Four Factors: Hanley Ramirez and regression


We all know that Hanley Ramirez has had something of a down year this season. Despite still being quite a valuable commodity (2.9 WAR is nothing to sneeze at), he has struggled on defense according to UZR and has hit just .284/.366/.451 (.356 wOBA) this year, well below his ZiPS-projected .320/.393/.527 line predicted before the season.

There have been two reasons for Hanley’s well-documented offensive struggles. One of those is BABIP; his current .305 mark is the lowest of his career and seems pretty out of line given his speed and ability to spray the ball. The other is a lack of power, as his .167 mark is also the lowest of his career. Which of these two parts is hurting his game more when compared to his average values, and what happens if he improves going forward? To evaluate this question, I’m going to use the tool made available by Fangraphs author and friend of the Maniac Jack Moore: the Four Factors spreadsheet.

The effects of the Four Factors

The Four Factors involved in the analysis are walk rate, strikeout rate (per PA), POWH (extra bases per hit, a stat I’ve been using a bit lately when talking about Mike Stanton), and BABIP. The two factors that have been a problem this season are BABIP and POWH. Let’s look at this season’s numbers compared to Ramirez’ career averages and ZiPS-projected rest-of-season line.


I also provided the ffwOBA for each Four Factors line; that is a projected wOBA based on just those factors. It is not as accurate as just calculating wOBA from projected component stats, but it serves as a decent approximation (credit Jack Moore for the math work deriving the equation for ffwOBA).

Again, the problems seem very obvious looking at these stats. The walk and strikeout rates are close enough not to be too concerned; in fact, Ramirez has struck out less this season than he has in any year since 2007. The drop-off in power and BABIP are quite staggering. But which one seems most damaging?

Let’s start by looking at that lack of power. I used Hanley’s 2009 and projected numbers for each of the other Four Factors and varied his POWH, starting from his current 2009 low up to his career high .797 in 2008. It produced this following graph, first using his 2009 numbers.

If Hanley could achieve the same sort of power mark he had in 2008, when he hit 33 homers and 34 doubles en route to a .301/.400/.540 (.405 wOBA) season, the Four Factors estimated wOBA would reach .384; given the approximation element of this method, this could be in the .380-.390 wOBA range. What if we take his rest-of-season projected values?

At a more typical Hanley BABIP of .333, you can see that his value would be dramatically increased. This lack of power would still result in a .373 estimated wOBA, about the same wOBA that he would get if he had a .720 POWH and a .305 BABIP. With Hanley’s career best POWH from 2008, he would have an estimated .404 wOBA, quite fitting given the fact that, in 2008, he had a .405 wOBA with a .329 BABIP.

What about if we changed his BABIP and kept his power stable? Let’s take a look at how his 2010 numbers would translate with increasing BABIP, starting from his career low .305 now and going up to his .379 mark from 2009.

It would seem that, with Ramirez’ ability to get hits on balls in play, he will probably benefit a lot more with regression in this regard than in the power department. With the other factors at 2009 values and his BABIP at a projected .333, the estimated wOBA is .379, which is almost 30 runs better than average in a given season. What would we expect to see in terms of projected numbers?

Obviously the ceiling jumps a bit, with the 2009 BABIP leading to an estimated .421 wOBA (50+ runs above average in a season). The basement wOBA of .358 shows just how much changes as his power goes back up to the projected value; at a .305 BABIP and a projected .660 POWH, Hanley’s numbers only go up from .355 to .358 (though there is an increase in strikeouts as well).

Which factor is hurting Ramirez the most?

Just looking at the graphs, it would seem that BABIP is the more damaging of the two factors with which Ramirez is struggling in 2010. Changing the POWH from the current .587 to the ZiPS-projected .660 yields an estimated wOBA increase from .355 to .365, or about five runs above average in a season. Bringing the BABIP up from .305 to the projected .333 brings a much greater increase in wOBA, from .355 to .379, a difference of about 12.5 runs in one season.

Which is more likely to regress? I’d say both are equally likely to be on the upswing. Ramirez is still pounding the ball on the ground at a career-high 53.6% rate, but that should normalize as we continue. Hanley’s also picking up fewer hits on both grounders and fly balls, and I’m confident that those will fall a bit more and up the average. While this season has been a bit of a disappointment for Ramirez, it is nothing to be worried about. He has already shown he’s getting at least his power back, as he’s hit .298/.385/.544 this month. Hopefully, he’ll continue to be on a tear in the coming weeks and end the season with at least a 4-WAR campaign.