The Impact on Hitters Who Change Parks

(Special thanks to Tom Tango for working through the conceptual and analytical issues on this article with me)

After seven outstanding seasons as one of the National League’s premier hitters, Prince Fielder signed a nine-year $214 million deal to play first base for the Detroit Tigers. During his years in Milwaukee, Fielder averaged a .391 wOBA, 32 home runs (.0546 HR/PA) and posted a .257 ISO. Certainly, no one could argue about his productivity. But with a change to a new team —and more importantly, a new park — there are questions about whether Fielder’s offense will be impacted.

If Park Factors are to be believed, he should be in for a decline. By just about any model, Detroit is roughly even offensively overall, but a much tougher hitting environment for left-handed hitters than Milwaukee. That means we should expect Fielder’s offensive performance to decline more than basic aging and regression would predict. Since the Park Factor change only impacts half of a player’s games each year, the theoretical ratio between change in factors and change in performance is 2:1. Essentially, we’d expect a wOBA to decrease by 1.5% and home runs to decrease by 15%. There are a number of different Park Factor formulas, but the general pattern looks similar regardless of the factors you look at.

To get a sense of the real impact that changing parks has on hitters, I created a data set with two consecutive seasons worth of figures for all hitters with more than 100 plate appearances from 2003 to 2010. From there, I  looked at changes in players’ wOBA and in home runs per plate appearance from the first year to the second. Then I compared Park Factors and Home Run Factors for the parks in which they played in both years. (Park and Home Run Factors were borrowed from Seamheads, as I have done previously.)

There are a number of comparisons below: players with more than 100 PAs in both years; players with more than 300 PAs in both years; players with more than 300 PAs in both years and a first-year ISO greater than .145; and players with more than 300 PAs in both years and a first-year ISO less than .145. For each segment, I looked at the percent change in Park Factors (or Home Run Factors) and the percent change in wOBA (or HR/PA) for the top 100 or top 30 players based on the size of the change in their Park Factors or Home Run Factors. I looked at both positive and negative changes in those factors in each category.

The lower the ratio between factors and metrics, the greater the impact of the park change. Here are the results for wOBA:

Park Factors Change wOBA Change Ratio (PF Change:wOBA Change)
AVE_%Change AVE_%Change
Top 100 [>100 PA both years)] Positive Change 18.30% 2.20% 8.3
Negative Change -15.80% -3.30% 4.8
Top 100 [>300 PA (both years)] Positive Change 14.20% 1.50% 9.5
Negative Change -11.40% -3.40% 3.4
Top 30 [>300PA (both years)>.145 (YR1)] Positive Change 16.40% 1.40% 11.4
Negative Change -15.10% -5.80% 2.6
Top 30 [> 300PA (both years)< .145 ISO (YR1)] Positive Change 20.50% 7.20% 2.9
Negative Change -14.70% 0.30% 54.7

Remember that the actual ratio of Park Factor Change to wOBA Change should be around 2:1. But it isn’t. The closest we find is 2.6:1 for regulars with a first-year ISO greater than .145 when moving to a more pitcher-friendly park (2.6 ratio). And what about for hitters moving to a more hitter-friendly park? That ratio is 11.4:1. The impact on wOBA is far more muted than theory would lead us to believe. Not only that, but the magnitude of the impact varies quite a bit depending on the type of hitter you are and whether you’re moving to a more friendly or less friendly hitting environment. Power hitters don’t seem to be affected that much by moving to a more hitter-friendly park (11.4:1 ratio). But move to a more pitcher-friendly park, and the impact is easier to see (2.6:1 ratio).

But do we get any closer to the theoretical ratio for HR/PA?

