Ordinary average bunts

Runner on first, no outs. The table is set for a mistake.

Managers often see this situation as a great opportunity to bunt the runner over, but the numbers show they’re giving away runs and wins even when the sacrifice is executed perfectly.

By giving up an out for a base, teams lose more than they gain. Everyday, run-of-the-mill bunts cost teams in expected run scoring and the probability that they will win.

The most common bunt can be the most damaging type of sacrifice over the long run. The median win probability added of all 2,114 sacrifice bunt attempts in 2013 was -0.02, according to an analysis of data from the Baseball-Reference.com Play Index. That means teams’ odds of winning drops by 2 percentage points when they lay down a bunt.

  • The Cardinals and Diamondbacks were tied at 9 in the top of the 13th inning on April 3, 2013, when Matt Carpenter bunted over Jon Jay to second. The Cardinals went from having a runner on first with no outs to having a runner on second with one out, and their win probability declined by -0.02. Jay never scored, and the Cardinals lost 10-9.
  • With runners on first and second, no outs and trailing 5-3 on April 15, 2013, Luis Cruz of the Dodgers bunted the runners over and was thrown out at first. The Dodgers WPA dipped -0.02, and the Padres went on to win, 6-3.
  • Denard Span bunted over two Nationals’ players to second and third in an effort to come back from a 4-2 deficit. Again, the bunt was successful, but the Nationals lost -0.02 in WPA and lost to the Cardinals 4-3 on Sept. 23, 2013.

If these examples seem similar, that’s because they are.

The same story repeats itself over and over. Each time, teams lose fractions of a win when they bunt runners over.

Bunts may be worthwhile when the pitcher is batting or when a position player is fast enough to bunt for a hit.

But in most other situations, teams shouldn’t bother with the bunt. They’re killing themselves with smallball.

Bunts do more harm than good for most teams

Most of baseball’s managers harmed their teams when ordering non-pitchers to lay down bunts last season.

Only 13 of 30 Major League Baseball managers improved their teams’ chances of winning with sacrifice bunt attempts, according to an analysis of information from the Baseball-Reference.com Play Index.

The other 17 managers would have been better off letting their batters swing away.

In the 2013 season, the managers and teams that were best at bunting — both the decision to try a bunt and the execution of a bunt attempt — were Terry Francona of the Indians, Joe Girardi of the Yankees and Ron Roenicke of the Brewers. The managers who made the worst bunting decisions were John Gibbons of the Blue Jays, Mike Redmond of the Marlins and Walt Weiss of the Rockies.

The most profitable bunts occurred when the batter reached base, either by beating out the throw or when the defensive team committed an error. Bunts also made a positive difference in win probability when they advanced runners from first and second to second and third with no outs.

For example, Juan Centeno substantially improved the Mets’ chances of winning on Sept. 29, 2013, when he bunted in front of home with a runner on first and no outs in the bottom of the 8th. The runner, Juan Lagares, came all the way around to score, tying the game at 2. The Mets went on to defeat the Brewers 3-2. With a win probability added of 0.37, that play had the highest WPA of any sacrifice bunt attempt in the 2013 season.

The least useful bunt attempt happened when the Marlins’ Rob Brantley hit a popup to third that turned into a double play at first when Donovan Solano was caught off the bag. That play significantly reduced the Marlins’ chances of winning with a WPA of -0.27, and one strikeout by Craig Kimbrel later, the Braves notched a 3-2 victory on April 9, 2013.

This study excluded pitchers in order to compare National League and American League teams.

Sacrifice bunt attempts are one of the more easily measurable decisions that managers make, but of course this ranking shouldn’t be used to solely determine whether one manager is better than another. Many other managerial decisions include bullpen management, lineup configuration, intentional walks, stolen bases, training, scouting, preparation and intangibles, such as player motivation.

There was a low to moderate correlation between the number of sacrifice bunt attempts and their win probability added. The correlation was about 0.38.