I think protection might be one of those things that could be anecdotally true but doesn’t really matter so much in practice. Not that I actually believe that. For multiple reasons, but to get into it let’s say the anecdotes are true and you do see more fastballs in front of a good hitter than you would otherwise. The numbers don’t support this theory, but it is the central tenet to the protection theory so let’s go with it for now.

To start are we really absolutely sure that ML pitchers necessarily locate their fastballs better than their curveballs? A pitcher who has so much trouble locating their curve that they can’t throw it when behind in the count probably doesn’t have great control of their fastball either, so it may not matter anyway. For that matter, is it really better as a hitter to get a fastball than a curve? Hanging breaking pitches are much easier to hit than a good fastball. If you’re facing a pitcher who is so unsure of where his stuff is going that he won’t throw a curve with a good hitter coming up, you’ve got to think that the chances of him hanging one are pretty high too. Maybe you want to get that curveball. For that matter, pitchers can’t automatically throw their fastballs for strikes, which is the crucial part of the protection theory, that you’re going to get a fastball for a strike and you jump all over that. And even if they could, no pitcher is going to sit there grooving fastballs to Joe Average because Manny Ramirez is hitting behind him, because Joe Average isn’t really average – he’s one of the best hitters in the world and you can’t groove fastballs to anybody in the majors except Bobby Crosby. But let’s say Fred Average the pitcher does anyway … what we have to consider now is the flip side that Fred Average’s fastball is a pretty great pitch. It comes in at 90 and has more movement than any of us have ever faced. So even if Joe is waiting for it it’s still pretty damn hard to hit. And Fred’s got eight fielders behind him, probably half of which are amazing, trying to catch the ball if you do hit it. And even if you do give up that base hit to Joe, you’ve still got roughly a 70% chance to get Manny out. Those odds aren’t bad. Why risk throwing Joe a bunch of stuff he might rake when you’re still more likely than not to get Manny out regardless? Throw Joe your best pitch (which might be the fastball anyway) and then throw them again to Manny. What else are you going to do?

I think this is where, anecdotally, the protection theory begins to break down. Even if there are a bunch of pitchers out there who are so lacking confidence in their breaking pitches that they can’t throw them behind in the count with a good hitter coming up, it might not really matter. The assumption is that behind in the count w/ good hitter coming up = a fastball is coming = the batter is going to/has a better chance of getting a hit. This might not be true (the numbers say it isn’t true). And even if it is, the effect might be so small that it doesn’t make any difference. After all, how many times during the year is Joe Average at the plate with a 3-1 count and Manny Ramirez behind him? 30? 40? A few more? A few less? Is 50 PA with a slight advantage on one or two pitches in the AB as compared to a regular AB enough to make a huge statistical difference over a year’s time? And is it even an advantage for all the reasons above?

Most likely I think the situation doesn’t really come about b/c you don’t see many pitchers in the majors who have to abandon everything but the fastball when they get behind in the count. Those guys are all in the minors and their name is Henry Rodriguez. (I joke, I joke. Sort of.) I think it’s more likely that ML pitchers throw everything as hard as they can at everybody because everybody in the majors is dangerous. Some more so than others but they’re all the very best hitters in the world.