Steven Souza's strikeouts: A PITCHf/x analysis

There's no denying the fact that Steven Souza Jr. has struck out too much to start the 2015 season. It's great that he hits for power (.173 isolated power and five towering home runs), and his 14.6% walk rate is pretty neat, but a 36.2% strikeout will make it hard for anyone's bat to contribute. Moreover, as the first top-level offensive statistic to stabilize, we can be reasonably sure at 130 plate appearances that the high strikeout rate Souza has shown is an accurate reflection of how he has played.

Now it's easy to say that Souza is a rookie and that he needs to adjust to major-league pitching. As I've pointed out before, Souza has only faced slightly over one full season's worth of double-A and triple-A pitching combined, so it makes sense that he'll need time to figure out how to battle high-quality hurlers. But simply saying that he needs to adjust isn't the most helpful observation. HOW does he need to adjust? WHY is he striking out so much?

Let's start with the PITCHf/x plate discipline numbers from FanGraphs. Below is a comparison of Souza's approach and results over his short career, compared to the entire league in 2014.

	The League (2014)	Steven Souza
Zone%	48.8%	48.1%
Swing%	46.2%	42.4%
Z-Swing%	63.1%	64.7%
O-Swing%	30.1%	21.7%
Contact%	79.3%	69.5%
Z-Contact%	87.5%	81.2%
O-Contact%	62.9%	37.3%

Zone%: So first off, he's not being pitched much differently than average. He's seen about the same total number of pitches in the zone as the rest of the league.

Swing%: As you would expect from someone who walks a bunch, Steven Souza takes a few more pitches than average. The overall swing% doesn't tell you all that much though, because it doesn't say whether he's taking the right pitches. That's next.

Z-Swing%: This is the percentage of pitches within the strike zone that a player swings at. You want your hitters to swing at these often, and we see that despite what appears to be a passive approach overall, Souza is actually slightly more aggressive than the average hitter within the zone.

O-Swing%: And here's where Souza makes up for his aggressiveness within the zone to arrive at the low overall swing rate. He offers much less often at balls, swinging only 21.7% of the time. So far, we don't have an answer for why he's striking out. These numbers look like a player with an excellent batting eye, who we would expect to succeed.

Contact%: Now we're getting somewhere. We can see that Souza makes less contact than the average player. That is to be expected, as there's a tradeoff between contact and power, so this isn't necessarily something to get alarmed about.

Z-Contact%: Once again, this is a high number of whiffs, but it's not a debilitating one for a guy with real power like Souza has. Plenty of useful players, like Matt Kemp and Todd Frazier, were right in this neighborhood in 2014.

O-Contact%: Ding ding ding! This number is abysmal. Of the players to receive 130 plate appearances (what Souza has so far) last year, only five made contact outside of the zone less often. They were Jordan Danks, Junior Lake, George Springer, Alex Avila, and Justin Ruggiano. That's not healthy company to be keeping.

Going Granular

So, having established that Steven Souza's biggest problem is an inability to make contact with the pitches he swings at outside of the zone, let's get into the minutia of how this is happening. To help, I created a visualization of all pitches Steven Souza's received, using data pulled from from Baseball Savant, but with slightly tweaked pitch types and results. I've drawn a rule book strike zone, but keep in mind that the real strike zone is a bit different. Righties like Souza generally tend to have slightly shorter zones than normal, and to need to protect a slightly wider area both on the inside and the outside of the plate than the rulebook would suggest (while lefties have to protect a significantly larger area on the outside).

Click over to the Souza Strikeout Explorer (TM) and play with the filters, and then I'll walk through some of the more interesting points.

Filters: RHP, Two Strikes, Batter Swinging

I like working with two-strike counts because it's clear what the batter needs to do—avoid taking strike three. Other counts can muddy the picture. For instance, if you were to include zero-strike and one-strike counts, you would see Souza whiffing on more pitches within the zone, but when he gets into a situation where he's trying to protect, he cuts down on those swings and mostly makes contact whenever he's presented with a strike.

