Why Are the Twins Grounding Into So Many Double Plays?

Yesterday, Jon Marthaler asked What's with the Double Plays?, offering three theories as to why the Twins have grounded into a Major League leading 62 double plays through Sunday. Jon provided a number of interesting insights and got the discussion started. Today, I plan to take the investigation to another level of detail, looking at the number of double play situations, ground balls hit in these situations, and compared the Twins results to a Major League average "expected" number of double plays based on where each ball was hit and fielded. I'll go into more detail below, but here's what I found:

1. Opportunity: The Twins have been in a Major League high 441 "double play opportunities", defined as a runner on first base with less than two outs.
   
2. Ground Balls: The Twins don't tend to hit ground balls in these double play situations as often as one might think. At 31.3%, Minnesota is slightly above the MLB average (30.6%), 12th in MLB.
   
3. Speed: Based on the number of opportunities, ground balls hit and location of the ground balls, the Twins should have "expected" to ground into 49.5 double plays (3rd highest in MLB). With 62 actual double plays, the +12.5 difference is by far highest in MLB (Kansas City is 2nd with +6.4).
   
4. Mauer and Cuddyer: Joe Mauer has grounded into the second highest number of double plays relative to "expected" (12 GIDP, 7.1 expected), and Michael Cuddyer has had the highest number of expected double plays in MLB.
   
5. Defense: With a fly ball heavy staff and few walks allowed, the Twins haven't turned a lot of double plays (41, 10th in MLB), but they have been efficient, turning 7.1 more double plays than expected, second in MLB to Detroit (+8.3).
   
6. Morneau: The Twins have turned a MLB high 7 double plays on ground balls fielded by the first baseman. Considering that Justin Morneau has fielded 12 ground balls in double play situations and MLB average double play rate is 19.8% when fielded by first basemen, that's nearly five extra double plays just due to Morneau.
   

More details follow after the jump.

Baseline

I used league wide data from the 2009 season as a baseline for my calculations. Reviewing every at bat from the season, I found there were a total of 38,443 double play opportunities. During these at bats, batters hit a total of 11,769 ground balls (30.6%) and grounded into a total of 3,791 double plays (9.9%).

I could have chosen to use the overall ground ball rate to calculate "expected" double plays, but I wanted more detail so I calculated double play rates based on whether the ground ball is fielded by the first baseman, second baseman, etc:

The results aren't terribly surprising. Double play rate is highest when the shortstop or second baseman field the ball, but third base isn't too far behind.

Double Play Opportunities

Which teams have had the most and fewest double play opportunities in 2010? Most:

Fewest:

The Yankees and Twins are #1 and #2 respectively in OBP, so it's not surprising to see that they have had the most double play opportunities in MLB. Oakland is more surprising, as they are 22nd in OBP. On the other side, Houston (#30) and Toronto (#28) are at the bottom in OBP, but St. Louis (#12) appears to be an outlier similar to Oakland. I'm not sure why these teams have had so many opportunities relative to their OBP though.

.

For individual batters, the number of double play situations depends primarily on the two or three batters in front of him. Considering that Michael Cuddyer hits directly behind Justin Morneau (.493 OBP, #1 in MLB) and Joe Mauer (.400 OBP, #12 in MLB), it's not surprising that he has had the most at bats in a double play situation in the majors.

"Expected" Double Plays

Now that I have double play rates for each position, I can use this to calculate an expected number of double plays based on number of ground balls in a double play situation are fielded by the first baseman, second baseman, etc. This allows me to determine which teams or players are most efficient (or inefficient) in grounding into double plays given opportunities.

Teams most likely to end up grounding into a double play:

Teams least likely to end up grounding into a double play:

The surprising San Diego Padres, who have the best record in the National League, are the anti-Twins when it comes to grounding into double plays. The Padres have built a team that relies on speed in spacious Petco Park, so it's not too surprising they are at the top of the list. But it's also entirely possible that luck is also a factor in these numbers, good and bad.

Players most likely to end up grounding into a double play:

Players least likely to end up grounding into a double play:

I know Carl Crawford is fast, but he's hit a total of 20 ground balls in double play situations and not a single double play has been turned. On the other side, Joe Mauer has hit a total of 18 ground balls and 12 have been turned into double plays. Part of this is due to bad luck, as 17 of the 18 ground balls have failed to make it through the infield, but as commenters noted in Jon's thread, Mauer also tends to hit balls up the middle (11 of 18 ground balls were fielded by either 2B or SS), where a double play is more likely.

Infield Defense

I'll go into more depth on infield defense and ability to turn double plays at another time, but the data indicates that the AL Central is a double play haven, as Detroit (+8.3), Minnesota (+7.1) and Cleveland (+6.0) have turned the most double plays relative to ground balls hit in double play situations, one-two-three in MLB. Of course, this is partially because Minnesota and Kansas City are the two worst double play offenders in the league. On the bad side, the Dodgers (-8.5), Royals (-8.1) and Pirates (-8.0) are the least efficient at turning double plays.

Conclusions

I posted the primary findings from this analysis before the jump, but the data indicates that while the Twins have had the most double play opportunities in the majors, it appears that lack of team speed and perhaps luck is a large factor as well. Given a high OBP, the Twins can expect to face a large number of double play situations, but even accounting for these situations around 12 double plays have been turned that otherwise would not have been turned given an average team and average luck. Since the expected run difference after a double play is around a run lost, double plays appear to have contributed to about a 10 run loss, which corresponds to around one loss so far this season, or four losses over the course of a full season.