Friday, September 16, 2011
Rays are getting big games from their MVP
By Mark Simon
If the Tampa Bay Rays complete an improbable comeback to make the 2011 postseason, who is their most legitimate MVP candidate?
The best case could be Friday's starting pitcher, James Shields, who notched his career-high 15th win Sunday and will face Josh Beckett in the second game of this four-game series.
The Rays are 20-10 when Shields starts this season; he’s won four straight starts, allowing three earned runs in 34 ⅓ innings pitched.
This season, Shields has lived up to his “Big Game” nickname in these regards.
He is 8-4 with a 1.70 ERA against teams with .500 records or better this season (7-6 with a 3.68 ERA against other teams). That includes two wins apiece against the Red Sox and Rangers.
He has four wins and a 2.17 ERA in 11 starts in which he’s received two runs or fewer of support. That’s tied for most such wins in the majors with Cole Hamels.
In Shields’ 15 wins, he’s allowed a total of 11 runs (10 earned) in 126 ⅔ innings pitched. His 0.71 ERA in his wins is the best in the American League.
It’s also reminiscent of something similar that a former teammate, Matt Garza, did in Tampa Bay’s AL championship season of 2008, as noted in the chart on the right.
Shields ranks fourth among AL pitchers in the next-level metric Win Probability Added, which looks at how a hitter/pitcher contributes to his team’s chance to win on a play-by-play basis. His 3.96 wins added trails Jered Weaver (4.74), Justin Verlander (4.63), and Beckett (4.05), the pitcher he’ll face Friday.
The strength of that comes from Shields’ performance in close games. A pitcher gains win probability points based on his effort in such contests. Whether a pitcher loses 8-1 or 10-1 doesn’t have much impact, because the scores were not close.
Shields’ success is also a testament to the predictive ability of the metric xFIP (expected Fielding Independent Pitching), a stat resembling ERA, based on a pitcher’s strikeout and walk totals, factoring in that his fly-ball-to-home-run rate will match the league average.
Last season, Shields had a 5.18 ERA, but a 3.55 xFIP (based largely on his 187 strikeouts and 51 walks).
The Rays did not do a good job converting his batted balls into outs (his opponents' BABIP -- Batting Average on Balls in Play -- was .341).
This season, he has 210 strikeouts, 57 walks, and his opponents BABIP is only .261. It’s made for one of the biggest season-to-season turnarounds in recent memory.
Shields does have one turnaround left though. He’s 1-8 in his career in Fenway Park. The Elias Sports Bureau tells us that only two pitchers have lost at least nine of their first 10 decisions in that ballpark -- Sid Hudson (1952) and Paul Foytack (1960). Shields will try not to join that group.