Offense and Power

I always hear things like "the Pirates need more power", and "this offense won't score without a middle of the lineup home run hitter", and there are several problems with that. First of all, power != home runs. Obviously home runs help power, but to be a power hitter you don't need to be a home run hitter. As a quick example of this I looked at the isolated power leaderboard from last season. And there was Andres Torres sitting with good power numbers, despite having only 16 home runs. Which put him above players like Prince Fielder and his 32 home runs. I think this is a fairly obvious assessment, so I don't feel a need to delve any deeper into it.

A second problem is that I don't believe power is a must for offensive performance. I believe that runs are runs, and if you get on base, you will score. So I decided to take a look at the data.

I took a look at the years 2008-2011, and looked at the correlation between runs scored and home runs, runs scored and isolated power, and as a point of comparison, runs scored and on base percentage.

First, I will look at the relationship between runs scored and home runs. I did the graphs in OpenOffice, so I don't know how to get them in here, but it isn't all that important. The correlation coefficients broke down as follows.

2008: 0.38

2009: 0.55

2010: 0.53

2011: 0.62

Oddly enough, the correlation of every stat I looked at increased yearly. That could be do to many things I'm not even going to try to get into right now. The point is, the average correlation is around .5, which is a statistically significant correlation, but not a very strong one. The graphs of all of these (that I can't show you) are quite scattered, but they show that home runs certainly help an offense, obviously.

Now, for the relationship between runs scored and isolated power. The correlation coefficients:

2008: 0.49

2009: 0.57

2010: 0.59

2011: 0.71

The average correlation here is around 0.58 (I'm just guessing here), which makes overall power a stronger correlation to scoring runs than home runs, but it still isn't a very strong correlation. These graphs appear significantly less scattered, with a few outliers throwing off the correlation rather than a more spread out graph like home runs. So power is helpful, and moreso than just home runs.

Now, as a bit of a comparison, we will look at the relationship between runs scored and on base percentage. The correlation coefficients are as follows:

2008: 0.7

2009: 0.72

2010: 0.77

2011: 0.8

Holy correlation, Batman! With an average correlation of roughly 0.75, clearly on base percentage is related strongly to runs scored. These graphs are very linear, with few outliers.

Now for a summary. Power is fairly important. Obviously it's useful to have, and teams with little power rarely score a lot of runs. Home runs are less important. If a team hits tons of doubles and triples, but few home runs, they'll be just fine offensively. On base percentage is very important.

As for how it applies to the Pirates. They need some more power than they're getting now. Theoretically, that should come from Pedro Alvarez. They don't need home runs as long as they produce adequate power numbers. They should produce middle of the road power numbers with the addition of Alvarez, and a lack of a black hole at first base. However, I do not think they need another power hitter. Many people advocate acquiring a power hitter to bat in the middle of the lineup, but I feel as long as the team has on base machines, like McCutchen, Tabata, and to an extent Jones have been this year, they should have a quality offense. In the future, I believe the offense is set to be competitive, providing the black holes are filled. However, that does hinge on getting massive power output from someone, most likely Alvarez. If he pans out, I am perfectly happy filling out the other "power" positions, like first and corner outfield, with high OBP players.

In short, I think the team, with the addition of Alvarez, should provide s good enough offense. There is no need for an additional power hitter or home run hitter. And the data seems to back up that theory to an extent. If the team is to have an elite offense, things would need to be different, but I don't foresee the Pirates having an elite offense.

And yes, I know correlation does not imply causation. I'm not saying that this tiny little quick study I did is anything definitive. More food for thought, especially for the statistically minded among us.