Last week a user on the Fangraphs Q&A board was looking for pitcher's stolen base rates and mentioned Johnny Cueto's notorious ability to prevent the stolen base. I don't believe that stolen base rates for pitchers are easily organized at any of the major baseball stat suppliers (though I could certainly be wrong about this), so this morning I've decided to upload that data for any one interested, as well as providing some context for Cueto's notorious ability at both holding and picking off baserunners.
Looking back over the last ten seasons, from 2002-2012, Cueto ranks fourth best at preventing runners from swiping the extra base among pitchers with at least 900 innings pitched (an admittedly arbitrary number chosen only to include Cueto himself):
Pitchers with fewest SB Against since 2002:
Cueto's best season with regard to preventing stolen bases was also his most recent. In 2012 he allowed just one stolen base in over 217 innings, while nine foolhardy baserunners dared to test Cueto's quick delivery to the plate that year and were gunned down accordingly.
(I should note that the only stolen base Cueto did allow in 2012 was a bit of a "cheapee." It happened during a botched squeeze play by the Dodgers back in early July. It appears that after Cueto recognized Luis Cruz was breaking for home from third, he may have decided mid-delivery to throw the ball up and in, hoping to make his pitch unbuntable. Unfortunately, however, this just led to what very much resembles a wild pitch, but for whatever reason the play was ruled an official stolen base.)
Cueto's stolen base prevention in 2012 was his best season to date, but it certainly was no fluke. The Reds' right-hander has posted similar numbers in earlier seasons, but it appears he has fine-tuned his skills as his career has matured:
Johnny Cueto Stolen Base Rates Against by Season:
As impressive as Cueto's 2011 and 2012 seasons might seem, there have been exactly six seasons in the past ten years where a pitcher has allowed zero stolen bases over the course of a season in which they pitched at least 150 innings:
Best Seasons at Preventing the SB:
Josh Tomlin's 2011 season stands out amazingly as the only season in which no base runner even attempted to steal a base in any of his 165 innings pitched that year.
This is impressive, sure, but how much does it really matter?
If we add Fangraphs run values for stolen bases and caught stealing to Cueto's SB/CS rates, we find that frightened base runners may not be preferred to over-zealous base runners.
On average, a Caught Stealing was worth about -.4 runs in recent seasons, while a stolen base has remained steady at +.2 runs. If we apply these values to all pitching seasons in the last ten years with at least 150 IP, Cueto's 2012 shows up as tenth best:
Greatest wSB seasons for pitchers since 2002:
I suppose a drawback to preventing runners from attempting a stolen base is that you also prevent runners from getting caught stealing a base. Most of these elite seasons are elite by virtue of their high CS total, rather than strictly an ability to prevent stolen bases.
I find this very interesting. I think most of us are aware that major league teams generally are incredibly ineffective with their stolen base game. Don't get me wrong, teams have become exponentially wiser with regard to stolen bases over the past ten years, but we're still nowhere near the point where preventing stolen base attempts gives a tremendous advantage to the pitcher.
On average, a pitcher who compiled 200 innings during the 2012 season only lost .9 runs* to the stolen base game (assuming my math is right, I encourage you to double check). That would suggest that Cueto saved himself 4.4 runs compared the average pitcher by preventing stolen base attempts-- almost a half of a win. This is certainly nothing to scoff at, but it's important not to lose sight of how often teams still run themselves off the base paths.
But what about Cueto's amazing pick off move?
Absolutely, it is a thing of beauty. Retrosheet tells me Cueto picked off six additional runners where the play was not ruled a "caught stealing" and an out was recorded in 2012. Adding those six additional outs to Cueto's Caught Stealing column earns him another two and a half runs saved (6 x-.4) in his 2012 season.
But his excellent pick-off move comes with a bit of baggage that is seldom talked about.
As often as Cueto picked off runners in 2012, he also threw the ball away and allowed the base runner to advance on an error. Four of these six instances Cueto himself was charged with the error (with Votto and Costanzo being charged for the fifth and sixth instances). Live by the sword, die by the sword, I suppose.
If we count these advancements as valuable as a stolen base, we must then subtract another 1.2 runs (6 x .2) from Cueto's final runs saved tally, leaving us at about five and a half runs saved for the season.
Just give me the stolen base rates for pitchers, please.
Of course, back to the point of all this-- here is a spreadsheet with each pitcher-season since 2009 (min 150 IP) and an account of their SB and CS against, alongside this preliminary base runners runs saved metric:
(SB+ PK_errors)*runSB + (CS + PK_outs)*runCS
You'll see that Cueto's 2012 ranks third, behind only Verlander's 2009 and Shields's 2011--- still very impressive.
I'd love to be able to expand on this idea to get a better idea of how pitchers may be lowering their ERAs by holding or eliminating base runners. I think the next step is looking at RE24 values rather than our averaged run values, but I'm hoping the community here can offer your own suggestions.
. . .
* I used this formula, which was quick and dirty** and should probably be tweaked if taken to the next step: [(lgSB*runSB + lgCS*runCS)/ lgIP]*200. For 2012 that's [(3229*.2 + 1136*.398) / 43337.7]*200 = .894.
**When Fangraphs calculates their "wSB" for base runners they use opportunities (1B+HBP+BB-IBB). I'm pretty sure I should have used something similar to this rather than IP because a pitcher is actually preventing stolen bases by not allowing base runners. But if we want to measure pure ability to prevent SB, opportunities is probably the best way to go. In the spreadsheet I've labeled this method as "Runs_from_avg2" for reference. Cueto gains almost two runs saved under this method. Something else to discuss down the road.
Thanks to Baseball Heat Maps and Retrosheet for the data.