26 May 2015

Why Hasn't Manny Machado Taken More Walks?

As many of the Orioles' position players have slumped in recent weeks, one has gotten stronger: Manny Machado. He's hit 34% than the major-league average in May; along with Jimmy Paredes, he's provided all of Baltimore's offense in the current month. His .287/.350/.427 overall line, good for a .357 wOBA and 127 wRC+, makes him a formidable hitter, and gives off the impression that he's finally realized his potential.

One of the major factors behind this leap? Walks. Writing for FanGraphs a few weeks back, Jeff Sullivan noted that:
[...I]n terms of out-of-zone swing rate, Machado is showing by far the largest decline between 2014 – 2015, at more than 14 percentage points.
Since then, Machado's regressed to the mean slightly, but he still maintains a superb eye at the plate. After yesterday's game against Houston, a mere 58.1% of the pitches he's seen have been strikes; thus, 9.6% of his 178 plate appearances have ended in a base on balls. This massive increase in walks (he took free passes at a 4.6% rate prior to 2015) accounts for most of his breakout.

Something about this doesn't smell right, though. Out of 173 qualified hitters, Machado's 2015 strike rate comes in at 12th; his walk rate, by contrast, only ranks 57th. A similar disparity existed before this year — from 2012 to 2014, his 65.0% Str% was 173rd in baseball, while his aforementioned BB% was 210th. In his first three seasons, he underperformed his (for lack of a more descriptive term) expected walk rate, and that phenomenon has carried over into his otherwise vastly different fourth season.

To figure out the causes of this, let's take a dive into Machado's splits. The first three years of his career saw him swing more often as the count went into his favor:

Balls in Count O-Swing% Z-Swing% Swing%
 0 25.1% 53.1% 36.2%
1 39.9% 77.4% 55.0%
2 47.7% 84.3% 63.8%
3 54.7% 81.3% 67.7%

Such increased aggression would certainly explain why Machado accrued fewer walks than expected — he'd work the pitcher, make him throw a ball or two, and then pounce. And lo and behold, he's done the same thing in 2015:

Balls in Count O-Swing% Z-Swing% Swing%
 0 20.8% 47.3% 31.7%
1 23.4% 58.4% 36.8%
2 25.3% 75.0% 45.7%
3 61.5% 71.7% 67.1%

Most of the numbers themselves have obviously gone down, as Machado's become more patient overall, but the trend remains. Additionally, we see that in three-balls counts, Machado hasn't altered his approach at all, and offers just as much as he used to. Thus, like before, he's earned free trips to first much less frequently than his strike rate would suggest.

Does Machado's output with more balls in the count justify this erosion of restraint? Earlier in his tenure, it did, to a certain extent:

Balls in Count wOBA*
0 .301
1 .318
2 .329
3 .223

Machado hit better and better as the count became more hitter-friendly — until he got to three balls, at which point it all fell apart. These results would seem to condemn his strategy as a whole; in some ways, it worked, but the net production could have been better.

In case the numbers above don't make a convincing argument, Machado's 2015 marks probably will:

Balls in Count wOBA*
0 .328
1 .300
2 .302
3 .299

*Excluding walks

Aside from a possibly flukish jump at the beginning of the count, Machado has hit just as well with one ball as he has with two or three. (The major-league average wOBA without bases on balls usually sits around .280.) Based on this, Machado doesn't appear to benefit from his swing uptick.

Perhaps, if Machado maintained his moderation as he went deeper into the count, he'd see fewer hittable pitches, and his production later in the at-bat would drop off. But would the losses there compensate for the gains made by walking more often? We won't know unless he decides to try, and with the Orioles swinging as much as ever, that looks less and less likely every day. While Machado has the ability to garner free passes at a high clip, he may never realize it with this club.

No comments: