This article started with a conversation at Mullins before the UA Game. I believe it was Denizen Kane who suggested it, but I'm not entirely sure. If I got it wrong, put it in the comments and I'll give you credit.
We've heard a lot that Pitching wins championships. The questions is, does history bear this out? Is it possible that hitting gets you to the postseason but pitching wins championships?
To answer this question, I put together a data set that contains hitting and pitching data for every team and every season from 1998 to 2013. The end dates were chosen because 1998 was the first season that included 30 major league teams -- many of you will remember that was the year the Diamondbacks and Rays entered the league. For 2012 and 2013, I had to deal with the wild card. I decided to simply treat the loser of that game as missing the playoffs. That way, I consider the performance of 8 postseason teams every year.
Next, I had to define the question. I decided to consider number of wins. Specifically, I consider whether good pitching leads to more wins than good hitting in the regular season? Then, I ask the same question in the postseason.
Next came the question of how to define hitting and pitching. I settled upon two different indicators for hitting: wRC+ and OPS. For pitching, I use FIP. (I also got results using ERA. They are almost identical to FIP and, thus, omitted.) Since FIP, OPS, and wRC+ are measured in different ways, I needed a way to directly compare the results. I decided to use the impact of a one standard deviation change in the underlying statistic. I've linked to a technical defintion of standard deviation for people that are interested. The short version, however, is that standard deviation measures how much higher or lower than than the mean the "average" observation is. As I'm using it, you can think of it as the impact of the same relative improvement in pitching and hitting on games won.
The results for wRC+ suggest that hitting is very slightly more important than pitching. A one standard deviation increase in wRC+ leads to 7.06 more wins in the regular season and .61 more wins in the postseason. A similar decrease in FIP leads to 4.86 more wins in the regular season and .43 more wins in the postseason.
When I consider OPS instead of wRC+, the results shift as well. In the regular season, hitting and pitching are equally valuable -- leading to improvements of 8.10 wins per season and 8.01 wins per season, respectively. In the postseason, pitching increases wins by about .72 wins per team while hitting only increases wins by about .58 wins per team.
Given this, it seems that hitting and pitching have similar impact on performance in both the regular season and postseason play. This could be deceptive, though, because we ultimately are interested not in how many games they win, but do they make it to the playoffs and, once there, do they win the World Series?
I can happily report that analyzing this question leads to a rosier conclusion for Cubs fans. (The results are similar for both sets of offensive numbers, so here I will focus on OPS.) A one standard deviation increase in either pitching or hitting has almost the same impact on making the playoffs: a 22% increase for hitting and a 21% increase for pitching. However, one they are in the playoffs, an improved offense increases your chances of winning the world series by 4% while an improved pitching staff increases your chances of winning the world series by less than 1%.
Taken together, these results suggest a couple things. First, good teams in the regular season tend to win more games and make the playoffs, regardless of whether they have good hitting or good pitching or, ideally, both.
However, once you make the playoffs, things subtly change. Pitching may help a team win more games by extending a series but, in the end, the numbers suggest that better hitters do tend to triumph over better pitchers.
Filed under: Uncategorized