He's clearly getting fooled on pitches over the outer half of the plate but below the strike zone, though. And it's not just on one pitch. In that mix are two fastballs, one sinker, one (or maybe three depending on what you count) changeup, four sliders, and two curves. That's the only location outside of the zone he seems to offer at in two-strike counts with any regularity, and when he does, he almost always misses.

Filters: RHP and LHP, Two Stikes, Batter Swinging or Taking a Ball

Now I've added in the pitches called as balls to the same graph. The point here is that while Souza shows a pretty good ability not to offer at pitches outside the zone in most of the areas, proving that he has a discerning eye, it's pretty much a 50-50 proposition down and away. Throw a pitch there—any pitch really—and he might take it for a ball as he should, or offer at it and miss.

This is the answer to the low O-Contact%, but let's press on because there's one other interesting thing I found.

Filters: RHP and LHP, All Counts, Batter Taking

This isn't the most useful set of filters, but I throw it in for comparison to what comes next. You can see what I mentioned before about the actual strikezone for a right-handed batter being a little bit wider than the rulebook zone on both sides. You can also see that Souza may have gotten the rookie treatment on a couple sliders down and away, but that's not the biggest of deals.

The real point I want to make is that Souza, like most batters, takes some pitches right down the middle. That's okay. It's normal. It means he was looking for something else, or he saw it wrong coming out of the hand. It's just a natural byproduct of being selective, and I don't believe Souza should radically change his approach in this case.

But look what happens with the next filter.

Filters: RHP, Two Strikes, Batter Taking

Once more I've limited the data to two-strike counts only. That's the important part. I've also limited it to right-handed pitchers, because the lefties don't matter at all for the area of concern I'm highlighting, and removing them makes it a little bit easier to read.

No longer do we see a player taking pitches within the entire zone. Rather, we see Souza swinging at everything that he should, except for one pitch that he's failing to recognize: belt-high fastballs on the outer half of the plate. Note that from a right-handed pitcher, the horizontal movement of a fastball will bring it back over the plate from the outside, so I think there are two things going on here. First off, Souza is not protecting a wide enough zone. Secondly—and I say this because we see this problem "take spot" with fastballs and not with other pitches—Souza seems to be having trouble accounting for the movement of the outside fastball. He sees it as off the plate out of the pitcher's hand, but then he's being surprised when it loops back over the black.

When you think you see a pattern in an individual player, it's always a good idea to compare that pattern to league average to make sure you're not simply overreacting to the way baseball works. For that, I really like Jeff Zimmerman's tools at Baseball Heat Maps.

Below is a graph of all the take/swing pattern of Steven Souza this season on all fastballs thrown to him by right-handed pitchers in all counts, compared to league average, and with a 20 percent regression rate. Green means that he has swung at a pitch in that location about the same amount as the rest of the league, hotter colors mean he was ore likely to swing at a pitch in that location then the league, and cooler colors mean he was less likely to swing at a pitch in that location.

Sure enough, we see that there's a big blue spot on the outer edge of the plate right where we saw it in the Souza Strikeout Explorer. The pattern is real. Souza does in fact take those fastballs for strikes more often than he should, and more often than we would expect him to.

Conclusion

There are two main issues driving Steven Souza's high strikeout rate. The first is an abnormally high whiff rate on pitches outside of the zone, and this is primarily attributable to getting fooled on pitches over the outer half of the plate but below the zone. That's a tough pitch for anyone to hit, and when Souza tries to hit it, he nearly always fails.

The second issue is that he's taking too many fastballs on the outer portion of the zone. This doesn't show up in the overall plate discipline numbers because Souza has such a good conception of most areas of the strike zone that he still looks above average in terms of discipline even with this persistent error, but it's a hole in his approach, and it's one that is being exploited.

I'm not a hitting coach. I don't know how to take these findings and present them to Steven Souza in a format he can process and use at the plate, but the goal is simple. He needs to be more discerning vertically so as to not offer at the pitches below the zone that he can't hit, and he needs to be more aggressive horizontally so as to cover the outer portion of the plate that will almost always be called a strike. If he can't do that, he might still be a useful player because of his walks, power, and speed, but if he can make the adjustment, Steven Souza Jr. will be a star.