talked about this in the first "Historical Abstract." He thought that maybe in the future (speaking in 1985) that leadoff men would combine the "LuApps" qualities. Get on base like Luke Appling but steal like Luis Aparicio. It seems like we are still trying to get this message out there.

  21. Cyril Morong Says:

    Probably what a team wants to do is create the lineup that will maximize scoring. So you have to set the lineup to do that and you can't just worry about one slot.

    But if you are just worried about leadoff, for the moment, you need to take everything into account (and the rest of your lineup is pretty traditional). If one guy has a .390 OBP but never steals, and another guy has a .385 OBP and steals 70 with 10 CS, you might want the 2nd guy. It might be the 2nd guy who you put leadoff. We need to discover the trade off between stealing and OBP for the leadoff man.

  22. Cyril Morong Says:

    This link has my regressionon lineup slot. I am not saying these are necessarily a guide to creating the best lineup, but if you are going with a traditional lineup, the last regression gives the run value for SB, CS, hit%, walk% and extra-base% (with walks + AB being the deonominator for the last 3 and per game for the first two).

    http://www.beyondtheboxscore.com/story/2006/2/19/192135/078

    You will see that the SB run value is not highest for the leadoff man. 4 slots have a higher value. But the negative value of a CS is only worse for one other slot. So the run value for SB and CS for the leadoff man are

    SB .202
    CS -.558

    But the leadoff slot had the 2nd highest value for hit% and the highest value for walk%. I think that means the leadoff man needs to be successful 73.42% of the time.

  23. Cyril Morong Says:

    I also did a post a few years ago called "Who Are The Good Leadoff Men?"

    http://cybermetric.blogspot.com/2008/06/who-are-good-leadoff-men.html

    It seems obvious: Hitters who are fast and get on base alot. You also probably don't want someone who hits alot of HRs, since you want those guys to bat with runners on. So I tried to devise a stat that would capture this. Here it is:

    (2B + 1.25*3B - HR + SB)/outs

    In other words, how many times a player gets into scoring position per out. Since triples are worth about 25% more than 2B's according to run expectancy tables, I multiply them by 1.25. By dividing by outs, the ability to get on base is taken into account since if you make an out you don't reach base. Also, outs include caught stealing. By subtracting HRs I am saying that guys that hit alot of HRs, even though they may have other good leadoff traits, are "penalized" here, since they might be better suited to batting lower in the order. But I also ran the numbers without subtracting HRs (the correlation between the two different formulas for all the players in the study was about .86).

  24. Michael E Sullivan Says:

    also, in general stolen bases are far far overweighted in the public's mind. Henderson wasn't a great because of his speed and base stealing, even though that might have been his most defining skill. That was a fairly small portion of his overall value.

    Even somebody who stole bases in Ricky quanitities, if they got caught significiantly more often, those stolen bases aren't worth much. Brock, the guy whose record Ricky broke, added a bit less value on the base paths than the guy who was in second by 250 SB back in 1980 when Henderson broke in: Joe Morgan, who got caught a lot less often. Tim raines didn't end up in second place on the stolen base list with only 808 career. But he *is* in second place on the rbaser list, because he rarely got caught.

  25. Neil Paine Says:

    #3 is correct - the 3-hole sees PAs with fewer men on and/or more outs than 1, 2, or 4, so he actually should be your 4th or 5th-best hitter.

    And Johnny really nails it in #7 w/ regard to #2 hitters. The run values per event are higher almost across the board for #2 compared to #3 hitters (meaning a single is, on average, more valuable from the 2 hole than the 3 hole, etc). Using a bat control guy who isn't one of your better hitters is a major inefficiency.

    Btw, teams were definitely using speedy guys at leadoff way before Rickey. But Rickey is the rare all-time great basestealer who was also an all-time great OBP guy.

  26. Neil Paine Says:

    Great links, Cy.

  27. Cyril Morong Says:

    You're welcome. Thanks for tackling this issue

  28. Johnny Twisto Says:

    (2B + 1.25*3B - HR + SB)/outs

    Bill James had a junk stat like this as well. Weighting things which are more useful early in the inning against those which are more useful with men on (power). I think it may have been something even simpler like (BB + SB) / (TB - H). You'd prefer a high ratio to get things started and a low ratio to bat later in the inning.

