First pitch key for Carmona

Fausto Carmona is off to a fast start. Through seven starts, Carmona is 5-1 with a 2.55 ERA. Last season, Carmona went 1-10 with a 5.52 ERA. What’s changed?

The short answer, surprisingly, is not that much.

After a tumultuous 2006 season, Carmona made the Indians this year originally due to Cliff Lee’s spring training injury, and did not pitch well in his first start, allowing six runs in 4 1/3 innings against Chicago.

However, since then Carmona has been amazing. In six subsequent starts, Carmona has an incredible 2.05 ERA. Something seemingly changed. Right?

There is little doubt that Carmona is a better pitcher now than he was last year. However, the numbers suggest that he really is not that much better. This is possible for two reasons: 1) last year, he was very unlucky, and 2) this year, he has been very lucky. Both years he has, essentially, been the same pitcher. However, there is one key element that Carmona has added this year: throwing more first-pitch strikes.

Last year Carmona allowed a whopping .336 batting average on balls in play – far higher than the league average of .300. This year, however, his BABIP is .233, far below league average. This year, Carmona has stranded 84% of the runners that have gotten on base. Certainly, this has to do with his inordinately low hit rate, as well as additional luck which figures to regress to the mean (last season, Carmona stranded 70% of all runners).

Last year, Carmona’s main strength was his ability to induce ground balls: 59.6% of balls in play were hit on the ground. This year, Carmona’s ground ball rate has risen to an extraordinary 62.6%. For comparison, fellow sinker-baller Jake Westbrook’s ground ball rate last season was 60.8%.

While Carmona’s strikeout rate remains dangerously low – and is much lower than it was last year – he has managed to reduce his walk rate. In fact, while Carmona is much the same pitcher as he was last year, there is one clear aspect where he has improved: throwing strikes.

Thanks to baseball-reference’s, handy pitch data, we can ascertain that this season, Carmona has delivered 58% of his first pitches for strikes, well up from last year’s total of 54%. He has thrown 63% of all pitches for strikes, rather than 62% last year. 17% of all plate appearances result in an 0-2 count (as opposed to 14% last year), and only 6% result in a 3-0 count (down from 7% last year).

For Carmona – as for most pitchers – throwing the first pitch for a strike drastically changes an at-bat. This season, in all plate appearances in which the batter has had an 0-1 count, batters have hit .161/.186/.237. In all plate appearances in which the batter has had a 1-0 count, batters have hit .352/.440/.521. Last year saw a similar trend: after 0-1 counts batters hit .231/.293.299, whereas if Carmona’s first pitch was a ball batters hit .383/.474/.625.

While Carmona is mostly the same this year as he was last year, the key difference is not just his reduced walk rates and ability to throw more strikes, but his ability to throw first pitches for strikes. While 58% first-pitch strikes is a large improvement over 54%, it still stands much room for additional improvement. For example, fellow Indian CC Sabathia’s first-pitch strike rate this season is a whopping 69%.

Think of it this way: if the first pitch Carmona throws is a strike, all batters turn into roughly equivalent of me, who never even played high school baseball, trying to get a hit against him. However, if the first pitch is a ball, all players suddenly become Tony Gwynn.

Carmona was unlucky to be so bad last season; however, the silver lining is that his poor performance as a closer caused the Indians to turn him back into a starter, where he is more valuable. This season, he has been very lucky to prevent hits on balls in play (how many great plays has Josh Barfield made with Carmona on the mound?) and to strand runners that do reach base. However, Carmona has also shown improvement in his ability to throw strikes – especially first-pitch strikes – which may be the most important thing for him, as a 539-point-difference in the OPS of hitters against him after an 0-1 count as compared to a 1-0 count can attest to.