SOX FEELING ALL-STARS’ PRESENCE
Trio of pitchers has added mightily to their win probability
The selections of Carlos Rodon, Lance Lynn and Liam Hendriks to the American League All-Star team gives due recognition to the White Sox’ dominant pitching staff.
That trio ranks high whether looking at traditional stats or newer metrics. Lynn, in particular, ranks high in win probability added (WPA), a measure of how much the plate appearance of each opponent has added or subtracted from the Sox’ chance to win.
Through Sunday, Lynn (2.02) and Rodon (2.37) are second and third behind the Rangers’ Kyle Gibson for the AL lead in ERA. Hendriks leads the AL with 21 saves while striking out 38.7% of batters and walking only 2.9%.
As for WPA, Lynn is the Sox’ leader and tied for fourth among AL starters at 1.88, with Rodon at 1.38 and relief ace Hendriks at .76. Compared with average results, Lynn has added 1.88 wins to the Sox’ probability.
Once a leverage index is included, Rodon moves up to 1.88 — fourth among AL starters — with Lynn at 1.69.
Let’s use Lynn’s latest start as an example. He went six innings, allowing one run and four hits, striking out nine and walking five in the Sox’ 8-2 victory Friday against the host Tigers.
The Sox entered the bottom of the sixth with a 2-1 lead and a 59.5% probability of winning. A leadoff walk to Jeimer Candelario walk reduced that probability to 53.3% and subtracted .062 from Lynn’s WPA.
How are the probability numbers derived? History tells a home team leading off the sixth by putting the tying run on first increases its likelihood of winning by 6.2%.
Later in the inning, with runners on first and second and two outs, the Sox’ win probability was up to 60.9%. Lynn struck out Jake Rogers, protecting the 2-1 lead and boosting the Sox’ win probability to 68.5%.
For the inning, the Sox’ win probability climbed from 59.5% to 68.5%, and Lynn’s WPA rose by .09.
Not all plate appearances are created equal. A home run in a close game will do more damage to the win probability of the pitcher’s team than a homer in a blowout. So the cost in WPA will be greater in the close game than in the blowout.
Similarly, a game-ending strikeout in a close game is a bigger WPA boost than a strikeout in the early innings.
Not all pitchers get the same opportunity to pitch in the same number of highleverage situations. There are unequal opportunities to boost WPA. Dividing the WPA by a leverage index (LI) puts pitchers on more even ground.
Once that’s done, Rodon takes the Sox’ lead with his 1.88 WPA/LI.
WPA is not a strong predictor of future success. All pitchers have ups and downs in how they fare in high- and low-leverage situations. Instead, WPA and WPA/LI are snapshots of what has happened with a pitcher on the mound in any defined time period.
For the Sox, the time is now. And their All-Stars have been about as good as it gets.