We've discussed the fact that Upton and Kinsler have huge home/road splits and pondered the implications. Below is a link to a recent article by Dave Cameron on this issue that focuses specifically on Upton, but deals with the issue generally.

Cameron makes some interesting points, some of which I and others have made. The most interesting point that he makes, which hasn't come up in our discussions, is the fact that home/road splits can be "noisy" even over seemingly large samples like 5 seasons. He notes that over the last 5 years, Andre Ethier hit substantially better at home despite the fact that the Dodgers have a pitchers park (or at least neutral park). And Adam Dunn has hit better on the road despite his home parks being hitters parks. Cameron concludes that even seemingly large samples aren't large enough to eliminate all of the noise in the data for home/road splits.

Another point that Cameron makes about Upton, which I have made about Kinsler, is that b/c of the division he plays in, a disproportionate percentage of his road games are in pitchers parks. So we should expect him to have a larger than average home/road split.

Cameron's analysis has assuaged some of my concerns about Upton. I was all for the Andrus-Upton trade before. Now it seems even more like a no brainer. Please, please, pull the trigger, JD.

http://www.fangraphs.com/blogs/index.php/justin-upton-is-not-a-park-effect-mirage/

I've always been of the belief that there are more factors that play in to the home/road splits. Some guys are just more comfortable at home for instance, sleeping in the own beds, etc.

Not surprised Carlos Gonzalez is at the top of that list.

I've made that point and so does Cameron. For the typical player, playing at home adds about 14 points to wOBA relative to playing on the road. This is not due to park factors, but non-park factors that create a true home field advantage.

The problem is that even over large samples, certain players exhibit extreme splits. Over the last 5 seasons, Upton's and Kinsler's wOBAs were 80 points higher at home. The question is why do these hitters have such extreme splits. Part of the answer is that they hit in hitter's parks. But even if you apply general park adjustment factors to these hitters, that doesn't come close to fully explaining their splits.

This could lead to the conclusion that these hitters receive some special benefit from their home park that most hitters don't and that if you removed these hitters from their home park, their production would collapse. Cameron is challenging that view. Basically, he argues that even large samples like 5 seasons might not be large enough to allow us to draw firm conclusions from home-away splits. I don't think he's arguing that the splits are meaningless. Just that they need to be interpreted with a lot of caution.

Ian Kinsler detractors cherry-pick stats against him all the time, but the fact is he's a fly ball hitter who hits in three extremely fly ball suppressive parks about 30 games a year. That type of sample (130-160 at bats) will screw road splits. Texas is tailor made for his swing, and places like Anaheim, Oakland and Seattle are deleterious to his statistics.

People who bring up Justin Upton and Carlos Gonzalez are incorrectly viewing the paradigm of road hitting, in my opinion. In Upton and Cargo's case, they play the majority of road intra-division games in San Diego, LA and San Francisco -- again, three pitching-oriented parks.

I'm not saying these numbers should be ignored, because that would be shortsighted of me. What I'm saying is, consider the park influences when you talk about players on a case-by-case basis.