See what I did there? One of my favorite parts about having a blog is being able to choose titles. You’re welcome for today’s gem. Genius, I know. I thought I’d try something new today. Since Twitter lends itself to spontaneous baseball debates, I thought I’d use a conversation I had with @andrewdmoses regarding his suggestion that Andruw Jones would be a nice pickup for the Cardinals’ bench. You can follow our entire back-and-forth on the subject to the left. If it’s too small for you to read, you should be able to click the picture and access a larger image.
Anyways, I argued that there was no room for Jones on the Cardinals’ bench since that would leave the team with six outfielders and necessitate a return to Memphis for one of Allen Craig or Jon Jay. I then remarked that I would have preferred a part-time Andruw Jones over Lance Berkman because it would have presumably afforded Craig more at-bats and saved the team money to spend on other needs (e.g. middle-infield, bullpen reinforcements). After making these off-the-cuff claims, I figured I should take a closer look at my position and see if the numbers back up my talk.
You’ll notice four scenarios in the table below that attempt to measure the run differential between Berkman getting everyday playing time, Berkman getting less playing time due to injury, make-believe Andruw Jones acquisition, and a straight platoon between Craig and Jay. In the Craig/Jay platoon scenario, I gave Craig a majority of the playing time since Mozeliak stated intentions of giving him 400-500 at-bats before the acquisition of Berkman. Of course, Jay would have netted more than 200 at-bats, but he also will pick some of them up in CF (hopefully, not too many).
I utilized Tom Tango’s formula for converting wOBA to Runs Above Average (RAA) which is: (player wOBA – league wOBA)/1.15*PA. I used last year’s league-average .321 wOBA in the calculations. In retrospect, I wish I would have used a more standard league-average ~.335 wOBA, but I don’t think this should matter much in the comparison of players since I used .321 in each of my calculations. Just know that the league-average .321 wOBA was uncharacteristically low in 2010 and will likely be something closer to the .325-.330 range in 2011. Therefore, players’ RAA represented in the table below are probably somewhat inflated.
Player wOBA’s in the chart are based on Bill James’ projections for the 2011 season (available at FanGraphs). Allen Craig’s projection was a little more optimistic than I would have expected. For what it’s worth, ZIPS also projects Craig to be slightly above league-average (106 OPS+). I had a difficult time deciding which numbers to use for Andruw Jones because the outcome is pretty different based on whether you think he will replicate 2010 or more closely approximate James’ projection. To keep things consistent, I just used the more conservative projection.
The fielding runs are a little less scientific (aren’t they always?). Basically, I eye-balled 2010 UZR figures and fudged them towards that given player’s reputation. For example, according to FanGraphs, Jon Jay was actually worth -4.9 runs in the OF last year. Since he is regarded as a pretty solid defensive player, I just made him an average defender for the purposes of this exercise. I pretty much did the same with Andruw Jones. Of course, he isn’t routinely accumulating 20+ runs with his glove anymore, but he’s been around league average in time spent in the outfield over the past three seasons, so I’m neither adding or subtracting value based on his leather. Anyone have suggestions about how to better incorporate fielding runs in the future?
|Player||Plate Appearances||Projected wOBA||RAA||Fielding|
|Lance Berkman Actual Scenario|
|Lance Berkman Injury Scenario|
|Andruw Jones Fake Scenario|
|Craig/Jay Platoon Scenario|
As you can see, I was wrong in my preference for Andruw Jones over Lance Berkman (depending on how much you value the couple/few million dollars that would have been saved). Even when factoring in somewhat disastrous defense in RF, Berkman’s offense seems strong enough as long as it regresses towards career norms as James predicts. If Jones were to replicate his wOBA from 2010, however, he’d beat out the Berkman scenario by a little more than one run (15.13 to 14.78). Of course, James’ prediction offers a much safer assumption given Jones’ hadn’t posted a wOBA greater than .360 since 2006.