Home Or The Range?

It's too hot in the summer months.

The jet stream turns weak fly balls into home runs.

Texas will never be able to sign an elite pitcher because it's too hard to pitch there.

If you have been a Ranger fan for any length of time, you have undoubtedly heard these excuses given for why the perennially pitching-poor Rangers never seem to be able to put together an elite pitching staff. On its face, the argument seems to carry a lot of weight. Any fan who has braved an afternoon game during July or August knows how absurdly hot it can get and most have seen at least one fly ball catch the jet stream just right and carry over the fences. With a mystery team (this time, the Phillies) having materialized just in time to grab the latest elite pitcher of the Rangers eye, the hope for an elite rotation seems to be as hopeless now as it ever has been in Texas.

Or does it?

While fans often lament a team's pitching, especially fans of this team, the truth is that their concerns are more aptly described as being with Run Prevention. Run Prevention is an all-encompassing term for anything and everything that prevents runs from being scored including, but not limited to, pitching, defense and home ballpark. Pitching is the largest component of run prevention, which shouldn't surprise many. However, many assume that the effects of the ballpark are the next biggest factor.

One way to address this is to look at an interesting ‘statistic' at FanGraphs called ERA-FIP, or E-F. As a reminder, FIP is an attempt to evaluate pitchers based just on factors that are firmly within their control (including strikeouts, walks, and home runs). This removes many of the factors, such as defense, home park, and luck, which often get lumped in with statistics that evaluate pitchers like ERA. Examining a specific pitcher's ERA-FIP can add more legitimacy to the case for a surprising pitching performance being sustainable (as in the case of Colby Lewis last year, who doesn't have a long track record but also did not have a fluke year last year) or can help identify a pitcher who is pitching well according to peripheral statistics and may be suffering from a fair dose of bad luck (like 2009 Derek Holland). A large negative value for ERA-FIP suggests that a pitcher's appearances are resulting in a lower number of runs than one would expect under average conditions, perhaps because they are playing in front of a good defense or in a pitcher's park or they are getting lucky.

Looking at a team's ERA-FIP helps to reduce the role of luck or variance due to the larger sample size than for just one pitcher, though it's impossible to eliminate luck entirely. Taking that into account, I have charted park factors vs. ERA-FIP, as well as showing a linear regression with R-squared value:

The park factors are taken from FanGraphs in 2008 and Baseball-Reference in 2009-2010, though there isn't very much variance from year to year. They range from 91 to 116, with 100 representing a neutral ballpark. Above 100 represents parks that favor hitters, while below 100 favors pitchers. Over this time period, teams' E-F ranged from -0.54 to +0.56. Looking at just the data points, there is not much of an obvious trend that shows up. Considering the R-squared value of 0.0632, it becomes apparent that there is no strong relationship between Park Factor and E-F. This suggests that home park is not having as much of an effect in helping or hurting a pitcher as some would have you believe. 

However, there is something that shows a stronger correlation than the park factors did -- defense.  Below is a similar chart using the last three years of E-F, only this time the x-axis is defense as measured by team UZR:

A casual inspection shows some sort of trend whereupon better defense, as shown by a higher UZR, generally relates to a lower or more negative ERA-FIP, or fewer runs given up than one would expect from that team's pitching peripheral statistics. The R-squared value of 0.4443 shows mediocre correlation between the two, but notably much stronger than between Park Factor and ERA-FIP. 

Why is this relevant?

With the standard statistical caveat that correlation does not equal causation in mind, what this suggests is that the best way for a team to help its pitching perform better is not to move to a cooler climate or build a more spacious park, but to field a better defense. It seems that somebody in the Rangers front office has figured this out as well, as the 2008 Rangers fielded a team with a -45.8 team UZR that helped make the pitchers look even worse to the tune of a +0.53 ERA-FIP, yet the 2009 and 2010 Rangers' improved UZR (+25.6 and +18.1, respectively) had the dramatic effect of helping to lower the team's ERA-FIP to -0.1 and -0.24 and helping to produce more wins. 

So it seems that the Rangers can, in spite of the heat and jet stream, create an environment behind their pitchers that helps to limit runs and make their lives easier. If only the Rangers could convince the elite pitchers of this!

"Texas probably finished second to be honest with you. Just as far as the quality of the team and the chance to win a World Series ring, I think they're a better team." -Cliff Lee