Perhaps the biggest decision new Mets general manager Sandy Alderson has made so far was to pick up the $11 million option for 2011 on Jose Reyes. Many considered this to be a no-brainer, but Alderson comes to the Mets fresh, with no emotional ties to the player who came up from the farm system. If Omar Minaya was still in charge, Reyes might even have an extension by now. However, Alderson is on the fence about keeping Reyes long-term.
Last year Reyes had a .321 OBP, his lowest since 2005. Recently, Reyes had been amazingly consistent in OBP, posting marks of .355, .358, .354 and .354 the previous four seasons. Reyes is a lifetime .286 hitter and last year he batted .282, meaning that his OBP plummet came essentially from a decrease in walks.
In 2010, Reyes had a 5.1 BB%, his lowest again since 2005. Let’s disregard 2009, as Reyes had just 166 PA that season. The other three years, he posted BB% of 8.7, 10.1 and 7.5. Even in his abbreviated 2009, Reyes had a 10.8 BB%.
My friend Matthew Artus wrote an article yesterday entitled, More Plate Patience Could Benefit Reyes and Mets. Artus talked about how the Mets as a team swung at more pitches outside the zone last year than they did in 2009. Normalized to league average, Reyes jumped from an 83 O-Swing% to a 110 O-Swing%, meaning he went from a disciplined hitter into an undisciplined one.
Generally, it’s not a good idea to swing at pitches outside the strike zone. In the particular case of 2010-vintage Reyes, it did not hurt him as much as you might expect. While Reyes’ BB% dropped noticeably, we’ve already seen that his AVG was essentially his career rate. Additionally, his ISO was virtually at career norms, a .146 mark last year compared to a .148 career mark. It was 30 points higher than the .116 he posted in Artus’ comparison year of 2009.
Furthermore, despite swinging at more pitches out of the strike zone, Reyes tied his career low in K% with an 11.2 mark, down from 12.9 in 2009. Interestingly, while Reyes had a career-worst 32.1 O-Swing% last year, his overall Swing% of 44.9 exactly matched his career rate.
So, Reyes swung at more pitches outside the strike zone and fewer pitches in the zone. Counter intuitively, it did not affect his AVG, his SLG nor his K%. It was still a bad tradeoff because his OBP took a massive hit. So, how did that happen?
The problem with Reyes last year is that he was too aggressive once he got to a three-ball count. Here are the breakdowns from FanGraphs on how Reyes did last year, along with his peak years of 2006-2008:
Reyes had fewer PA in 2010 than the other three years, so the percentage columns are the ones on which to focus. We already know that he had both lower BB and K percentages last year but it is interesting to see the actual raw numbers and percentages. The key thing is that the percentages of PA through the various counts are nearly identical.
Swinging at more pitches outside the strike zone, on a percentage basis, did not affect Reyes getting to a three-ball count, which is obviously necessary in order to draw walks. Let’s do another chart, this one comparing 2010 Reyes to 2006-08 Reyes.
|2010 PA%||0608 PA%||2010 BB%||0608 BB%||2010 K%||0608 K%|
The greatest discrepancy in three-ball opportunities between Reyes last year and Reyes at his peak was 0.6 percent thru a 3-0 count.
Of course, there is another way of looking at the data. FanGraphs gives splits based on through a certain count. Baseball-Reference breaks it down for every count. With the FanGraphs method, we have some double-counting. If Reyes got to a 3-0 count and took a strike, his PA would be counted in both through 3-0 and through 3-1.
In 2010, FanGraphs shows 148 PA through the various three-ball counts for Reyes. Baseball-Reference shows 99 total PA in which the count reached three balls. Here are the PA% for the Baseball-Reference method:
2010 – 16.4%
2008 – 18.5%
2007 – 19.6%
2006 – 14.4%
0608 – 17.6%
Here we see Reyes getting into fewer three-ball counts, as many as 3.2% in a single year and 1.2 percent compared to his three-year peak. This indicates that getting to a three-ball count was a bit of a problem. But how big was it? Was it more of an issue than what he did once he got to three balls?
Last year, Reyes had 99 PA where he reached a three-ball count and he drew a walk in 31 of them, for a 31.3 percent ratio. Here are the other numbers:
2008 – 46.8%
2007 – 51.3%
2006 – 52.5%
0608 – 50%
Let’s hold Reyes’ PA constant and give him the same three-ball percentage for his three year peak. So Reyes has 603 PA and gets to three balls in the count 17.6% of the time, or in 106 PA. If we hold his 2010 BB% of 31.3 constant, that gives us 33 walks, or two more than he actually recorded.
If instead we give Reyes his 2010 PA and three-ball percentage but now give him his three-year peak BB% of 50.0, he now has 50 walks (49.5) or 19 more than he actually recorded. What he did when he got to a three-ball count, even by the more-precise Baseball-Reference numbers, was by far the biggest determinant in his reduced walk rate.
There is one more way to look at the issue and that is by actual swing numbers. Last year, FanGraphs has Reyes with a Swing% of 44.9, his highest since 2006 but, as mentioned earlier, his career average. But when were these swings occurring?
TexasLeaguers.com gives the necessary data to break out your adding machine and figure this out. I plugged in all of Reyes’ three-ball counts and got the following for 2010:
3-0 count: 0-27 (0%)
3-1 count: 31-57 (54.4%)
3-2 count: 64-75 (85.3%)
Unfortunately, the site does not have the data for the rest of the years we’ve been examining, having it just for 2008. Here are those numbers:
3-0 count: 0-22 (0%)
3-1 count: 30-80 (37.5%)
3-2 count: 87-109 (79.8%)
The difference is not as large as I would have expected but it’s certainly possible the numbers were even greater in 2007 and 2006, which would make our three-year numbers differ more. Also, you will note that the numbers referenced from TexasLeaguers do not match the numbers from either FanGraphs or Baseball-Reference, perhaps another reason for a lower difference than I would expect.
Regardless, I feel comfortable at this point stating that Reyes’ OBP problems in 2010 were mostly the result of what he did when he got to a three-ball count. In his piece, Artus concluded in the comments section that Reyes’ more-aggressive approach was evident in his BB% through all favorable counts (1-0, 2-0, 2-1, 3-0, 3-1 and 3-2).
Artus said, “So Reyes converted only 31.3% of those 3-ball opportunities into walks, which is well below his 42.6% career average. It also means that he converted 15.6% of favorable counts into walks, which is also below his 21.6% career average.”
It is my belief that using favorable-count numbers as a comparison basis is the wrong method. We saw with the FanGraphs numbers the problem of “double-counting” and here that issue is magnified. Not every 3-0 count ends up 3-1 nor does every 3-1 count end up 3-2. But every 3-0 count started out 2-0 and every 2-0 count started out 1-0, so the effect is almost one of compound interest.
Ideally, Reyes would swing at more pitches in the zone and fewer outside. If he wants to remain with the Mets, he will have to do whatever it takes to get his OBP back into the .350s or higher, as Alderson will have no use for a leadoff hitter with a low OBP. The most effective way would be for him to be more selective when he got to a three-ball count.