Home Run Factors Change HR/PA Change Ratio (HRFactor Change:HR/PA Change)
AVE_%Change AVE_%Change
Top 100 [>100 PA both years)] Positive Change 30.10% 21.00% 1.4
Negative Change -23.10% -21.40% 1.1
Top 100 [>300 PA (both years)] Positive Change 22.00% 9.10% 2.4
Negative Change -16.90% -18.00% 0.9
Top 30 [>300PA (both years)>.145 (YR1)] Positive Change 28.60% 3.60% 7.9
Negative Change -23.20% -23.70% 1
Top 30 [> 300PA (both years)<.145 ISO (YR1)] Positive Change 28.80% 59.00% 0.5
Negative Change -19.10% -8.60% 2.2

The short answer is yes. In fact, we see a number of scenarios where moving to a park with a higher or lower home run factor has an impact of 1:1 or better on a player’s HR/PA. What’s interesting, though, is that power hitters (>.145 ISO) moving to a more hitter-friendly park don’t reap the same benefits as weaker hitters who do the same. On average, above-average power hitters in the sample moved to parks that were on average 28.6% more home-run friendly, but only increased their HR/PA by 3.6% — an almost 8:1 ratio. But for hitters with less-than-average power? They moved to parks that were 28.8% more home-run friendly and increased their HR/PA  by almost 60% — a .5:1 ratio.

For example, take a hitter who is below average in power (say, five home runs in 600 PAs) and a hitter who is above average in power (40 home runs per 600 PAs). Move both of them to parks that are 10% more home run friendly, and we’d would see the less-powerful hitter to increase his home runs by about 50% (from five homers to eight homers per 600 PAs) and the more powerful hitter to increase his home runs by only 1.3% (from 40 to 41 per 600 PAs).

A change to a less-friendly offensive park produces slightly different results. The less-powerful hitter in the above example might see his home run total stay roughly the same; the more powerful hitter would see the park have a greater impact on his total.

So what does this mean for Fielder new digs in Detroit?

Using the data above to estimate, we might expect Prince’s wOBA to increase by roughly .4% — from .408 to .410 — since Detroit had an overall run park factor 5% higher than Milwaukee (101 versus 96). However, with natural aging and regression his wOBA will likely decline from last year, but park factors may not play as big of a role as we think. But  if we look at home run factor, things begin to change.

Overall, Comerica is 8% harder on home runs than Miller Park. That means Fielder could see his home run totals dip from 33 HR per 600 PAs to 30 HR per 600 PAs. The other way to look at it is that Comerica is 23% harder on left-handed hitters when it comes to home runs. I didn’t run the analysis with home-run-factor splits, but just to throw it out there, Fielder could see his home run rate drop from 33 per 600 PAs to 25 per 600 PAs. So, we may not expect a massive drop in Fielder’s overall production, his home run totals could see a decline.

None of this is to say that the ratios I’ve listed are concrete. Still, they suggest that applying park factors in a straightforward way (i.e. assuming a 2:1 ratio between factors and performance) can be problematic for hitters depending on the type of player (above-average power or below-average power) and whether that player is moving to a higher or a lower run environment. The data suggest that we need to do more work in this area if we want our park-neutral estimates for players to truly reflect reality.


To make life easier on folks, I have created a Google Doc with all the data I used for this study as well as a calculator. For the calculator, all you need to do is input the wOBA or HR/PA from YR1 in the green cell and then the percent change in park of home run factor in the other green cell. Again, I used Seamheads factors for the study. You will automatically see the percent change and estimated YR2 outcomes for each type of player calculated for you based on the ratios noted above.

Print This Post

Bill leads Predictive Modeling and Data Science consulting at Gallup. In his free time, he writes for The Hardball Times, speaks about baseball research and analytics, has consulted for a Major League Baseball team, and has appeared on MLB Network's Clubhouse Confidential as well as several MLB-produced documentaries. He is also the creator of the baseballr package for the R programming language. Along with Jeff Zimmerman, he won the 2013 SABR Analytics Research Award for Contemporary Analysis. Follow him on Twitter @BillPetti.

Comments Are Loading Now!