Easy. Pitching is the vast majority of what produces those 27 outs, and the difference between the best fielder and the worst fielder at a given position is rarely anywhere nearly as big as the difference between the best hitter and even an average hitter--especially in the outfield. Dunn is a huge offensive producer and a liability in the field at possibly the least important defensive position. It's not a matter of having it "both ways." When a hitter is at the plate, the only person who can make an out is himself. That is not the case in the field. Outs will be made via strikeout, in the infield, or in other parts of the outfield that he has no effect on whatsoever, but he has an effect every time he comes to the plate. The situations are not parallel, and that's why not making outs at the plate is more important than making outs in the field. it absolutely is a matter of having it both ways. the concept is that outs are precious. avoid them on offense, get them on defense. but you want to say x outs are really important, y outs are not so important. it's all about winning ballgames. both are incredibly important. Really? Is that what I want to say? That's surprising. I was pretty sure I wanted to say that an individual player's ability to avoid outs offensively is more valuable than his ability to make outs in the field in regards to winning games. But hey, what do I know? Why is this fair to say? Because there is a predictive difference between the two? Even the article says that just there can be many other factors, from other fielders to the pitchers to the type of park the player plays in. You're making a rather bold assumption here. The model only takes into account outs recorded versus non-zero out-making opportunities. It doesn't have any quality to suggest that they're anything specific--not bloopers or warning track line shots, or just fly balls he loses in the sun. Assuming that half would be doubles is pretty much just unfounded speculation on your part. I agree that a player should be held accountable for what he surrenders defensively in being evaluated, but you seem to be employing a highly faulty means of doing so. It's also noteworthy that despite your suggestion of this being the difference between a "bad defender and average defender," the stats you got this from suggest it's the difference between an average defender and a good defender. Equally good defenders on the list are Carlos Lee and Moises Alou, who sandwich Mench near the top of the list. And you're not taking into account the difference in the predicted data, which is the entire purpose of the data in that article. I haven't seen anybody "pretend" that his fielding has no effect. I've seen people suggest that his fielding isn't as inept as frequently suggested, and that his offense far outweighs whatever weaknesses he has in the field.