1871 Troy Haymakers / Franchise: Troy Haymakers / BR Team Page
History, Comments, Contributions
Did Troy play in a stealth "hitters park"? I think so.
A word here is probably appropriate about park factors. Based on standard analyses looking at runs scored at home and on the road, the Chicago White Stockings had the most offense-friendly park in the league in 1871, and the New York Mutuals, the most "pitcher friendly" park. BUT, in the early National Association, one cannot only look at "park factor," one also needs to consider "ball factor."
You see, in the early years, there was no standardized ball, and a number of manufacturers made varying degrees of "live" and "dead" balls over the years. Now here's the catch -- the visiting team generally supplied the ball. So, if your team favored the dead ball approach to hitting, you'd use the weak, soggy ball in all of your road games, and a mix of balls in your home games (supplied by the visitors). That could make your park seem to favor hitters even if it didn't. Conversely, Troy was infamous during its brief history for using a "live ball" on the road, effectively bringing the mountain (a 19th Century Coors Field) to Muhammed. Call it the "anti-humidor" effect.
Troy appears to be a neutral park in 1871. Under my "ball effect" theory, it may in fact have been a hitter's park, but run scoring was depressed compared to a live ball being used in road games and an "average" ball (the average of whatever balls the visiting teams brought) during home. But how could we test that?
As anecdotal evidence, Troy score 351 runs in 1871, the second most in the league (second to the league champion Athletics), while allowing 362, the most in the league. Possibly just the result of good hitting and poor pitching, but a point that makes the question of "ball factor" more interesting.
The easiest way would be to see how Troy's pitchers ERA+ changed after leaving the Haymakers' Grounds. If Troy was in fact a stealth hitters park, ERA+ should increase when a pitcher is traded to another team. Unfortunately, our sample size here is practically non-existent. John McMullin pitched every one (249.0)of Troy's innings in 1871, but only 15 innings in 1872 and 8 innings in 1873 (for New York and Philadelphia respectively). His ERA+ did increase (from 76 in 1871 to 96 in 1872 and 152 in 1873), but we can't draw any conclusions from that single, insignificant data point.
Another approach is to look at the hitters. Eight Troy hitters had at least one at bat in 1871 for Troy and for another team in 1872. In total, there were 747 common at bats (using the lower number of at bats for each player). Our primary data points here are Mike McGeary, Bill Craver, Tom York, Lip Pike, and John McMullin (who, after his Troy pitching debacle, became the regular left fielder for New York). All five had at least 100 at bats in 1871 and 1872. Over 747 at bats, the eight men who left Troy had a combined OPS of .742 in 1871 against a league average of .704. Thus, as a group their OPS was 6.8% higher than the league's. Looking at the common 747 at bats in 1872, these 8 men put up an OPS of .657, which happens to be exactly the league OPS for the year.
Did the "ball factor" increase Troy's offensive performance by almost seven percent in 1871? There, not enough data to decide conclusively, but the evidence is at least persuasive, if not conclusive, that it did.