Is Z-O Swing% a Better Indicator of Plate Discipline Than O-Swing%?

In some FanGraphs articles, Z-O swing percentage is thrown around as a measure of plate discipline. That makes sense because generally when a hitter swings at strikes, good things happen, and if he swings at balls, bad things happen.

To test if that stat is really better, I looked at the 2017 leaderboard. I looked at the wRC+ of the top 30 and bottom 30 hitters with Z-swing%, O-swing%, and Z-O swing%. Here is what I found:

wRC+
	z-o swing	z swing	o swing
top30	122	112	122
bot30	103	105	96
all qualified	110	110	110

There is a slightly positive effect of Z-swing, but a much stronger effect of both Z-O swing% and O swing%. At the top, the low-chaser and high-differential guys do about the same, while the bottom chasers do even worse than the bottom differential guys.

If you widen the search for top half and bottom half you get that picture:

z-o swing	z swing	o swing
119	110	117
102	110	103
110	110	110

Z-swing has no effect at all, and the differential is slightly better than the chase rate, but not by much.

Overall, the Pearson value for differential was a positive .42, for the chase rate it was .32 (used 100 – O-swing% to get positive value), and for Z-swing there was almost no effect (.07). So the differential is a bit better, but the effect isn’t huge; it is probably like with OPS+ and wRC+ where one is mathematically more elegant and correct but the actual values won’t differ much.

I also dissected the hitting into the components OPB, ISO and BABIP.

 

ISO
	z-o swing	z swing	o swing
top30	.220	.210	.200
bot30	.170	.180	.163
all qualified	.193	.193	.193
OBP
	z-o swing	z swing	o swing
top30	.360	.335	.370
bot30	.324	.350	.317
all qualified	.341	.341	.341
BABIP
	z-o swing	z swing	o swing
top30	.308	.309	.306
bot30	.304	.303	.306
all qualified	.306	.306	.306

The result is quite interesting. The differential (+27 ISO points) does clearly better in the power department than chase rate (+7); in fact, even Z-swing had a more positive effect (+17) on power than a low chase rate.

With OBP, that is reversed. Here, the chase group does better than the differential group, while a high Z-swing rate has a negative effect.

With BABIP there was a very small positive effect of differential and Z-swing, and no effect of the chase rate, but the effects are almost non-existent.

So we seem to have two opposing effects here. Being more aggressive in the zone helps the power but seems to slightly hurt the OBP (of course there probably is a bias that aggressive hitters in the zone are often also aggressive outside, but still). And for OBP, chase rate clearly is king, while it doesn’t really have an effect on power.

Still, that might have an effect for certain hitters and especially pitchers, but overall the advantage doesn’t seem to be big, even though it is a bit due to coincidence due to the opposing effects.