Here we are approaching Labor Day and Todd Helton of the Rockies is pushing to be the 1st big-leaguer since Ted Williams to hit .400 for a season. Some folks disparage his efforts, pointing out he plays half his games in the best hitting ballpark in the bigs. OK, fair enough. But Williams played in a great hitting ballpark as well and I recall Wade Boggs hitting .420 or so at Fenway several years ago. I did a little digging and found home/road splits for Helton and Nomar Garciaparra.
Todd Helton
g ab h r rbi bb k tb ba slg
all 124 451 179 114 110 80 42 324 397 718
home 61 229 99 75 70 46 19 185 432 808
away 63 222 80 39 40 34 23 149 360 671
Nomar Garciaparra
g ab h r rbi bb k tb ba slg
all 106 393 145 80 74 52 39 242 369 616
home 54 190 74 45 39 31 20 114 389 600
away 52 203 71 35 35 21 19 128 350 631
Todd picks up 72 points in average and a ton of power at home. Nomar is better by 39 points at Fenway, with a little more pop on the road. Another way to look at it is, Todd’s rate of hitting is 20% higher at Coors against a 11% increase for Nomar at home.
My question is, how much was Williams helped/hurt by his home park in 1941? I’d be amazed if Fenway hurt him. I’m guessing he hit close to .435 there, which would mean he hit only .375 on the road. Poor guy.
FTR, here’s Ted’s seasonal line.
Ted Williams
g ab h r rbi bb k tb ba slg
all 143 456 185 135 120 145 27 335 406 735
This is always a tough question to answer definitively. The answer seems to be “helped some, but probably not as much as Helton benefits from Coors”. In their 1984 book The Hidden Game of Baseball, John Thorn and Pete Palmer calculated a “park factor” for Fenway in 1941 of 1.09, meaning that scoring by visiting teams in Fenway that year was ten percent above level of scoring by visiting teams in all parks that year. They also point out that the average batter hits about 10 percent better at home than on the road, all things being equal, for what that’s worth. On the Total Baseball web site, which also uses Thorn and Palmer’s statistical methods (albeit refined somewhat from their earlier form), the park factor number for Fenway in 1941 is 102 (100 being a league-average park that year) – this was the lowest park factor for Fenway of any year in Williams’ career.
By comparison, Coors Field’s park factors according to Total Baseball for 1998 and 1999 were 123 and 119, respectively.
Now you have to keep in mind that the park factor is calculated by its effect on runs scored, not batting average or slugging percentage. So it’s possible that there’s a discrepancy there between what’s been measured and reported and what you’re asking, but I think the magnitude of the effect is roughly equivalent on BA and SLG at least.
Jim Furtado has an interesting article at the Baseball Think Factory site in which he discusses various aspects of calculating park effects on various types of events.
None of this should take away anything from what Todd Helton’s already achieved this year or anything he may achieve the rest of the way. He is still hitting .360 on the road, after all (though admittedly his power numbers are significantly lower).
Sure but someone’s got to have the home/road splits.
I don’t know how to consider Palmer’s PA figure of 1.09 for Fenway 1941. I pulled these numbers from Big Mac and adjusted AL values to exclude Bostons’. In each case Boston topped the league.
BA 283/264 SLG 430/383 Runs 865/720.
BA up 7% relative to league, SLG 12%, runs 20%. Red Sox must have hit the ball at every park that year. They did have some nice sticks in addition to Ted: Jimmy Foxx, Bobby Doerr, Joe Cronin, Dom Dimaggio.
I mentioned Boggs in the OP. Here’s some data from a variety of sources.
So for this period Boggs hit 21% better at Fenway. If we use the same number for Ted (I know, I know) we have him hitting x on the road and x * 121% at home, which gives a road BA of 367 and at Fenway a cool 445.
I’m someone does, but none of my print sources at home do, nor do any of the online sources I use.
The 1941 Red Sox’s offensive prowess would to some degree artificially depress the park factor for Fenway that year by slightly increaing the scoring at the rest of the parks in the league, but that effect should be minor. And the park factor for Fenway remained in the range of .99 to 1.05 for nearly every season in the late '30s and '40s, so it’s not as if 1941 were really anomalous.
In the absence of actual home/road splits for Williams’ '41 season, however, we’re left with extrapolating from data for other hitters. While this can be a fascinating exercise and can yield useful insights about overall trends and the effect of parks on the mythical “average player”, I believe that such extrapolations have the potential to be especially misleading with regard to players whose performance is well above (or below) the mean. In other words, while the park factor in general may increase or decrease total offense in a particular park, the effect does not necessarily apply equally to all players and is least likely to apply in the same way to a player whose stats are more than a few standard deviations from the mean.
According to the book Hitter by Ed Linn, Williams hit .428 with 19 HR’s at Fenway, and .380 with 18 HR’s on the road in 1941. He also lost the MVP race 291-254 to Joe DiMaggio.
Blunt: Thanks very much for the information. Can you imagine sitting in Fenway all summer and watching some guy put together a 428 year? Do you have any other home/road splits for Ted in 41 you could post?
Thanks also for the steer on Hitter. Although I’ve read Linn on Durocher and Veeck, I’ve never heard of this one.
rackensack: I’m not sure I take your point on the Sox’ bats depressing the numbers for park factor. Couldn’t you say as well that having an above-average collection of hitters would drive up PF, regardless of whether it was already high? Or are we getting into a chicken and egg thing?
