Rollins Was Better Than You Thought

I’ve been spending some time at Baseball Reference and FanGraphs, trying to figure out if it’s reasonable to expect Jimmy Rollins to rebound from a disappointing 2009 season. PECOTA is particularly bullish on Rollins, projecting him to put up an .808 OPS. The other projection systems have him in the .765-.780 OPS range, which is about his career average.

In perusing his stats, something has jumped out at me, an inescapable conclusion: 2008, not ’07 or ’09, was the outlier season for J-Roll. In fact, it seems to me that his ’07 and ’09 performances were eerily similar, just with different end results. Rollins’ OPS in those seasons were .875, .786, and .719. It appears at first glance that his ’07 was otherworldly, his ’08 was relatively normal, and his ’09 was part decline, part bad luck.

Let’s start with the walk and strikeout rates:

2007 6.3 11.9 0.58
2008 9.3 9.9 1.05
2009 6.1 10.4 0.63
Career 7.2 12.9 0.61

His BB/K in ’08 is at 1.05, the only time in his career he finished higher than 0.78. His ’07 and ’09 rates are very similar.

Batted ball rates…

2007 19.9 35.9 44.2 7.5 10.7
2008 24.0 45.4 30.6 11.8 7.2
2009 19.2 39.7 41.1 13.7 8.5
Career 21.5 41.8 36.7 10.6 7.9

His LD/GB/FB rates are similar again in ’07 and ’09, but he has nearly doubled his infield fly ball rate (IFFB%) from ’07 which signifies that Rollins is making worse and weaker contact, potentially a sign of decline.

Plate discipline…

O-Swing% Z-Swing% Swing% O-Contact% Z-Contact% Contact% Zone%
2007 23.7% 60.8% 42.9% 68.5% 93.6% 86.9% 51.6%
2008 20.3% 58.6% 39.7% 78.0% 95.1% 90.8% 50.6%
2009 22.9% 64.8% 43.8% 72.4% 94.4% 88.7% 50.1%
Career 19.9% 64.0% 43.1% 66.1% 92.7% 86.9% 52.5%

*Note: The prefix O- means outside the strike zone; Z- means inside the strike zone.

Once again, ’07 and ’09 look like each other. As his K/BB rates showed above and as his Swing% shows here, Rollins had improved his strike zone judgment. Somehow, he lost it last year.

Finally, to cap it off, defense.

DP Range Error UZR/150
2007 1.1 0.7 4.8 6.3
2008 0.4 8.2 4.3 15.0
2009 -0.4 -1.4 4.5 2.9

Surely the 15.0 UZR/150 from ’08 sticks out at you, but look at the individual components: double plays, range, and errors. Double plays and errors don’t differ much. However, Rollins’ range wasn’t much in ’07, expanded considerably in ’08, and then shrunk down again in ’09. It is possible that this is a data error on the part of Baseball Info Solutions (from which FanGraphs derives UZR), but more likely that Rollins simply had a fluky year.

So we now see the similarities, but there’s still that glaring discrepancy to sort out. In ’07, Rollins had an .875 OPS; last year, he had a .719 OPS. If those seasons were so similar, why the 156-point difference in OPS?

I have frequently pointed to BABIP when talking about pitchers such as Cole Hamels and J.A. Happ, explaining why their results aren’t in line with their performances. Pitchers only have a tiny bit of control over BABIP and there’s almost no year-to-year persistence, meaning that when we see a very high or very low BABIP, there’s no reason to expect it to stay that high or low going forward.

For hitters, it’s different — they do control their BABIP. David Wright’s BABIP, for example, is .345. Pedro Feliz’s, on the other hand, is .269. Rollins is about in the middle of the two at .293. However, last year, his BABIP was at a staggeringly low .251. While his ever increasing IFFB% is a factor, it doesn’t totally explain why his BABIP was so low. The projections agree that it was a bit too low, as all five systems project his 2010 BABIP at .274-.285. That seems reasonable.

It may sound weird, but going forward, expect Rollins to perform like he did in ’07 but with results somewhere in between ’07 and ’09.