thanks.  Actually, one of the things I've always wondered is just how accurately can a pitcher throw a pitched ball. 6 inches? 8? A foot?   I think there is a tendency to overestimate how precisely a pitch can be located.

I think PFX can shed some light on that, although I don't think it will give us a definitive answer.   It's easy to measure where the pitch went, not see easy to determine where it was aimed, of course.

Would these facts explain why some pitchers with great control hit so many batters?  Pedro, for instance.   Sure, he throws inside, and sure, he has a rep for throwing at people, but maybe this extra dimension would explain.

Ex-MLB pitcher Dr. Dave Baldwin tells me that a backup slider is a pitch that was supposed to break horizontally but doesn't; hence the catcher has to "back up". As Hernandez says, it is a "mistake" pitch. The reason why it doesn't break is that the spin axis is accidentally oriented too far forward, like a modern-day gyroball. He and I wrote an article about that for SABR's current edition of the Baseball Research Journal (http://webusers.npl.uiuc.edu/~a-nathan/pob/BRJGyro.pdf, see p. 79).

I have the Hernandez book but have never read it seriously. Based on John's recommendation, I will now do so.

Keith is correct on all counts.  Although, when he says NEVER, that is probably not true. There are probably some pitchers at some times that intentionally throw a backup slider or an inside change up.

But as a general rule, pitchers don't do that, and there are very few exceptions.

The other reason you don't throw backup sliders besides it being very difficult psychologically is that you cannot control the break precisely so that you will hit too many batters when you try and throw it.  You'd rather leave a slider over the middle of the plate when aiming for the outside corner than hit a batter when aiming for the inside corner.

Dr. Baldwin is compeltely wrong with his definition, although "backup slider" is only a term.  It has nothing to do with whether the slider breaks or not.  It is exactly as Keith says - a slider that breaks over the inside corner of the plate from a same side pitcher.  Now, Baldwin may have HIS own definition but it is not the definition that everyone else uses, which is really the only one that counts (again, since it is only a term).  For example, I can't say that the definition of "fast" is "slow" because that is the way that I use it.  That is silly.  The only definition that counts is the "offcial" one or the one that is pervasive in everyday use.  Poll 100 broadcasters and 98 or 99 of them (maybe 100) will tell you the Hernandez definition.

The reason you rarely throw a change up inside is two-fold:  One, the batter needs to swing early on an inside pitch and swing late on an outside pitch in order to square it up.  That is just the physics of  "bat meets ball."  Which one will the batter tend to swing early?  Obviously the change up. Therefore you don't want to throw it inside.  Two, since the change up tends to tail away from an opposite side hitter, and you throw your changeup mostly against an opposite side hitter, again, you want to throw it away and then let it tail even more away.  If a batter swings early at a pitch that is outside and tailing even more outside, it is almost impossible to hit.  If you throw a changeup inside to an opposite side batter and it tails away from him, where the heck do you want to start it?  It just ain't gonna work.

You cannot tell where a pitcher meant to throw a pitch from the pitch f/x data.  Impossible.  What you can tell, as David says (and correctly so) is the AVERAGE location of a certain pitch at a certain count for any one pitcher or all pitchers collectively.  You just look at the scatter plot and find the point which gives you the "least sum of squares" or something like that.  But...

That will not tell you, for example, if a pitcher occasionally purposely throws to a different location.  For all pitchers, you will see a scatter plot of the change up around a location down and away.  But you still won't know if he occasionally tries to throw a changeup inside.  He generally won't.  You MIGHT try and throw a change up inside to a same side batter, but rarely if ever to an opposite side batter.

If you are watching a game, you can always tell where the pitcher intends to throw, a fact that seems to illude a lot of broadcasters (like when they say "great pitch" when the pitcher completely misses his location and then gets lucky), by looking at the location of the target.

Someday, hopefully, the pitch f/x data will include that location. 

mgl8 says "...the batter needs to swing early on an inside pitch and swing late on an outside pitch in order to square it up.  That is just the physics of  'bat meets ball.'"

I am not understanding that statement (which is not to say it isn't true, only that I don't understand it).   For the bat to meet the ball squarely, the axis of the bat must be roughly aligned perpendicular to the pitcher-catcher line, regardless of where the pitch is located.  So, I don't see why the batter must swing early/late on an inside/outside pitch.  Can someone explain this to me.

Alan, aren't you the physicist?  I probably used the wrong words when I said "meet the ball squarely."  And "the physics of bat meets ball" was also a poor choice of words.

If a pitch is outside and you swing early, you will hit the ball off the end of the bat or miss it entirely.  You need to swing late to hit it on the sweet spot, and not more towards the end of the bat.

The opposite for an inside pitch (if you don't swing early, you will hit it towards the handle).  It has nothing to do with hitting it on the top, middle, or bottom part of the barrel of course.

I thought that what I was saying was obvious, but I guess not. 

No, I don't think you are missing my point, but I don't think we are in agreement.  The problem with your analysis is that the bat does not rotate along a fixed point.  If that were the case, then yes, a late or early swing would hit the bat at the same point regardless of where the pitch is.

On an inside pitch, the batter needs to "pull his hands in" (and clear his hips) in order to hit the sweet spot of the bat.  If he is late, he does not have time to do this.  

On the outside pitch, the batter has to move his hands towards the ball.  After a certain point in the swing, this is impossible.  You cannot do that if you swing too early.  You can still do that if you swing early (by "insiding out" the swing), but it will be difficult.  If you swing late, either on purpose or by accident, you will naturally be able to throw your hands at the ball (on an outside pitch) and swing "inside out."

I guess I did not make that clear at all and I apologize.  You have to set up a tee and ball and take some swings to see what I mean. 

When you teach baseball, you teach players that to hit the outside pitch, you have to swing a tad light and to hit the inside pitch, you have to swing a tad early.  Anyone who has ever played baseball (and remembers what I am talking about) knows that when you are late on an inside pitch, you will hit the ball off the hands (and your hands will hurt), and that in order to hit the inside pitch, you have to swing early, get your hips out of the way, and pull your hands in.