Carlos Carrasco has made adjustments and is having success

I was among those not in favor of giving Carlos Carrasco another go at the rotation. The guy had seemingly settled in to the bullpen and while a good (or even average) starting pitcher is more valuable than a good middle reliever Carrasco had certainly not demonstrated that he could be a "good starting pitcher" before.

I am happy to admit that my disapproval A) was ignored by the Indians front office and B) has given way to optimism, thanks to some research I've done.

Usually I have a pretty good sense of what I'll find when I dig into a player, but in this case I expected to see that Carrasco was basically just getting luckier now than he was in April. There is a little truth to that idea (his .344 BABIP as a starter in April has given way to .161 in August and neither of these numbers are sustainable), but there is a lot more to it.

In April, he struck out 23 batters in 22 innings, compared with 9 over 12 innings in August. And while that is a drop off, in terms of K/9, it is a perfect example of why K/9 is flawed. Carrasco was striking more guys out, but he was facing way more guys too. K% is a better way to look at it, and in August, Carrasco has struck out 22.5% of the batters he faced, while in April, he struck out 23.5%, meaning he's striking out guys almost exactly as often now as he was before.

Strikeouts are only part of the equation. Data shows that one of the best predictors of ERA is K% minus BB%, basically, how many more batters you strike out than you walk. In April, Carrasco posted a 9.2% BB% in his starts giving him a K%-BB% of 14.3. In August, he has yet to walk a batter as a starter, so that 22.5% K% is also his K%-BB%.

For context, here is a complete list of MLB starters with a K%-BB% north of 22.5 this season: Clayton Kershaw, Chris Sale, David Price, Felix Hernandez, Stephen Strasburg, Masahiro Tanaka. Six pitchers, all of them among the twenty best in baseball.

Why the heightened success? I see a four causes:

1) Control: In his four April starts, Carrasco threw 53% of his pitches for strikes. In August, that number has jumped to 67.5%.

2) A change in his pitch mix: In April Carrasco threw 56.7% hard pitches (sinker and four-seamer), 28.9% breaking pitches (slider and curve) and 14.4% off-speed pitches (changeup). In August, he is using the hard stuff more (62.8%) and the slower stuff less (26.3% and 10.9% respectively). Within that mix, he is also using his slider (which has always been his most effective pitch) more often. He never uses it much vs. lefties (3% in April, 0% in August), but he's nearly doubled his usage (from 23% to 45%) against right-handed opponents.

3) Velocity: In April, that four-seamer was sitting just under 95 mph. It's up to over 97. The sinker was at just over 93. Now, it is closer to 96. The same holds true for his other pitches.

4) A change in his release point: Carrasco is now releasing the ball a couple inches lower and a few inches closer to first base than he was in April. Before the season, there was a lot of talk about Carrasco's mechanics, and while I haven't heard or seen much about that lately, there was some noise around the time he was moved to the pen. After his April 25 start, Carrasco complained that his new mechanics - with a raised arm slot- were causing him to lose velocity. Terry Francona disagreed.

"I know he said something about when his arm's up, he was throwing 92. This is what we have to get to, because maybe that's what it feels like, but that's not the case."

Tito went on to say Carrasco was throwing harder than the pitcher thought, but that he wanted Carrasco to talk to pitching coach Mickey Callaway to get things figured out.

I wasn't able to find any quotes about his mechanics since then, but if Carrasco didn't like extending his arm, and decided to keep himself more compact through his delivery, that would show up in exactly the way we're seeing with his release point: lower to the ground and a bit closer to the body.

We are talking about 40 batters and 12 innings over two starts and there are all sorts of small sample size caveats we have to keep in mind. But there are also some really positive signs. The better results can be explained by the increased velocity and improved control. Both of those can be explained by the changed pitch mix and the new release point.

They can also be explained by random fluctuations in the data, or happening to go on a good day each time, but when Carrasco takes the mound Friday night, watch his control and his velocity (and I'll be looking up his release point postgame). If the new patterns continue, Carrasco may be Mickey Callaway's latest successful reclamation project.