What is the impact of Carlos Lee's loss?

Whenever we talk about whether the Astros can continue to climb the standings during the remainder of this season, Carlos Lee's injury last weekend always comes up.  And for good reason.  Lee's offensive performance was excellent and will be irreplaceable.

Rather than just speculating, let's put some numbers to it.

Surprisingly, whomever has played LF in Lee's absence has been hitting very well...so far. We can use baseballmusing's day by day data base to find that out. Prior to Lee's injury, the Astros' left field position produced a .308 batting average and a .903 OPS.  All very good, of course.  In the short period subsequent to Lee's injury, the LF position has produced a .429 batting average and 1.169 OPS.  Wow, I didn't realize the left field replacements had been that good.  Of course, we are seeing a very brief time frame, and we can't expect the LF replacement to continue that rate of offense.

In evaluating the future impact of Lee's absence on the Astros, we must evaluate both offense and defense.  While the loss of Lee will hurt badly on the offensive side, the Astros should gain some on the defensive side.

For the defensive measurement, I use the Baseball Think Factory's defensive runs saved (above/below average for the position) here and here.  BBTF's calculation is one of several range based metrics which are available, and I'm not saying it is necessarily the best one.  But it does produce a "runs" calculation, and it is openly available on BBTF's web site. 

Cecil Cooper appears to favor a lefty / righty platoon of Wigginton and Erstad in LF, with Blum used at 3d base when Wiggy is the LFer.  In effect, the offensive differential, resulting from Lee's absence, is between Blum and Erstad versus Carlos Lee.  I assume that Wigginton is about the same as Carlos Lee defensively.  When Erstad plays, he is much better than Lee as a fielder.  When Wiggy plays LF, Blum provides a defensive improvement at 3d base.  So, the defensive improvement is based upon Erstad vs. Lee in LF, and Blum vs. Wigginton at 3d base. 

I used Erstad's defensive runs saved in both LF and CF this season (in order to get a larger sample size).  BBTF's defensive calculation overstates how bad Lee is on the fielding side, because of peculiarities related to MMP's high left field wall.  Therefore, to be conservative, I adjusted Lee's terrible negative "runs saved" value, nearly -20, to -15 runs saved. 

For offensive performance, I used Runs Created from the Hardball Times stats, and turned the defensive and offensive runs into rate stats (per inning and per plate appearance).  Here are some results.

Runs Created Per 100 Plate Appearances

Lee  17.9

Blum  12.4

Erstad  12.95

Defensive Runs Saved Per 100 Innings

Lee  -1.7

Erstad  +1.2

Wigginton (3d base)  -0.1

Blum (3d base) +0.67

For guesstimate sake, I assumed 369 innings and 164 plate appearances  remain this season.  With the "Wiggy in LF/ Blum at 3d base" configuration, the Astros lose 6.14 net runs for the remainder of the season.  With Erstad replacing Lee, the Astros actually gain 2.5 runs over the remainder of the season.  That's because the defensive improvement of 10.63 runs saved exceeds the offensive loss of 8.1 runs scored.  If you assume a platoon in which Erstad plays 70% and Blum 30%, the net runs lost due to Lee's absence is minimal (less than 1 run).

Of course, this ignores the fact that Lee was blazing hot when he went down.  His OPS in August was over 1.5.  But, then again, how long could we have expected Carlos Lee to stay THAT hot?

I realize that a large margin of error surrounds this type of calculation; so I'm not going to try and convince you that the Astros won't miss Carlos Lee (even if these numbers say so).  Moreover, it is possible that the performance of Erstad and Blum will decline if they are over-exposed (meaning that they won't produce well if they are overused).  On the other hand, Cooper's platoon scheme may prevent either player from being over-exposed.  However, I think the exercise demonstrates that the Astros may not be hurt by Lee's absense as much as we think.  Runs which are allowed or saved by the defense count just as much as runs scored by the offense.