Sgt Schwartz
Wooosh.
What is that supposed to demonstrate? That game-winning runs can be scored in the bottom of the ninth, just like at any other time? What’s the evidence that the Mets would have been less likely to have scored that run if they had batted in the top of the ninth rather than the bottom?
Some teams practice and use silent counts more than others, but false-starts caused by crowd noise are common. The bigger problem is getting verbal instructions to wideouts. Loud noise can make it simply more difficult to concentrate, too.
It’s tanglble, but often overstated.
Well, earlier in the game, the Mets might have been less likely to sacrifice a run in, preferring to try and get a big inning.
So in the few cases where the home team is tied/behind in the bottom of the ninth, and are in a situation with a choice whether to play for a big inning or try and get just one run, then they have an advantage in batting last (examples: if the game is close, you might try and stretch a long single into a double, or be more willing to bunt a runner over, whereas if you’re down more than one then your only hope is a big inning, so you need to baserun cautiously and play for hits at every at-bat).
Of course, this advantage has to be decreased by the fact that the fielding team also occasionally has the opportunity to make choices about whether to play tightly or give up a run or two to prevent a big inning. Examples are playing the outfield in with a runner on third and less than two outs, giving up intentional or semi-intentional walks in the same situation, or the classic ‘defensive indifference’ to a base-stealing attempt, which is so common it gets its own scoring category.
I had sort of assumed that batting last was a net advantage, but looking at how often the batting team has a decision affected by the ‘bottom of the ninth’ situation versus how often the fielding team has a decision affected, I’m wondering whether it’s not an advantage to field last [leaving aside psychological considerations]. Does anybody know what the Sabre-heads consensus is?
I’m not sure this is true any more, although, in the Negro baseball leagues, according to Wikipedia, the “last professional club, the Indianapolis Clowns, operated amusingly rather than competitively from the mid-1960s to 1980s.”
When comparing home-field advantage across different sports, it’s also important to compare winning percentage in general across the sports. Baseball games are significantly less deterministic than American football or basketball: A very good baseball team might have a record of only .600 or so, while a comparable football or basketball team might be in the .900 range. So in baseball, the home-field advantage is nearly half as large as the good-team advantage.
Well, if having the last at-bat was a significant factor in winning percentage, I would expect extra-inning games to show a much better winning percentage for the home team than overall win percentage. However, data from extra-inning games over the past 10 years show a 1082-987 record for the home team, or about .522 winning percentage. That’s not much different than usual, and may in fact be a little lower (IMO this is because crowds tend to diminish in size for EI MLB games, what will beer sales cut off before the 9th inning 
).
Regarding last at-bats, there is a common maxim that anyone who has played organized baseball has heard repeatedly: “Play for the tie at home, the win on the road.” Here’s an example of it in practice: 9th inning, runners on 2nd and 3rd, 2 outs, batting team down by 1. Batter hits a line drive base hit to right, scoring the runner from 3rd and allowing the trail runner to round the 3rd base bag. All else being equal, if you are the visiting team, the 3rd base coach is more likely to send the runner than if you are the home team.
Road teams therefore take more chances late in the game than the home team. Since this strategy wold come into play more in one-run games, it’s interesting to look at how teams have fared in one-run games over the last century. The home team has won nearly 60% of these type of games, which alone accounts for almost all the HFA. It’s naive to blindly attribute this to the all-or-nothing strategy of “tie at home, win on the road”, but it surely argues for it being a factor (along with, e.g. the crowd being more “into the game” if its close).
I think you may have misunderstood my post. I was referring to equality in scheduling rather than in quality of play.
Yep, probably just a funny coincidence.
To tease apart the different aspects of HFA, it would be most interesting to look to see if HFA exists in minor league baseball. Clearly, the teams there would not be built to accomodate the local stadium. Also, the crowds are smaller than for major league baseball, especially at the lower levels, thus reducing any influence the crowd may have on referee intimidation or amping the home team.
I would be surprised if someone from SABR had not already looked at some of this, but I cannot find the information now.
What would be most interesting would be for one of the independent minor leagues (e.g., the Northern League) to play part of their season in reverse order (home team bats first) to see how much this contributes to HFA.
Re: HFA in baseball.
For the last ups contributing to the HFA, I would point you to the 2001 World Series.
With one out in the bottom of the ninth in game seven, with the bases loaded and the game tied, the fabled New York Yankees brought their infield in to the infield grass. Luis Gonzalez blooped a single over Derek Jeter’s head to drive in the runner from third.
If this situation had occured earlier in the game, probably even in the top of the ninth, Jeter (SS) and Soriano (2B) would have been playing for the double play and only the first and third basemen (Martinez and Brosius) would have been playing in.
