Pronk vs. Proon: The Fight for the [Baseball] Soul of Travis Hafner

by crazymoloh on Apr 25, 2008 7:13 PM EDT

During the years 2004-2006, Travis Hafner hit like an MVP candidate. That version of Pronk disappeared in 2007. Instead, we were treated to Proon. Proon hit fewer line drives (17%), sent fewer fly balls out of the park (15%) and hit a lot more ground bals (48%). Not surprisingly, Proon wasn't an MVP candidate.

With Hafner's slow start beginning to worry people, we find his 2008 numbers to be a confirmation of our worst fears - further decline. However, I beg to differ. Strongly. What were are witnessing is not a deterioration, but a fight between Proon and Pronk for the baseball soul of Travis Hafner, like in Superman III.

Batted Ball Numbers:

Line Drive Rate
2008 - 25%
2007 - 17%
2006 - 21%
2005 - 20%

GB Rate
2008 - 41%
2007 - 48%
2006 - 39%
2005 - 43%

With the small sample size we have thus far, we are seeing signs of Pronk winning the battle. However, there are a couple of points where his numbers are off.....

Infield Fly per Fly Ball
2008 - 6%
2007 - 4%
2006 - 7%
2005 - 14%

Home Runs per Fly Ball
2008 - 27%
2007 - 31%
2006 - 16%
2005 - 14%

As bad as that looks, there is some good news here. First off, EVERYONE is popping up the ball at alarming amounts early in the season. Secondly, since Hafner's fly ball rate is back to pre-Proon levels, once he gets his timing in order a lot of those infield flies should (hopefully) turn into outfield flies (and hopefully a higher HR rate).

So why aren't those line drives and outfield flies turning into more hits? Believe it, or not, Travis Hafner is suffering from bad luck this season. His BABIP thus far is 0.267. Now, some players have a terrible BABIP because they hit very few line drives and a ton more grounders and fly balls, both of which are more easily converted into outs. 2008 Hafner isn't one of those players.

Using THT's 2007 Batted Ball Data (subscription only), I used a simple linear regression to find a correlation between LD%, GB%, IF%, HR%, non-HR OF% with a player's BABIP. The results weren't great, but they were good enough to give us an idea if a player is getting completely screwed by luck.

BABIP by Batted Ball Type (2007)
GB = 0.200
LD = 0.730
IF = 0.000
HR = 1.000
non-HR Outfield Fly = 15%

BABIP - expected BABIP Results
Mean Difference = 9.29%
Median Difference = 7.55%
Std. Dev. of Difference = 7.32%

Clearly the standard deviation isn't what one would hope for. Still, if a player's BABIP is off by his expected BABIP by more than 15% (mean + deviation), we can say, with confidence, that he has been the recipient of luck (good or bad).

From the results below, we can see that Pronk, Gutierrez and Dellucci have BABIPs much lower than you'd expect from their batted ball distribution. Look for their numbers to rebound. The opposite may be true for Garko......

0 recs | Comment 12 comments

Story-email Email Printer Print

Comments

Addendum

IF% = (IF/F)
HR% = (HR/F)(FB%)
nHR% = FB% – IF% – HR% (this is the number of fly balls that were neither infield flies, nor home runs)

NBR (No Borowski Run): Refers to that run which puts the Indians up by 4 over an opponent, thereby drastically reducing the chances of Joe Borowski pitching in a game.

by crazymoloh on Apr 25, 2008 7:16 PM EDT 0 recs

Once more....

IF% = (IF/F) x (FB%)
HR% = (HR/F) x (FB%)
nHR% = FB% – IF% – HR% (this is the number of fly balls that were neither infield flies, nor home runs)

NBR (No Borowski Run): Refers to that run which puts the Indians up by 4 over an opponent, thereby drastically reducing the chances of Joe Borowski pitching in a game.

by crazymoloh on Apr 25, 2008 7:16 PM EDT up 0 recs

This is really, really good stuff, crazy. Much more analytically sound than my stuff.

Couple of points… Even with a return to a normal amount of LD and BABIP luck for Hafner, can we reasonably judge at this point that his ceiling is something below that of 2006? I’m blisteringly tired right now, or else I’d do something to try to determine this on my own. But my gut assumption is that no player who is going to post an OPS of 1.000 over the course of a season is ever going to have a month of .670, no matter the luck.

