In the bottom of the seventh inning of Game One of the National League Division Series between the Diamondbacks and Brewers, Arizona’s starting pitcher Ian Kennedy faced Milwaukee slugger Prince Fielder with the Diamondbacks losing by two, a runner on second, and two outs. Fielder hit a home run. Almost immediately after the event, commentators questioned the wisdom of letting Kennedy pitch to Fielder in that situation. I even heard the suggestion that the Diamondbacks should have given Fielder a free pass to first base. Did Arizona make the wrong (non-) move?
From a number-crunching perspective along the lines of the one found in The Book (the basic method of which I will follow here), a number of things should be taken into consideration when deciding whether or not a pitcher should “pitch around” a particular hitter (I will simplify by treating an “unintentional intentional walk” as functionally equivalent to a regular intentional walk). Among the issues to take into account are the true talent of the hitter, the true talent of those hitting behind him, the inning, the base-out state, and the score.
Read The Book to get the full details as to how why, but to get to the point: in the bottom of the seventh, two outs, a runner on second, and the pitching team down by two runs, the batter at hand needs to have an expected wOBA at least 1.21 times that of the batter on deck for an intentional walk to be a good idea.
Obviously, Prince Fielder is a great hitter. The most recent update of the Oliver projection system estimates him to be a .408 wOBA true talent hitter. Rickie Weeks was the hitter on deck, and while his projected .366 wOBA is good, Fielder’s .408 is obviously far better. However, the ratio of Fielder’s wOBA to Weeks’ is only 1.11, lower than the 1.21 needed to make an intentional walk (or other “free pass”) a good move.
Before closing the book, however, we should take another important factor into account. Assuming the right-handed pitching Kennedy would be pitching to both (Brad Ziegler and Joe Paterson were warming up in the pen), we should take the expected platoon skills of the hitters into account. Using the method outlined here combined with the Oliver’s overall wOBA projections for Fielder and Weeks as given above, we would expect Fielder to have a .424 wOBA and Weeks a .358 wOBA versus right-handed pitchers. That makes the gap between the two wider, but the ratio of Fielder’s expected wOBA versus right-handers compared to Weeks’ 1.18 — still not quite there. Although the general quality of pitcher is not as important in this case, ideally we would want to take the pitcher’s platoon skill into account. However, for his career Kennedy has actually pitched better against lefties, so doing the more complicated work necessary for estimating pitcher platoon skill and how it would interact with the hitters skill still probably would not make deliberately walking Fielder a good idea.
This is not to say that these are the only considerations to take into account. Perhaps Arizona manager Kirk Gibson or his coaches could have seen something in Kennedy at that point that would have indicated he should be pulled (I don’t know). Maybe Gibson would have been better served to bring in an appropriate lefty-killing reliever to get Fielder out.
It is easy to speak with the benefit of hindsight — after all, Fielder did hit a home run, a play that was the third-greatest contribution to the Brewers’ victory. But it simply is not the case that Fielder was likely to hit a home run off of Kennedy, or even get on-base. Even the best hitters make outs in 60 percent of their plate appearances, and Kennedy is a better-than-average pitcher. I seriously doubt people would have even mentioned Kennedy pitching to Fielder if the plate appearance had ended in a lazy fly to left field, or even a walk followed by a Weeks ground out to end the innings. The ratio of Fielder’s versus Weeks’ projected wOBAs versus right-handed pitchers is close enough that having Kennedy intentionally walk or pitch around Fielder would have been understandable. However, on the basis of the numbers, there was nothing wrong with letting Kennedy pitch to Fielder in that situation.