One of the common arguments about valuing free agents is the fact that no deal is made in isolation. For every player that a team doesn’t sign, there’s some alternative. It’s almost certainly not an alternative they wanted to exercise and often it’s a substantial downgrade but establishing the baseline performance of that alternative gives us an idea of what the dropoff is. Starting off, we’ll take a quick look at left field.
There are a couple of options in the minors to fill in left field should Matt Holliday be signed elsewhere and the Cardinals choose not to pursue other free agents. You could make a case for players like Nick Stavinoha or Shane Robinson, whom have both already had a bit of time in the majors, to be considered. Both come with significant downfalls, defensive ones and offensive ones respectively. Jon Jay makes for a better discussion as he hits more than Robinson while fielding quite well. But when talking about left field, the most likely in house alternative seems to be Allen Craig.
A player who is supposedly without a defensive position (though I still can’t fathom the move away from 3rd), Craig has been a consistent producer at each stop in the minors. Over the last three years, he’s posted the following wOBAs beginning in 2007: .400, .386, .400. It’s worth noting that the last two come in hitter friendly environments but it’s obvious that Craig is doing something right with the bat. Maintaining an average over .300 with a high-OBP from decent walk rates, Craig has flashed plus-power hitting 20 HRs each year and no less than 25 doubles.
So what does Craig offer? Taking his statistics from last season in which he hit .322/.374/.547, I’ve gotten translations from both Minor League Splits and Baseball Prospectus. I don’t believe either set of minor league equivalencies (MLEs) is “open source” and even if it was, I won’t pretend that I’m personally in a position to evaluate its efficacy. Nonetheless, it’s been my experience that BP has very optimistic translations (both regular and peak) and Minor League Splits seems more realistic. That’s a totally subjective and anecdotal evaluation but it’s what I’m going to use to try and baseline his performance for next year.
His 2009 stats using BP’s regular translation paint a rosy picture of a player that hits almost as well in the majors posting a line of .296/.345/.516. That .364 wOBA seems very optimistic even to someone who has been a long time Craig advocate. I personally think that Craig could easily replicate a league average offensive season of .330 wOBA. Let’s assume that’s the high end of his projection meaning that Craig has a 20% chance of posting that line. We’ll use the Minor League Splits MLE for a low projection with a translation of .270/.315/.438 good for a .318 wOBA. We’ll assign a 50% chance that chance of that. We’ll also assume a 10% chance that he’s a total bust with a .297 wOBA.
Converting each of those to runs we’ve got +20, 0, -7 and -20*. Using the percentages above, that would give us an overall offensive run production of about half a run. That jives with my personal projection nicely. If you want the slash stat equivalent, think .275/.330/.425 or Melky Cabrera 2009.
Of course, that’s only one piece of the puzzle. Defense is far more of a crapshoot given that there’s little to no data on Craig in the outfield. He was rated as something around average at 3B by TotalZone (a psuedo play-by-play statistic available at Minor League Splits) but his skillset doesn’t seem like it would translate all that well to the outfield. He’s not particularly fast and his arm doesn’t rate any better than average. Moving from a more difficult fielding position to a corner outfield position we’d expect his defensive numbers to go up but I’m unwilling to make that assumption. I also have a hard time envisioning him as a worse fielder than some of the current MLB corner outfielders. We’ll call him a nuetral defender in left but you could rate him +/- 5 and I wouldn’t argue with you.
The positional adjustment for LF is -7.5 runs. Replacement adjustment is +20 runs.
Together that gives us a player who is about 13 runs above replacement or about 1.3 WAR. This seems like an eminently reasonable assessment of Craig. It could be worse; it could also be better but ~1 WAR seems like what should be reasonably expected.
So whenever the options in left field are listed and the rumors fly thick and fast, ask if the incremental WAR increase from free agent A or trade target B is worth the incremental cost increase from Allen Craig.
*Note: All run totals are based on 700 PAs or an equivalent set of defensive innings. Pro-rate the final number based on playing time.