On May 29, the Rays faced the White Sox and lefty John Danks for the second time. In the first matchup, Danks struck out 9 and gave up just 2 hits in 8 innings. In the rematch, Danks only struck out 2 and the Rays would score 8 times. One difference was that Joe Maddon stacked the lineup with left-handed batters. Last week we saw a similar scenario in which JoeMa loaded the lineup with right-handed bats against the right-handed Shaun Marcum of the Blue Jays.
Why? Joe Maddon is kinda crazy. But also because Danks and Marcum throw changeups that are tough on opposite-handed batters.
Joe Maddon does a lot of unorthodox things. Some work. Some don’t. Filling a lineup with a bunch of lefties against a left-hander is certainly one of the more unusual moves. But does it work?
Tommy Rancel of DRays Bay dubbed this “The Danks Theory” and offered this explanation of the theory…
If an opposing starting pitcher has a specific above-average pitch that he throws to a certain type of batter, then let’s put a lineup together that neutralizes that pitch as much as possible, if such a line-up is possible.
The question is: How effective is The Danks Theory? While Maddon is neutralizing the strength of the pitcher, he appears to be doing this at the cost of neutralizing the strengths of his batters. That is, typically, a batter is more effective against an opposite handed pitcher (e.g. RHB vs LHP).
Rancel admits that testing the theory is nearly impossible based on sample sizes.
One way to analyze the theory would be to look at how the Rays performed in games under Maddon’s tenure in which The Danks Theory was employed. But there are problems with this method.
While some games may be obvious, others may not. How would we know if Maddon employed The Danks Theory with only one or two batters? And even in the obvious games like May 29, we can’t be absolutely certain how many of the batters were utilized for The Danks Theory. For example, Dioner Navarro batted right-handed against Danks. Do we include his stats or not?
Switch-hitters present an interesting scenario. Anytime a switch-hitter bats right-handed against righty or left-handed versus a lefty, then we know The Danks Theory (or something similar) is being utilized. Also, consider that a right-handed batter faces righties all the time. But switch-hitters are truly doing something unique under The Danks Theory. That is, a switch-hitter never bats right-handed against a righty.
So one way we can test The Danks Theory is to see how switch-hitters do under these circumstances compared to their normal production levels.
So let’s take a look at every switch-hitter* that has played for Maddon and how they have performed under The Danks Theory…
OK, not a lot of data here, so we really can’t draw any conclusions. But we do learn one thing: switch-hitters under Maddon have never batted left-handed against a left-handed pitcher. So all of the above data is when switch-hitters batted right-handed against a righty, as was the case for Navarro against Marcum.
So what can we do? Well, we can look at how switch-hitters as a group perform under The Danks Theory. Let’s take the total from the above table and compare that to how the group performs under normal conditions…
To calculate the career totals, we weighed the contributions of each hitter. For example, Ben Zobrist accounts for 15.8% of The Danks Theory plate appearances, so his career stats contributed to 15.8% of the overall totals.
So switch-hitters appear to be improving their performance when they bat right-handed against a righty. Again, still not a lot of data. In theory, we would like another 100 plate appearances before we can be certain that switch-hitters do better under The Danks Theory. But what is really important from this data, is that we see the switch-hitters are not performing worse under The Danks Theory.
And as they say, so far, so good.
Then again, maybe all this says is that Navarro needs to stop switch-hitting. But hey, we’ve been saying that since 2008.
*Four other switch-hitters have played for Maddon (Greg Zaun, Elliot Johnson, Greg Norton, Raul Casanova) but never had a plate appearance under The Danks Theory.