- FanGraphs Baseball - http://www.fangraphs.com/blogs -

The Byrd/Ortiz Play

It wasn’t the most important play of the game, but I’m willing to bet that the play that generates the most talk around the baseball world will be this play, from the bottom of the 9th.

(Jonathan Broxton pitching) John Buck reached on fielder’s choice to right (Fly). David Ortiz out at second.

In case you didn’t see it, you can watch it here. Ortiz was waiting to see if Buck’s short fly ball would drop. It did, and Byrd managed to field the ball on a hop and fire to second in time for the force out. A player with any sort of speed would have easily been safe, but Ortiz was out by a slim margin.

Instead of two runners on base and one out, the AL was in the undesirable situation of two men out with a two run deficit against possibly the best pitcher in baseball. Let’s examine the impact that this play had on the All-Star Game.

According to Table 8 of The Book, with an average pitcher on the mound, the home team will score 2 runs and tie the game 11.1% of the time with runners on 1st and 2nd and one out. They will score 3+ and win 16% of the time, for a total win expectancy of 16% + 5.5% (they will win half the time in extra innings), which comes out to 21.5%. With a runner on first and two out, this number falls to 4.9% (5% tie, 2.4% chance win in 9 innings),which means that the difference in Ortiz making it safely to second and making the first out at second is 16.6% of a win – quite significant.

The numbers here differ slightly from what you’ll see in the Game Graph as The Book is based on the 1999-2002 run environment, but since it”s a difference of only about .4% of a win, it doesn’t make much difference

It’s key to note that this is if the AL All-Stars were facing an average pitcher, which they were not: CHONE projects Broxton for a 2.69 ERA, or 2.92 RA per 9 innings. (ZiPS projects a much lower FIP, but, for the purposes of the exercise, this works better. You can mentally adjust the numbers down if you want.) The Book also presents a run expectancy table for a team which would allow 3.2 runs per game, which is reasonably close to what Broxton and the NL All-Stars would probably allow, although a little high.

With runners on first and second and one out, the AL would be expected to win 16.3% (10.6% extras, 11.0% win) of the time, about 5.2% less than with the average pitcher on the mound. With a runner on first and two outs, that number falls to only 2.9% (3.4% extras, 1.2% win), or 13.4% worse than if Ortiz reaches second safely.

Broxton’s presence on the mound actually makes the impact of the play, purely in terms of the WE difference, lower than with an average pitcher, as it’s more likely that an out is made with Broxton on the hill. With an average pitcher, however, the AL would still be left with about a 1 in 20 chance of winning the game. Against Broxton, that falls to only 1 in 34.5.

Obviously, the biggest play in last night’s game was Brian McCann’s three RBI double, but this play will certainly be remembered as well. Against Broxton, a gaffe of the sort made by David Ortiz on the basepaths will almost certainly doom your team. Of course, Marlon Byrd deserves heaps of praise for coming up strong with the baseball and firing a near-perfect throw to second base, and there will certainly be much made of the decision by Joe Girardi to leave Alex Rodriguez on his bench as Ortiz plodded on the basepaths. But even somebody with Ortiz’s speed should have been able to reach second in that situation; with a proper read, we may not even be having this discussion.