Trading outs for runs is commonly done when your team has more at bats. In the bottom of the last, you must stop runs from scoring in close games and therefore your defensive positioning is different. Had Jeter been playing for the double play (far back, almost to the outfield grass), or even “half way” (about even with the bases), he would have been able to catch the bloop single and the runners would not have been able to advance.
There is some field grooming that can be tailored to your strengths and weaknesses. Infielders a tad slow? Grow the grass a bit longer. Your guys can’t bunt? Slope the dirt on the foul lines so more bunts roll foul. Your guys great bunters? Do the opposite. Other guys have a great base stealer? Soak the heck out of the area around first base. The Twins even did some twiddling with the air conditioning ducts in the Metrodome to make a little breeze in or out as the situation demanded.
However, they’re playing better at home (23-12) than on the road (25-15.) So they don’t play as well on the road; they have the best road record simply because they have a much better overall record than anyone else. And Red Sox fans have been following the team for a few years now, and they played much worse on the road in 2003, 2004, 2005 and 2006. But in 2002 they were better on the road. It’s random chance.
I really don’t think major league ballplayers are much influenced by cheering.
The August 2005 issue of By the Numbers (Volume 15, Number 3), the SABR Statistical Analysis Committee newsletter, reports on an analysis, published in the Journal of Sports Sciences, of the last-batting advantage, or lack thereof, and conclusively asserts that there is at BEST no advantage to batting last. My bad; report follows.
Intuitively, it seemed (and still seems, but what can I do) to me that knowing that you can put the game away with one right hit is a huge advantage; tie game, one out, bases loaded, what the hell, put in the guy with his foot in a cast who can always hit a long fly. Similarly, knowing that you have to get two runs just to tie is pretty damn useful. I defer reluctantly to the facts.
////_/
Steven R. Bray, Jeff Obara, and Matt Kwan, Batting Last as a Home Advantage Factor in Men’s NCAA Tournament Baseball, Journal of Sports Sciences, Volume 23 Number 7, July 2005, pp. 681-686
This is another stab at determining the factors behind home field advantage. Along with crowd influence, travel, and familiarity with home field, it has been proposed that the very fact of batting last can lead to an advantage to the home team. It follows that a test of this proposal might be possible if a context exists in which the team batting last is not characteristically the home team. Bray, Obara, and Kwan examine such a context: NCAA tournament baseball, in which teams often play one another at a neutral site, and where the home team does not necessarily come to bat last. Using data from four-team tournaments at Divisions I, II, and III from 1999 through 2003, Bray et al. first found that neutral teams won exactly half the time batting both first and last. In addition, they noted that that home teams actually won more often batting first (67.3%) than last (59.7%); it follows that visiting teams did so also (40.3% versus 32.7%). Thus, there is no support for the hypothesis that batting last in and of itself is advantageous. One additional finding relevant to a slightly different issue was that home teams did not have a higher probability of winning a tied game or staging a come-from-behind victory in the bottom of the ninth inning than visiting teams.
There have been a number of articles attempting to correlate home field advantage and testosterone levels of players.
This WebMD article from June 2006 discusses increased testosterone levels among ice hockey players. While testosterone levels are up before any game, they are higher before home games than away games. While there is some babble about a primal urge to protect “home territory”, the more plausible (to me) suggestion is that they know they are playing before friends and family.
This May 2002 Australian transcript discusses similar findings holding true for soccer players. They also propose doing the same study with women soccer players, as women do have measurable amounts of testosterone.
I submit that the probability of the visiting team achieving any sort of victory in the bottom of the ninth inning is zero. 
No, it’s just a reflection of the fact that when the home team goes ahead by a run in the bottom of the ninth or the bottom of an extra inning, play halts and the game is over (except in the case of a walk-off home run, when up to three additional runs can score).
The extra-inning statistic is a more even-handed measure of whether HFA is stronger in games which are close and decided late, and as you note, HFA is actually weaker in extra-inning games.
Not in the study being discussed, in which the home team did not necessarily bat last.
This is true; in extra innings at least, home teams win by 1 run far more often than by 2 or more. Over the past century, for every extra-inning game won by 2 or more runs by the home team, there were about 10 extra inning games where the homers won by exactly one run. By contrast, 1942 extra-inning games won by the visitors had a 2-run or greater margin of victory, while 2374 extra-inning games were one-run extra inning wins.
Clearly that 10:1 ratio is a result of the home team quitting once they score the go-ahead run. If the same data applies to the 9th inning (I don’t see why it wouldn’t), it certainly skews the overall one-run results. In short, perhaps one-run winning percentage is misleading: A large chunk these games are won by the home team in the 9th or extra innings and are only “one-run wins” because the teams stopped playing. If they hadn’t, there may have been a greater margin of victory.