I agree that it is generally useless to try to extrapolate the elite’s performance from that of the rabble. Granted that neither is a Williams that’s why I looked at Garciaparra and Boggs. Playing with the numbers is a part of baseball that I’ve always found ‘fascinating’ but I try not to take them to the bank.
Here’s the splits for William’s 1941 season:
G AB R H TB 2B 3B HR RBI BB HBP AVG OBP SLG PRO
Home 75 243 72 104 186 21 2 19 62 80 3 .428 .574 .765 1.339
Road 68 213 63 81 149 12 1 18 58 65 0 .380 .535 .700 1.225
Total143 456 135 185 335 33 3 37 120 145 3 .406 .551 .735 1.286
Sorry I have no idea how to post a table and it looks horrible. If someone can explain how to make a real table let me know and I’ll try to repost this. If someone can understand it and wants to post it as a table, go ahead.
The book Hitter has a great appendix with all kinds of William’s stats, including a day by day recount of his 1941 season wich has his average going into each day. It also has a lot of his stats from the Pacific Coast league, and with the Minneapolis Millers in the minor leagues.
Park factor is computed by calculating the statistics for all the Red Sox and their collective opponents stats’ at Fenway, and comparing them to the same “population” of road games.
In other words, for this year, you’d sum the runs scored in all games between the Red Sox and Yankees at Fenway, and sum the runs scored for all the games between the same two teams at Yankee Stadium.
You would do this for each team, and compare the summed results. Park factor attempts to cancel out the influences of better or worse hitters by computing the same player sets as much as possible.
Certainly there are extraneous influences–Pedro Martinez might have pitched more games in Fenway than on the road. But these influences, as a whole, are usually statistically small.
Thorn and Palmer’s method of calculating park factors uses only the scoring by visiting clubs, in an attempt to compensate for the possibility that a club may well attempt to field a team that is particularly designed to take advantage of the home field’s characteristics. The scoring by visiting clubs in each park is compared against the scoring by visiting clubs in other parks, to arrive at the amount by which visiting club scoring in a particular park is higher or lower than the mean. Thus, a particularly strong offensive club in a certain year might account for higher overall scoring in other parks throughout the league, reducing the difference between their own park and the others. With only eight teams in each league in 1941, an individual club’s effect on the park factors overall would represent one seventh of the whole, while now it would be around one fourteenth (given that we now have fourteen teams in the AL and interleague play.
Other methods, such as the one by Jim Furtado referred to in my previous post, don’t distinguish between scoring by the home and visiting teams, so a strong offensive club wouldn’t ultimately influence the numbers one way or the other.
I don’t know how much of a difference it makes (and it certainly can’t be proven empirically to make any difference at all), but today’s ball players have to play more night games (lights = bad vision on the field) and have to travel more and over longer distances.
Again, I don’t know if it makes any difference at all, but Helton’s achievement is certainly noteworthy. To try to denigrate it because of his home field or any other contrivance does him a disservice. After all, if Coor’s field is such a hitter’s paradise, why isn’t the entire Rockies team flirting with .400?
Also (and I think this is a real consideration), we’re not talking home runs here, which could be accounted for by the conditions at Coor’s. We’re talking hitting safely, with eight guys out there positioned to stop the batter from doing so. Does Coor’s Field actually encourage hitting safely, or just the long ball?
Taking a look at a comparison of Teddy Ballgame and Todd Helton at espn.com, one big thing I noticed: Ted drew far more walks than Todd does. This is a pretty big thing, as if you draw 145 walks, that’s 145 at-bats you don’t have to hit in, and thus 58 hits you don’t have to pound out. The difference between Teddy then and Todd now is 65 walks, so let’s assume that Tod draws 80 walks (pretty decent estimate). That’s 60 at-bats more in which he’ll have to hit safely to get .400, and 24 more hits in those 60 at-bats. Not at all easy for anyone to do.
That said, anyone who bats .300 or better is perfectly fine in my book. Lord knows 99 percent of the human population of the Earth couldn’t do what MLB players do.
Sigh . . . . when in doubt, espn.com usually has what you’re looking for.
Practically every offensive event occurs with greater frequency at Coors Field, not just homers. In an attempt to partially compensate for the effects of thinner air, the park was designed with the fences placed somewhat deeper than usual, increasing the amount of territory the outfielders have to cover. Typically, most play deeper than they would in other parks, and hence a lot of balls fall into the gaps between and in front of them. The amount of ground Rockies outfielders must cover is one reason the team’s decided to place a greater emphasis on speed in the outfield (Jeffrey Hammonds, their experiments with Tom Goodwin and Brian Hunter, newcomer Juan Pierre, etc.).
Also, the thin air causes baseballs not only to travel further, but to do so in a lesser amount of time–meaning that balls arrive at their eventual destination a bit faster. This cuts down on the time an OF has to catch said ball. So not only do fielders have to cover more ground, they’ve got to do so in less time. Not good.
[Hijack] I can’t believe the stupid Dodgers were dumb enough to be fooled by Tom Goodwin’s Coors inflated on base percentage. As a lead off hitter, Goodwin makes an excellent defensive replacement and pinch runner. [/hijack]
The way this was explained to me and others (by one of the founders of the Baseball Prospectus, by the way) was that the effect on baseballs hit would be far greater than the effect on humans’ abilities to move: You and I are still limited by our reactions and by physiology. Also, presumably the effect on a baseball hit 200 feet (or however far) into the outfield would be greater than it would be on a human running far less distance than that to catch a ball.
Mind you, the above reasoning is a WAG. I should ask the guy who came up with the theory.