  29. BSK Says:

    "It seems obvious: Hitters who are fast and get on base alot. You also probably don't want someone who hits alot of HRs, since you want those guys to bat with runners on."

    Cyril, I have to take a bit of an issue with this. If given the option of a leadoff man who puts one over the fence versus a leadoff man who draws a walk, you'd take the former every time. The former is a guaranteed run where the latter isn't, no matter what it does for the batting circumstances of your later hitters. I'm thinking back to McCarver's "I can't believe a leadoff HR leads to more multi-run innings than a leadoff walk." Now, if your leadoff guy is also your best HR hitter, than you would be right that this likely isn't the ideal scenario. But the point is not to get your best bats up with runners on. The point is to score runs. If your team does its best run scoring by setting the table for your heavy hitters, so be it. But if your leadoff man can drive himself in, there is no reason to discourage him from doing so.

  30. Cheese Says:

    For the cleanup spot, if your best power guy walks a lot (ie Bonds, Ruth, Williams) should they go here, or the #2 spot?

  31. Cheese Says:

    er. to clarify by 'best power guy', i meant 'best hitter with power'. Cheers.

  32. Cyril Morong Says:

    BSK

    i definitely understand what you are saying. That is why I also did it without subtracting HRs. But given that a team has only so many players to work with, your highest OBP and SB guy might also be your best HR hitter, and, depending on who else you have, you might not bat this guy first.

    Cy

  33. Tomas Says:

    Off topic, but these are both excellent 'on this date' from the bbref bullpen:

    Bullpen:Today in Baseball History

    1902 - Baltimore Orioles infielder John McGraw is hit by pitches five times, but home plate umpire Jack Sheridan refuses to allow him to take first base. In the ninth inning, McGraw is hit for the last time and sits down in the batter's box in protest. American League president Ban Johnson will suspend McGraw for five games.

    1913 - Wearing the uniforms of the Chicago White Sox, the Cincinnati Reds lose to the Chicago Cubs, 7 - 2, at West Side Grounds. Cincinnati forgot to pack uniforms and has to don those of their opponents' crosstown rivals.

  34. Jimbo Says:

    When Bonds was the beast, I always thought the Gaints just needed ONE other high OBP guy and they'd be set.

    If Bonds batted 2nd, behind a player with a .400 obp, then BOOM, he is pretty much guaranteed to bat with a man on at least 40% of the time. And there is no reason at all for that leadoff hitter to have speed, he should NEVER be stealing unless Bonds was taking the day off. Havnig 2000-2004 Bonds bat with a man on 40%+ seems so strong I wouldn't even care about the rest of the lineup, though the next best hitters should probably bat 3rd, 4th, and 9th.

    It seemed when they batted him 4th, 2 of the guys in front of him usually had low OBP's, and the other guy would be batting into DP's constantly (Grissom). Not only would Bonds be batting with nobody on in the 2nd alot (although his high obp to lead off the inning had its pluses), he also had to bat immediately after running in from left field. As the older player that he was, I could see this possibly having drawbacks.

    I also think batting 4th has a negative effect on the hitter. He gets ready in the 1st inning to hit, but there is alot of uncertainty as to whether he will actually hit in that inning. In Bonds case (I was a big fan of his) I imagined if he batted 3rd, he could have better approach for his first AB, always knowing exactly when that AB would occur. So for this reason I propose there is another advantage to batting your best hitter 3rd instead of 4th. He can get himself into a better mental/physical rhythm, because he knows exactly when he will hit.

    Which brings me to a question I've never seen asked. Is there any research data showing how well catchers bat when leading off an inning after catching for the defensive half? I just imagine taking the gear off and all that right before hitting would possibly be distracting from your hitting approach. And the same question about outfielders being affected by having to run in and then leadoff an inning.

    I've never seen this discussed.

  35. Neil L. Says:

    @22 @23
    Cy, particularly enjoyed your blog on good lead-off men. Hope to read a lot of your posts here.