I could be dead wrong about that. Again, really tired right now. Damn 3:30 alarm.

Regarding sample size, my concern is that 2008’s early trends could be an aberration, and he’ll regress to 2007-like numbers when it comes to LD. Man I hope not.

Anyway, thanks again. May the over-analysis of my favorite Triber continue.

by tabler84 on Apr 26, 2008 12:37 AM EDT 0 recs

Agreed. This batted ball distribution might just be a blip. But this blip resembles 2004-2006 more than people realize.

NBR (No Borowski Run): Refers to that run which puts the Indians up by 4 over an opponent, thereby drastically reducing the chances of Joe Borowski pitching in a game.

by crazymoloh on Apr 26, 2008 1:07 AM EDT up 0 recs

Look, it’s pretty simple. Hafner’s not hitting the ball hard now – or for the past 11 months -or at least not as hard as he was previously. Unfortunately I don’t get to actually see enough of him to come up with a reason for this. In fact I imagine if I saw every at bat he’s ever taken I’d know what the issue was. But aside from the IF% and nHR and such here’s what I wanna know: what’s his bat speed like now compared to 2006? You know they can measure this just like pitch speed. Somewhere in the Indians FO some baseball nerd’s got this data, but it looks like they haven’t let him out of his mother’s basement long enough to tell us what it is.

"the most vehement Yankee-hating guy I know" - Jay

by mauichuck on Apr 26, 2008 12:18 PM EDT 0 recs

I hereby submit that until further notice (or when he begins OPS-ing like a big boy), Pronk be known as Travesty Halfner.

by emd2k3 on Apr 28, 2008 12:45 PM EDT 0 recs

I took a look at Hafner’s numbers at fangraphs.com.

BABIP peaked in 2004/2005, and has trended down ever since. This seems to roughly coincide with the “shift.” He continued to hit Flyballs at a relatively high rate, and FB’s are affected by a shfit. But when you take a look at 2007/2008, his FB rate decreased (an lots of those FB’s are Infield Fly’s this year).

The GB/FB/LD breakdown is a reason for optimism, at least in regards to getting hits. He seems to be hitting the ball hard, just having some bad luck.

I say this having watched just 1 game, since I’m not a local. It sure is nice to get some real life opinions, instead of just looking at numbers.

What do you think about his broken hand from Sept 2006? It seems possible that a hand/wrist injury could have some long term effects that he’s just getting over.

Coincidentally, Ortiz had a little slide on BABIP from 2004-2006, which he reversed in 2007. His FB rate has remained high, however and his Isolated Power and Slugging Pct has remained incredibly high (2008 not withstanding).

by rogue409 on Apr 28, 2008 1:42 PM EDT 0 recs

I think it’s funny how often I found myself comparing Hafner to Ortiz. It wasn’t good enough that we had a 1000 OPS hitter. He had to be better than Boston’s 1000 OPS hitter. I’ll tell you something, these days, that Diamond Jim Gentile comparison’s looking mighty eerie.

by fleerdon on Apr 28, 2008 3:46 PM EDT up 0 recs

That really is scary. I had forgotten that he bowed out with Cleveland in 1966, washed up at 32.

by ken from alexandria on Apr 28, 2008 4:25 PM EDT up 0 recs

Ortiz was just the first reasonable comparison that I thought of. Power-hitting, “slow”, left-handed DH/1B.

I thought about looking at Giambi too, but there’s a whole ‘nother can of worms there.

I’ve heard “The Shift” called “The Papi Shift” and “The Giambi Shift,” but I’ve never heard it called “The Hafner Shift.” I think ESPN just hates you guys.

by rogue409 on Apr 28, 2008 5:13 PM EDT up 0 recs

Hafner is not particularly slow, but Ortiz is a good comp anyway.

by Jay on Apr 28, 2008 5:42 PM EDT up 0 recs

I learned about Projected OPS today (from fangraphs.com)

It projects a players OPS based on their batted ball composition. All the Indians hitters except for Martinez have been a little bit unlucky this year.

With average “luck”, Pronk should have an OPS of 0.832. In other words, he looks like Pronk-2007, not Pronk-2006.

Sorry.

by rogue409 on Apr 29, 2008 11:51 AM EDT 0 recs