One of the great things about the explosion of data that has taken place in MLB over the last several years is that it’s pretty easy to check on things for which we previously would’ve had to rely on instinct or guesses. Want to know how often Vinnie Pestano throws a fastball to a right handed hitter? Boom. Want to see how Carlos Santana’s BABiP progressed through the minor leagues? Sure you do. What about the last 20 times a pitcher threw a complete game shutout with no strikeouts? Look no further.
This is all super neat, but if we’re not applying the data against any assumptions—not using it to test hypotheses—then it’s really just daturbation.
Which is why I like reading stories from Spring Training that suggest a player has “changed his approach”, and then checking months later to see if any change is evident. Needless to say, this is usually nonsense fed to a reporter looking for a story. Orlando Cabrera is in the best shape of his life! Matt LaPorta finally has an approach against breaking balls! Ryan Garko looks great shagging flyballs in the outfield!
But sometimes, those Spring Training stories actually contain a kernel of truth, and I wonder if we didn’t witness such an instance this past March. I remember reading during Spring Training that Tomlin had decided he might have become a bit too predictable, and was thinking of reducing the rate of 1st pitch strikes he’d offer to opposing hitters. It seems that Tomlin thought he might be able to induce more eventual swings and misses if he threw more pitches out of the strike zone.
So I thought I’d check on some numbers, to see if there might be any evidence of a change in approach. I was a bit surprised by what I found:
This would be the time to mention the requisite warning about small sample sizes, but you really are tempted to see two different pitchers in these numbers, or at least two different iterations.
Without confusing correlation and causation too much, let’s note that Tomlin’s swinging strike rate is up considerably, while his 1st pitch strike rate and overall strike percentage are both down. 1 This could be evidence that by throwing fewer pitches in the zone, Josh is getting batters to chase pitches, resulting in more swings and misses.
Of course, one of the consequences of throwing fewer strikes is that you’re bound to walk more hitters, and that shows up in the data as well. Last year Tomlin was historically stingy with the free pass, so we might expect some regression in that department, and sure enough he’s up more than 2 percentage points over 2011—from 3.2% last season to 5.4% so far this year.
So if his strikeout rate is up AND his walk rate is up, we might think that nothing substantial has really changed. I have spent a good deal of time trying to convince you that the ratio between a pitcher’s strikeouts and walks is one of the most important measuring sticks we have at our disposal, no matter what “type” of pitcher he might be. For Tomlin’s career he’s at 3.33 and so far in 2012 he’s at 3.44. In other words, no big change, since both rates are up.
But the other factor that plays into a pitcher’s success—outside of his strikeouts and walks—is his ability to limit home runs. You’ll notice that so far in 2012 Tomlin’s HR/9 is down from last year, a good sign for someone who seems destined to struggle with the long ball.
Could it be that by throwing more balls and keeping batters off-balance, not only is Tomlin increasing his K-rate, but reducing his liability against home runs?
Well, sure. It’s possible. But it’s not all that likely. A deeper look will reveal that the rate at which Tomlin’s flyballs have become home runs this season is 11.9%—UP from his 11.1% from last year and his 10.4% career rate. It wouldn’t seem likely, if Tomlin were effectively reducing solid contact, that more of his flyballs would become home runs. We’d tend to think the opposite might be true.
In all honesty, it’s probably far too early to draw any substantial conclusions from all this data. If Tomlin allowed only one more home run this season, his HR/9 would be higher than it was last year; one fewer and it would be below the league average. We’re still in that territory where one or two events can drastically shape the numbers, and we’d be wise to remember that before we go telling ourselves stories about being “effectively wild”, even if it feels right.
On the other hand, I do wonder if the jump in Tomlin’s strikeout rate isn’t something that might be sustained. He’s recorded 31 strikeouts in less than 40 IP so far this season, and while that’s not anywhere close to impressive, it does lead the Indians among starters not named Zach McCallister. That tidbit might be more of an indictment of a particularly weak staff than praise for Tomlin—only the Twins’ starters have a lower K/9 than the Tribe’s 5.31.
But Tomlin was not likely to continue the other-worldly control he exhibited last season, when he led the majors in BB%. Even a slight uptick in his strikeout rate will give him the cushion he might need to make it through the regression he’s likely (already) seeing with regard to walks.
- Just to be clear, I’m using the BIS strike%, which measures whether a pitch was actually in the strike zone or not. This figure wouldn’t count a pitch out of the zone that a batter swung at, and it doesn’t care at all what an umpire says. In other words, it won’t correlate with the figures you’re used to seeing on the telecast. Also: FOOTNOTES!! [back]