    @28
    Johnny Twisto, I hadn't heard the phrase "junk stat" in a long time. What differentiates a legit stat from a junk one, in your view?

  36. John Q Says:

    @34 Jimbo,

    That's some interesting points about Bonds. Bonds batted third in 2001 and Aurilia batted second with a .369 on base which is fine. But what really didn't make any sense is that the Giants batted Calvin Murray .245/.319/.356 and Marvin Benard .265/.320/.439 in the leadoff spot. They were only 2 games out of the division. They probably win the division if they moved Aurilia into the lead-off spot and Bonds in the second spot and Kent third. Heck, they might have won with J.T. Snow (.370 on base) in the leadoff spot instead of Benard & Murray.

    In '02 it was an odd mix of Shinjo, Bell, Goodwin in the lead-off spot and none of them had a good on-base. Things didn't really get better until they got Kenny Lofton. Aurilia had a pretty terrible year as well. That '02 Giants team is a strange team for a team that almost won the WS. Basically it was Bonds & Kent with a bunch of good defensive guys and a good
    bullpen.

    '03 had Durham and Snow in front of Bonds with good on base percentages but Grissom spent a lot of time in the lead-off spot with something like a .320 on base.

    '04 had Durham and Snow and Tucker all of whom had good on base percentage in front of Bonds.

  37. Johnny Twisto Says:

    Umm...like Justice Stewart, I know it when I see it?

    To be clear, I don't intend it to be pejorative. It's not that it's not a "legit" stat. I guess I see it as numbers thrown together to describe something, without any rigorous science behind it. Bill James came up with lots of junk stats and described them as such. I think quality starts is a junk stat, but nevertheless I refer to them.

  38. Neil L. Says:

    @37
    JT, gotcha. So would gamescore for pitchers be another such junk stat? After all it is used in here which confers a certain legitimacy.

    I agree that there are lots of gimmick stats, but what about something like James' runs created. Where do you draw the line, objectively? Saying that you know it when you see it is sort of a non-answer.

  39. Neil Paine Says:

    Bill James used to say certain stats had achieved "the power of language." They were numbers, sure, but at the same time they told a story about something tangible, something bigger than themselves. For instance, runs created is a simple offensive metric derived from hits, walks, total bases, and at-bats, but it also has the power of language because of what it's a proxy for -- the literal number of runs a player contributed to his team's total. That fact in and of itself conveys meaning to a baseball fan beyond the context of a formula. If you can understand replacement-level theory, WAR is another stat with the power of language.

    I think a "junk stat" has the opposite of linguistic power: it is nothing but a number, and thereby means nothing outside of the very specific context of its own formula. Like Johnny said, it's not necessarily pejorative. It's just a statement in part about how rigorous the science is behind the formula, and in part about how much the stat transcends the narrative limitation of raw numbers.

  40. Neil L. Says:

    @39
    Neil Paine, wish I could use language like you do. {~sucks up~}

    Perhaps all the straight-forward, intuitive, language-transcendent statistics are gone and already taken but there is something to be said for simple elegance in a baseball performance metric, either offensive or defensive.

    Cy's "lead-off hitter efficiency index" or whatever you want to call it has the appeal of simplicity. That is to say it is "graspable" by casual fans, let alone ML managers and GM's.

    My point is that simplicity in a statistic is as alluring as complexity, even it is at risk of being relegated to junk status.

  41. John Autin Says:

    @34, Jimbo -- For his career, Bonds hit almost the same in the #3 and #4 spots:
    #3, .306 / .457 / .670 = 1.127 OPS
    #4, .317 / .503 / .670 = 1.173 OPS
    He had a higher walk rate (especially IBBs) hitting cleanup, but a lot of that was just timing; he batted almost exclusively #4 from 2003 on, after Kent left and after he had firmly established himself as the deadliest hitter of a generation (at least).

    He did have his record HR season mostly in the #3 spot, posting a 1.370 OPS in that position in 2001. And in 2002, splitting between the two spots, he did better at #3 (though his 1.301 OPS at cleanup was hardly embarrassing).

