from Amazon
Friday, June 27, 2008
Are home teams cheating this year?
So far this year, the home team has won 56.64% of its games. The last 3 full years (05-07), the home team WP has been 54.2%. The difference in 1174 games so far in 2008 is 1.65 SD. The difference may even be less than that because in 2005, while the home team won 53.7% of its games, the home and road team run scoring suggested that the true home team WP was 54.4% (using a basic pythag formula).
This year, the home and road run scoring suggests a true 56.4% home team WP (as opposed to an actual WP of 56.6%), so perhaps we can say that the home team wp is 2% higher than over the previous 3 years.
Technically, if we want to know the SD of the difference between the last 3 years and this year, we have to take the square root of the sum of the variances between this year and 05-07 combined. That is 1.6%. So a 2% difference is 1.27 SD, not a whole lot to get excited about, but worthy of some further investigation.
It has been suggested (by Jason Stark who spoke to some baseball insiders) that road team players are no longer taking “greenies” and that they had been doing that for 40 or 50 years (I don’t know what the home team WP was prior to 40 or 50 years ago, but the travel schedule was completely different back then, so we might not be comparing apples to apples anyway). Thus the road team players are more fatigued this year than in years past.
The answer to the question, “Why suddenly this year?” is two-fold (I am just speculating here - I am not necessarily suggesting that the 56.64% home team WP is “real” or that “greenies” is the reason): One, HFA has actually been climbing a little the last few years (I think), and two, with stricter testing and penalties, and the off-season publicity surrounding the Mitchell report and PED’s in general, virtually all players have gone cold turkey on PED’s this year.
There is another possible explanation that I have not heard at all. Again, I am speculating and thinking out loud. Earlier in the season, I read somewhere that starting this season MLB is allowing teams to store baseballs in climate controlled conditions, much like the Rockies and the humidor.
I have not heard anything else about this. I have no idea which teams, if any, other than the Rockies are doing this, or even if it is true. The only other team I can think of that might want to do this for the same reason as the Rockies, is the D-Backs. They are at a high elevation and it is very dry in Phoenix. Perhaps in the humid cities, they are storing balls in a “dry-ador.” I don’t know.
As soon as I heard this (that all teams can store baseballs in climate controlled environments), I immediately thought, “What is to prevent them from cheating by using livelier balls when they are batting or when they are behind late in the game?” I have no idea who is in charge of storing baseballs and then providing them to the umpire during the game. Do they have to give all of the balls to the umpires before the game starts, which limits their ability to cheat? Are the balls “rubbed up” in the umpire’s room after they are taken out of any climate controlled environment? Even if all of the balls are given to the umpires prior to the game, can the home team “mark” the balls? That seems a little far-fetched, even when I put on my “conspiracy hat” as I am now wearing.
Anyway, I took the STATS data and looked at average fly ball distance (which is a proxy for ball “liveliness,” all other thing being equal) for the home and road teams this year. For 2005-2007, the home team averaged 324.6 feet per fly ball, and the road team, 324.4, for a difference of .2 feet, in favor of the home team. We expect this of course, as a natural consequence of the HFA.
What about in 2008 so far? The home team is averaging 324.0 feet and the road team 323.0, for a difference of 1.0 feet, quite a bit different than 05-07. Now, that could be a consequence of whatever is causing the home team to score more and win more so far this year, whatever that might be - greenies or whatever. It does suggest, one way or another, that maybe the extra HFA this year is NOT a fluke. Could some teams be pulling out livelier balls when they are hitting? I guess that is possible.
The other thing I thought of was that maybe home teams are pulling out livelier balls when they are losing late in the game. If I were going to cheat and did not want to get caught, I would probably do that rather than try and use different balls every half inning. Plus, it is possible that the ball boy goes into a room somewhere every couple of innings or so and grabs a bucket of balls. Again, I have no idea how that works. Maybe someone can illucidate that for us, who has worked at an MLB stadium.
So of course I looked at average fly ball distances late in a game (> 6 inning) when the home team was losing and when they were winning (and not when they were tied). The results are pretty alarming, I think.
In 05-07, when the home team was winning in the 7th inning or later, the average fly ball distance for both teams was 323.9. When they were losing, it was 323.5.
In 08, when they were winning, it was 320.7 and when they were losing, it was 322.5. Wow. In 05-07, when losing, the ball actually travels .4 feet less, for whatever reason. This year, when losing, the ball travels 1.8 feet further, a swing of 2.2 feet. I don’t have to tell you how much that is worth. It is worth a lot. My guess is that each foot is worth an extra .005-.007 HR per fly ball or maybe .1 runs per game (a WAG). So 2.2 feet is worth maybe .25 runs per game, which is a lot of course.
What about if teams are cheating, but they are only “monkeying” with the baseball in the 9th (or later) inning? Let’s say that they bring in their closer and bring out the “dead balls.”
In 05-07, when the home team was winning in the 9th or later, the average fly ball distance was 323.4. When losing it was 323.0. So again, the ball travels .4 feet less when the home team is losing at the end of the game. That was in 05-07.
What about in 08 (obviously the above >6 inning data includes the >8 inning data)? When the home team is winning, the ball travels 317.9 feet. When losing, it travels 321.2, or 3.3 feet further! That is a 3.7 feet swing as compared to 05-07! Wow!
Looking at it another way, in 05-07, the average fly ball traveled 324.5 feet. When the home team was winning late in the game, it traveled 323.9 or .6 feet less. With the home team losing, it traveled 323.5, or 1 foot less than overall.
05-07
Overall Home team winning 7th or later Home team losing
324.5 323.9 323.5
08
Overall Home team winning 7th or later Home team losing
323.5 320.7 322.5
05-07
Overall Home team winning 9th or later Home team losing
324.5 323.4 323
08
Overall Home team winning 9th or later Home team losing
323.5 317.9 321.2
As I said, I find these numbers alarming and troublesome. More research and more investigation has to be done. I would like it if someone with access to fly ball distances can verify the numbers, especially for this year. I would also love it if someone could shed some light on whether teams other than the Rockies are actually storing baseballs in some kind of climate controlled environment and what the procedures are for giving them to the umpires and disseminating them during a game.
I find it hard to believe that home teams are cheating in the sense that you mention. If that was going on, someone on an opposing team would get wise to it and blow the whistle. Tightly organized conspiracies can work without detection, but a conspiracy involving several teams? There are just too many people who would know, from club officials, to at least some players, and even the ball boys. It’s the ball boys who takes balls out to the umpire, isn’t it? If it is, he would notice. There are just too many people involved to keep it a secret.
I have no idea why this is the case, but it could just be the way things are working out this season. Even if the statistical methods that you apply suggest that something is going on, that doesn’t mean it isn’t just luck. I remember reading one of the Bill James abstracts back when. He talked about running one hundred seasons of Willy Mays on his baseball simulator. He found that in one season Mays was miserable, only hitting about .250. Stuff happens. Anomalies occur. I think you’d need more than one season’s worth of data to be certain.
That said, from a business point of view, it’s in the interests of all sports leagues to give as much of an advantage to the home team as is reasonable. A team that wins at home brings out the fans, or so they say. It seems to be the consensus among those who claim to know.