Though we all groan every time “columnists” like Bill Plaschke and T.J. Simers decide to focus their tired shtick on the Dodgers, I’ve always felt that beat writers Tony Jackson, Dylan Hernandez, and Ken Gurnick consistently produce solid work, often being the first to break news and generally providing an informed, balanced viewpoint to readers. That doesn’t mean they’re perfect – it was only a few months ago when I got on Jackson for employing “blown saves” when discussing Jonathan Broxton, for example – but even if the Dodgers don’t have the quantity of beat writers they once did, the quality is still there.
It’s with this in mind that Jackson deserves credit for putting pen to paper on a topic that’s long been known in the stats community, but which the casual fan has been slow to accept: sacrifice bunts are almost always the wrong choice.
The sacrifice bunt is the most counterproductive strategic ploy in the game.
And Mattingly, in apparent attempt to conform to conventional baseball wisdom, put it on twice in the final innings against the Diamondbacks, resulting in the Dodgers giving away at least one and possibly two outs at critical junctures of what became a one-run loss.
Yes. YES. A million times, yes. Jackson is referencing the final two innings of yesterday’s embarrassing loss to Arizona, a game in which Chad Billingsley and Kenley Jansen held the Diamondbacks to just one hit and an unearned run (partially due to a blown pickoff play which Jamey Carroll took blame for). The Dodgers, as has become an epidemic this season, managed just three hits from the regular lineup (Billingsley’s double was the fourth), dooming Billingsley to another undeserved loss on his way to a 10-13, 3.20 ERA season.
In the 8th, James Loney led off with a double (a momentous occurrence that probably deserved a post of its own), before being replaced on the bases by Tony Gwynn. Mattingly had Rod Barajas bunt him over to third, despite Gwynn being fast enough to score from second on most hits and despite Barajas being one of the only Dodgers showing any semblance of power.
Jackson summed up what happened next:
Except that the guy behind him, righty hitting rookie Jerry Sands, didn’t come up. Instead, Mattingly sent Dioner Navarro to pinch hit — now he takes the unconventional route, burning his backup catcher — to face the right-handed Hernandez because Navarro is a switch hitter, with all of his hits this season coming from the left side.
Only this time, Navarro didn’t get a hit. This time, he struck out on three pitches.
So much for giving yourself two chances to get the run in. Now, having chosen to simply give up one of your three outs, you are left with one.
Jay Gibbons popped out to left, and the threat was wasted. In the 9th, Carroll led off with a single, and Mattingly tried unsuccessfully to employ the sacrifice again, asking Aaron Miles to sacrifice Carroll to second. Miles couldn’t handle that task twice, and once he was finally allowed to swing away down 0-2, struck out. There’s a good chance Miles wouldn’t have grabbed a hit anyway, but two of the final six outs the Dodgers were given were simply thrown away by the employment of the sacrifice.
Here’s where I can help Jackson, though:
I have no data to support my case here, other than more than a decade of watching a big league baseball game almost every day and a long-building skepticism about the effectiveness of this age-old strategy. One of the arguments that could be made in Mattingly’s favor is that the Dodgers have such a weak offense that they have to try to create runs. I say the exact opposite, that their offense is so weak they simply can’t afford to be giving up outs when they have only three of them per inning to work with.
Jackson’s completely correct, but even if he doesn’t know it, there is plenty of data to back him up. Let’s head over to the run expectancy charts over at tangotiger.net, which allow you to select a baserunner/outs situation and is defined as “presenting the average number of runs that scored, from that base/out state, to the end of that inning”, based on the results of all games played over a certain time period.
When Loney (replaced by Gwynn) doubled to lead off the 8th inning, the Dodgers had a man on second with no outs. After Barajas’ sacrifice, they had a man on third with one out. Which situation usually results in more runs, based on every time each has happened from 1993-2010?
Avg. number of runs scored, 1993-2010
Man on second, no outs: 1.170
Man on third, one out: 0.989
How about in the 9th inning, when after Carroll’s single, they had a man on first with no outs? Mattingly’s goal was to get Carroll to second with one out, but when Miles failed, the result was a man on first with one out.
Avg. number of runs scored, 1993-2010
Man on first, no outs: 0.941
Man on second, one out: 0.721
Man on first, one out: 0.562
As you can see, in each case, the highest likelihood of scoring a run was to allow the next batter to swing away, not to give away an out in the hopes of securing one more base. In the 9th inning situation, the damage was particularly severe, since the sacrifice wasn’t successful.
There are, as Jackson notes, some rare times when the sacrifice bunt is appropriate. If, for example, the pitcher had been batting in either of these situations, then you can certainly make the argument to try to get an extra base to go along with the out the pitcher is almost certain to make. But as ineffective as Barajas, Sands, and Miles have all been at times this year, the better choice was to let them hit – particularly in the 8th inning, when the fleet Gwynn was already in scoring position and likely able to score from second on most hits.
I don’t know if Mattingly knows those numbers, but he disappointingly said after the game that it was an easy decision, one he’d do again. I’ve been generally pleased with Mattingly’s performance so far this season, but if there’s one area he’s proven to be infuriating, it’s with his use of bunts. As we saw yesterday, such decisions can directly harm the run production of a team that’s already terribly struggling in that arena.
No official decision has been announced yet, but it sounds more and more likely that Blake Hawksworth will be headed to the disabled list, and that his replacement will be Javy Guerra, who would be making his major league debut. As I said yesterday, there’s not much in the way of immediate reinforcements ready in ABQ, so it’s not a total surprise to see them dip down to AA Chattanooga instead.