When we talk about regression to the mean, we’re really conflating two concepts— regression of skill and regression of luck.

An obvious example of the former is, say, Manny Ramirez showing up for a Sunday day game with a hangover. He has a splitting headache, he can’t see the ball, he’s exhausted and the crowd noise is driving him nuts. Not surprisingly, he goes 0 for 5. On that day, his true OPS was zero. But, he’s Manny Ramirez— so the next day his true OPS is right back to normal. He has regressed to his mean.

Most injuries, slumps where guys have psychologically bad at-bats, and so on are not one-day affairs, of course. But the principle still applies— when a guy has played very well for a prolonged period, and then suddenly starts sucking, he’s likely to improve even though his bad results are entirely caused by bad skill and not bad luck. And on the flip side of this, if a guy has been “unlucky” but lacks skill, he’s not terribly likely to improve substantively, because one form of regression will mostly cancel out the other.

Salb probably knows all of this already, being a stat madman and whatnot, but I thought I’d point it out.

To apply this to the guys in front of us: Hannahan has been “unlucky” but appears to lack the real skill to hit for average, so he’s not likely to improve much. Ellis,by contrast, has a career BABIP of .290 or so, so unless his dropoff is related to his injury and that injury doesn’t get better, he’s probably going to be fine. Barton certainly showed skills in the minor leagues which suggested the ability to be better than barely league-average, but it’s hard to tell in his case where his true skill level is because he doesn’t really have an extensive pro track record at higher levels of play.

He has a line drive rate of 10% and struck out in more than 1/4 of his at-bats.

To go off what I wrote above, while we might hope for some skill regression on his part, there’s no luck regression involved at all. He got what his at-bats this season deserved— a horrible hitting line.

between flyballs and line drives.

Barton’s home run per flyball rate is 6.4. That’s around what a middle infielder would do. For comparison, Ellis’ homerun per flyball hit is at 7.3 % , Crosby at 5. For a guy who hits as many flyballs as Barton, and who is a 1b, that’s a pretty damn pathetic lack of power.

As sal posted, even if his BABIP were normal, in line with his line drive rate, he’d still need to either find (much) more power, or change his approach and start hitting more grounders and run faster, to be a decent hitter at 1b.

I forget what it’s called, but certain PGA telecasts give you the ball speed, launch angle, and track the flight of the ball.

This technology just needs to be installed in major league stadiums and then someone needs to compile the data. I would imagine this information will become publically available sometime in the next few years.

My subjective eye tells me that Barton is a guy that develops poor ball speed; that has been my concern with Barton for the past few years as I’ve gotten the chance to see him live — he just doesn’t hit the ball hard. Even in batting practice he rarely hits a ball hard. He hits balls squarely, but he rarely creates great ball speed to my subjective eye. Even his home runs — to my subjective eye — appear to be the confluence of a solid strike with optimal launch angle rather than high initial velociity. Line drive percentage won’t account for that, because the current data system is too crude to evaluate ball speed. In other words, not all line drives are equal in value, but the current system weighs them all equally. Barton could be unlucky, or the current data classification (which is itself nothing more than a subjective system in which we have decided to create three categories) is a poor predictor of Barton’s ability to avoid gloves of the high level of defenders that populate MLB.

My prediction is that ball speed data on Barton would show that the ball comes off his bat slower than the average hitter, and thus his “line drives” end up being caught more. Scouts are particularly adept at evaluating ball speed with their eye, they just call it something different, i.e., “the ball jumps off his bat.” Scouts have consistently questioned Barton’s power potential because he just doesn’t generate much ball speed. It is entirely possible that his overall mediocre numbers in AAA (look at the MLEs on his AAA stats) the past two years and the struggles this year are a result of his inability to generate ball speed.

Clubhead speed in golf, and the concomitant rise in ball speed, can be developed in golf, and it can be developed in baseball. Barton’s selective eye will hopefully allow him to succeed despite his lack of true impact ball speed — assuming my subjective eye, and those of the dozens of scouts that have said the same thing about him is correct — but my sense is that Barton is simply going to have to get stronger or more mechanically efficient if he is ever going to be an impact player. He may need to do so to even remain in the MLB. I eagerly await the data to test my subjective opinion, assuming of course that Barton has not increased his ball speed prior to the data being collected and available.

That’s meant for Fat Pubis (is it a CGV to type that?).