# Different Factors For Different Folks Part II

A little while ago, in Part I of this series, I looked at how there appeared to be different homerun park factors at work for high and low percentage HR hitters, when comparing their records in Japan and in the States.

Since then, I’ve had a chance to update my park factors with the release of RetroSheet’s 2008 events files. As part of that, I rounded each park’s homerun factor to the nearest 0.05, or 0.1 if greater than 1.35 or less than 0.70. I also coded the all the batters from 1953 to 2008 for their career percentage of homeruns per batted balls.

AA .080+ A .060 – .080 B .045 – .060 C .035 – .045 D .020 – .035 E .010 – .020 F .000 – .010

The below chart cross references these ratings of batters and MLB ballparks, and shows the observed HR factor for each combination. The colors indicate the sample size, with dark green above 50,000 batted balls; light green 30,000-50,000, yellow 15,000-30,000 and orange below 15,000.

C is centered on the current mean rate of .040. As you can see in the chart the C batters HR factor (road rate divided by home rate) was just about the same as the factor for all batters. D, E and F had ratios increasingly further from 1 (effected more by the park) while B, A and AA batters had ratios increasingly closer to 1 (effected less).

This larger scale study agrees with my earlier study of Japanese batting stats, where it is generally acknowledged that the JPB parks as a group are a much easier HR hitting environment than MLB. By how much? In Part I, I created five groups of homerun hitters. The highest group had a JPB/MLB factor of 1.18, while the lowest had a factor of 2.27. On the chart, this corresponds to the Japanses parks as a group having a HR factor of 1.40-1.50 compared to MLB parks.

The main thing I wanted to get out of this study was a more precise way of measuring how each ballpark changed the homerun rates. Unfortunately, I haven’t wrapped my head around those numbers yet, which is part of the reason this article has remained a draft for several weeks. What we know going in is how many homeruns were hit in each ballpark, and who hit those homeruns. Some players hit more homeruns than others, but how much of that is due to their own talent at hitting a baseball a long way, and how much of it was the dimensions of the ballpark they played in? I have verified that players who hit a lot of homeruns are much less effected by their ballparks than players who hit few, but I have to avoid the circular logic of having to know what a hitter’s HR% is in order to calculate his HR%. Perhaps something along the line of calculating a player’s personal home/road factor, and then comparing that to the factors of the parks he played in.

While I’ve been pondering this, Greg Rybarczyk of Hit Tracker posted an article at Baseball Analysts offering a new approach using detailed batted ball data. Going forward, this is an approach I favor – look at the trajectory (distance, direction and speed off bat) and type (grounder, flyball) for each ball hit in each ballpark. Each classification of batted ball will have it’s own set of probable outcomes in each ballpark. Put a batter in a different set of home and road parks, and calculate how much the expected outcome changes based on those details of how each ball was actually hit. However, when looking back at past seasons, we still need to fine tune the normalization of batting stats with the data that’s available.

HR Factors by overall factor of ballpark vs career HR% of batter

 Factor AA A B C D E F 0.30 0.52 0.58 0.40 0.31 0.37 0.18 0.36 0.40 0.60 0.56 0.50 0.47 0.47 0.45 0.34 0.50 0.69 0.59 0.58 0.59 0.53 0.52 0.34 0.60 1.20 0.69 0.69 0.54 0.66 0.55 0.51 0.65 0.79 0.77 0.67 0.66 0.64 0.72 0.75 0.70 0.92 0.79 0.75 0.69 0.68 0.68 0.71 0.75 0.75 0.83 0.77 0.75 0.76 0.72 0.76 0.80 0.80 0.86 0.83 0.85 0.80 0.77 0.75 0.85 0.96 0.93 0.89 0.86 0.83 0.93 0.79 0.90 0.98 0.91 0.92 0.96 0.92 0.92 0.81 0.95 1.00 1.00 0.98 0.95 0.96 0.96 1.00 1.00 0.97 0.97 1.03 1.04 1.07 1.04 0.95 1.05 1.05 1.12 1.05 1.05 1.10 1.10 1.07 1.10 1.01 1.07 1.11 1.14 1.15 1.18 1.36 1.15 1.11 1.11 1.20 1.16 1.20 1.23 1.46 1.20 1.12 1.16 1.12 1.33 1.29 1.29 1.61 1.25 1.23 1.08 1.19 1.32 1.34 1.44 1.63 1.30 1.17 1.35 1.27 1.34 1.35 1.46 2.21 1.40 1.15 1.23 1.43 1.36 1.59 1.86 1.21 1.50 1.32 1.12 1.43 1.51 1.80 2.14 2.27 1.60 1.56 1.45 1.25 1.83 1.85 1.45 4.05 1.70 1.38 1.63 1.71 1.60 1.75 1.89 3.33 1.90 1.29 1.59 1.93 1.58 2.68 2.90 3.08

### 3 Responses to Different Factors For Different Folks Part II

1. KJOK says:

Good stuff. So, by the ratio method, good HR parks help poor HR hitters more than they help good HR hitters, and vice versa.
What if you measure the parks and hitters by
+/- actual HR’s instead of by ratio of home runs?
I’d guess than in a league with the HR average of 15, if you had a 5 HR ave player and a 25 HR average player, that although a high HR ratio park increases the 5 HR player’s RATIO more, on an absolute basis he still might pick up fewer HR’s overall?

2. Brian Cartwright says:

Tango has recommended the +/- system for HRs, and this is evidence that it might be a better model

3. MattS says:

I liked this article, and I’ve been thinking about it all day. I think that KJOK hit on a good point– it should be more like +/- since more powerful hitters may be affected more by park size in terms of how valuable they are, but have fewer of their home runs taken away percentage-wise.
Ryan Howard may have a smaller percentage of his long flyballs between 340-360 feet than Ichiro who may rarely hit flyballs more than 360 feet, but Ryan Howard probably hits more flyballs between 340-360 feet and more of his value probably derives from having those hits land for homeruns. Thus, Howard is relatively more valuable in Citizens Bank Park than Safeco compared with Ichiro, even if Ichiro’s homerun total would practically double in CBP.