Happy Valley Hardball with a sabermetrics twist

Penn State’s bottom of the ninth against Youngstown was so extraordinary that I’d like to take a few minutes to analyze it. For those you didn’t see or hear the game, the gist of it is that YSU had built a four run lead going into the bottom of the ninth but PSU came back to tie it in nine and win in thirteen.

First some sabermetrics background. Several years ago, sabermetrician extraordinaire Tom Tango analyzed several years worth of major league baseball data and developed a situation analysis view of a teams chances of winning a ball game based on the score, the inning, the number of outs, and what bases, if any, are occupied. So, he’ll tell you that if a team is batting in the bottom of the seventh, trailing by one run, but with the bases loaded and no one out, they will win the game 68.2% of the time even though they are trailing at that moment. Tango called this measure “Win Expectancy”. (UncleLar Note: Tango is just one of many who have developed similar stats – for an good read on the subject try Dave Studeman’s “The One About Win Probability” from the Hardball Times).

Using Tango’s formulas it’s possible to plot a ball game based on the results of every play in the game. There were 114 plays in the marathon 14 inning Youngstown State/PSU game on Wednesday night. Here’s a plot of Penn State’s winning expectation following every one of those 114 plays.

Basically, Penn State did little for the first eight innings. The Lions got mighty close to the bottom of that chart before turning it around in the ninth inning. In fact, there were basically two low points worth noting in the game.

In the top of the ninth, YSU was leading 4-1 and they had the bases loaded with no outs (that would be at play #72 in the chart). At that point, things are looking pretty bleak for the Lions. According to Tango, the Nits had only a 1.3% chance of winning the game in that situation (basically the odds were 75:1 against them). That’s when pitcher Mike Pierce really stepped up, striking out the next batter, throwing out the YSU runner at the plate on a squeeze bunt attempt, then getting the third out of the inning on a fly ball out.

While they dodged on huge bullet, the Nit’s chances of winning didn’t improve that much, only going up to 4.15% (or about 25:1) as they came to bat.

Here’s the play-by-play of that half inning:

Penn State 9th – M.. Yodice grounded out to ss. Ernst reached on a throwing error by 3b. Wine singled up the middle; Ernst advanced to second. Marlin walked; Wine advanced to second; Ernst advanced to third. Kelley singled to right center, 2 RBI; Marlin advanced to second; Wine scored; Ernst scored, unearned. Boonie pinch hit for Heath. Boonie grounded out to p; Kelley advanced to second; Marlin advanced to third. Borden intentionally walked. Nakata walked, RBI; Borden advanced to second; Kelley advanced to third; Marlin scored, unearned. Blackburn grounded out to 2b. 3 runs, 2
hits, 1 error, 3 LOB

Yodice’s ground out to start the inning is what led to the second low point I noted (play 77 in the above chart). Trailing by three runs, with one out and no one on in the bottom of the ninth, a team has a 1.77% chance of winning (a little less than 60:1). That’s when things really got turned around. Here’s a chart of the rest of that half inning (plays 77 through 85 in the graph above).

That ninth was a remarkable inning. PSU went from twice having less than a 2% chance of winning the game to actually being a 65% favorite to win as Blackburn was up with the bases loaded and two out. They didn’t pull it out then but they continued to play well in the extra innings (Pierce kept retiring the YSU batters and PSU kept getting runners on) eventually winning 5-4 in thirteen.

Huge props to the team for coming back that way.