    But in 2003-04 he spent the vast bulk of his time at #4, and was still utterly devastating.

    I can't see that where he batted made any difference. And I really can't see a big challenge in leading off an inning after jogging in from LF -- not in theory, and definitely not in the stats. Here are his OPS figures when leading off an inning from 2001-04:
    -- 2001, 1.421
    -- 2002, 1.257
    -- 2003, 1.227
    -- 2004, 1.199
    Bear in mind that he's not getting any IBBs in those situations, which tends to bring down his OPS. His HR rate when leading off in those years was astronomical: 54 HRs in 495 PA, 392 AB. That's significantly better than his HR rate when not leading off an inning in those years, even when comparing on a per-AB basis (not per-PA).

  42. John Autin Says:

    Neil Paine transcends the narrative limitation of raw numbers.

    Or maybe he narrates the numeric transcendence of raw limits.

    Either way, he's definitely electable.

  43. Jimbo Says:

    Thanks John.

    You definitely debunked my theory that leading off an inning after coming in from the outfield has any effect. I thought maybe having less time for pre-at bat routine would be a negative.

    I suppose most likely it makes no difference for catchers either then.

  44. BSK Says:

    Cy-

    Great point. If you have a great leadoff guy who also has power, no problem. If your best power hitter would also likely be your best leadoff hitter, it might behoove you to bat him elsewhere.

  45. Cyril Morong Says:

    Thanks, Neal L.

    Jimbo, I like your idea about studying catchers. It might be a little time consuming but it seems like we could just check the catchers who play the most and see how they hit when they leadoff an inning (I guess that BR has this split). Then compare it to the league average (I guess that BR has this split, too). Then see if catcher's normal hitting differs more or less from their leadoff hitting than other guys. Maybe this is what John did.

  46. John Autin Says:

    Jimbo -- I do think it would be interesting to look at catchers leading off an inning after playing in the field. I did not mean to dismiss the possibility of an effect on some batters just because it didn't seem to affect Barry Bonds.

    I don't have time to do the full study right now, but here's a piece of it -- the 2010 stats for all batters, when leading off an inning after playing in the field, and in all other situations. (Note that I did not try to filter out the DHs.)

    Stat* -- 1st** / Other***
    BA ---- .257 / .257
    OBP -- .317 / .322
    SLG -- .414 / .400
    OPS -- .731 / .722
    HR% -- 2.84 / 2.39 (HRs as % of PAs)

    * Rates were calculated after subtracting intentional walks.
    ** 1st batter of an inning but not of the game.
    *** All other situations.

    The differences are small, and it's just one season's data.
    But it's still interesting that the OBP and SLG results show the opposite of what one might expect: The leadoff OBP is 5 points lower, while leadoff SLG is 14 points higher (and that all based on XBH, not BA). The HR% is about half a point higher in the leadoff situations.

    Of course, even if we detect real difference in these situational results, we can't know how much might be due to the batter having played in the field, as opposed to the batter and/or pitcher having a different approach to a leadoff situation.

    I'll try to look at catchers when I have time.

  47. Johnny Twisto Says:

    Neil L, I thought Neil Paine's #39 was well-put. I don't objectively draw a line for "junk stats," and perhaps I should be more careful in my descriptions. I called that lineup ratio I described in #28 a junk stat because it "means nothing outside of the very specific context of its own formula." Rickey Henderson might have a ratio of 3.0, and Vladimir Guerrero a ratio of 0.3, and we can understand that means we'd rather have Guerrero batting behind Henderson if possible, but the numbers don't necessarily mean anything on their own.

  48. Jimbo Says:

    Wouldn't a lower OBP with higher SLG make sense because pitchers throw more strikes to leadoff hitters?

  49. John Autin Says:

    @48, Jimbo -- Maybe you're right about that. I may have been looking at the leadoff situation too much from the hitter's perspective.

  50. kds Says:

    I agree with Jimbo. The "other" column includes all situations with runners on. In general, pitchers in these situations will pitch in such a way as to lesson the chances of runs being driven in; hence the lower SLG. When you try to give the batter less than the usual in hittable pitches you will walk more; hence the higher OPB.