Baseball is full of platitudes. Play the right way. Speed never slumps. Pitching wins titles. Pound the zone. Much of the work done by sabermagicians is centered on testing those platitudes. And many times their research shows us that there are many different ways to play, that speed does slump, and that you need offense to win titles, too. Just because it’s been said for decades doesn’t make it true.
Then again, there are platitudes that are shown to be true. “Get strike one” might be one of them.
Nathaniel Stoltz over at Beyond the Box Score is looking towards creating a defense-independent pitching number using per-pitch and zone numbers right now. In other words, he’s trying to take the results from something like “strikeout rate” down into the components of strikeout rate — things like swinging strike rate, and zone percentage. It’s possible he could be better served by correlating his work to runs allowed instead of another secondary metric, but he’s on a mission.
If you’ve been following along, you know that there’s a possibility that the swinging strikeout is more important for forecasting future strikeouts than the called strikeout. This could improve our current understanding of pitching estimators. So good luck to Stoltz.
What I found interesting for our purposes was the fact that, of all the possible pitcher peripherals related to control that he could have used, he used first-strike percentage. It correlated well enough with SIERA, but that got me thinking. What if we could create an expected walk rate? That would be interesting to all sorts of baseball fans, and certainly to fantasy freaks.
There aren’t a ton of peripherals at our disposal if we were going to suss out the truth in a pitcher’s current walk rate. The two that came to mind immediately were zone percentage and first strike percentage, thanks to the original article. So I ran some correlations trying to figure out how these related to a pitcher’s walk rate.
The answer? Zone percentage isn’t great — perhaps because great pitchers can get swinging strikes on balls that aren’t in the zone to begin with, turning a ball into a strike — but first strike percentage is great. It explains almost half of the variance in walk rate! I didn’t make it all the way to an expected walk rate, but I did identify the situations that should result in a good walk rate: a high first-strike rate, and if you have a low zone percentage, a low contact rate and high o-swing percentage.
Yes, those describe ‘good pitching.’ But we proved a platitude, so of course we were going to come full circle in some way.
Let’s make this roto-relevant. Are there any pitchers that satisfy those conditions this year but still have a bad walk rate? Perhaps they could improve that peripheral and be more useful in the future. Here’s a list of all the players with a walk rate above eight percent and a first-strike percentage above 59.5%. League averages are listed last.
|Name||BB%||F-Strike%||Zone% (pfx)||O-Swing% (pfx)||Contact% (pfx)|
Using first-strike percentage alone, Matt Moore and Erik Bedard leap out from the list. They’re a full ten percent better than league average at getting strike one, and we’ve seen how important that is. If only Matt Moore got more swings at pitches outside the zone, he’d be perfect for this analysis. As is, he has a really nice contact rate, an above-average zone percentage, and gets strike one. He’s a great candidate for a better walk rate in the future.
Trevor Cahill actually satisfies all of the requirements. It’s been since his rookie season since he’s had a better-than-average walk rate, but right now, he’s getting strike one and whiffs at pitches outside the zone. Maybe that walk rate will improve as we go forward, too.
Some of these pitchers satisfy the requirements, but it’s hard to see them improving much more than they already have. Perhaps it’s the fact that Jeff Samardzija is doing these things that has lead to his current walk rate — he used to put up walk rates that were much, much worse. You might say the same about some of the other pitchers who already have a walk rate close to the league average. How much better is Chad Billingsley going to get, for example. You do still have to consider a pitchers’ previous walk rate (Matt Moore’s were excellent in the minor leagues, just saying).
And even though he didn’t make the list, it’s worth mentioning that Tim Lincecum is pretty close to having the right numbers to suggest his walk rate could improve — he’s walking a career-worst 11.9% of the batters he’s faced, and he has a slightly below-average first-strike rate (58.9%), but his o-swing % is great (29.9%) and batters are still failing to make contact (73.6%).
This research has not yet resulted in an expected walk rate. But that doesn’t mean we can’t remember the importance of strike one when we’re looking at pitchers struggling with a bad walk rate. After all, you probably hear about it all the time on your local sportscast.