In 2012, Edward Mujica pitched for the Miami Marlins and was a below replacement reliever until the trade deadline. He was then moved to the St. Louis Cardinals for what was a failed prospect that has since been designated for assignment and outrighted to the minors. After the trade, Mujica pitched great for the Cardinals, and has been good enough in 2013 that the Cardinals have moved him to the closer role, bypassing several extremely hard throwers.
First, let’s compare his stuff and pitch selection from 2012 and 2013 and see if we see a difference in approach:
We really see three pitches here, a slider between 80-86 MPH, a splitter between 84-90 MPH, and a fastball between 90-96 MPH.
Here is his 2013 chart:
Now we see just two pitches, the splitter at 84-89 MPH, and the fastball at 91-95 MPH. Velocity obviously isn’t the difference, but pitch selection obviously provides a big difference. He seems to have quit throwing the slider completely, and is throwing more splitters than fastballs. Why would Mujica do this?
In 2012, he threw 86 sliders (according to my count/simple classifications using the spin and velocity). He got just 4 swinging strikes with them, less than 5%. He threw it for a strike less than 57% of the time, and 22, over 25% (!) of them were put in play somehow. It wasn’t a pitch he could throw for strikes or miss bats with. So it seems that he has ditched his worst pitch and moved to a pretty strict two pitch diet. By getting rid of his worst pitch, he started having more success. It is a simple and convenient narrative, and maybe one that is a little too simple. The fact is that Mujica kept throwing his slider after he was traded, throwing it 22 times with the Cardinals after the trade. So this doesn’t seem to be the only reason he is having more success, unless the success he is having so far in 2013 is different or for different reasons than the success he had at the end of the year last year.
Let’s look at the location of all his pitches and see if it helps us.
Overall, he is not a pitcher that likes to throw the ball high, and maybe one could say he is throwing a few more balls gloveside than he was last year, but these kind of maps can be deceiving. He likes to keep the ball low overall
We can rule out a release point change for Mujica, as he was 5.9 vertically and -.96 horizontally in 2012 and 5.93 vertically and -.87 horizontally so far in 2013.
So what is the key to his success in the 2013 season so far? With little doubt, it is the splitter.
29 out of 125 splitters have been whiffs so far in 2013. With the Cardinals in 2012, that is, after the trade, Mujica threw 233 pitches that seem to be splitters, and got 30 whiffs with it, a difference of over 10 percent. So Mujica’s success at the end of 2012 is a little strange, as it isn’t batted ball related, and he kept throwing his worst pitch, and his splitter wasn’t getting as many misses as he is getting with it now. First, let’s see if there is an actual difference in his splitter from 2013 to 2012. I wouldn’t be comfortable using his movement data without park adjustments for each pitch if we were comparing the differences from his time with the Marlins and the Cardinals. However, to avoid having to do that, let’s look at his movement data just at home with the Cardinals, both in 2012 and in 2013. In this way, especially since there is no real velocity difference, we may be able (assuming there hasn’t been some kind of recalibration from 2012 to 2013 in St. Louis, which I am not aware of) to see if there is a difference in his splitter, an improvement. Conveniently, he threw 100 splitters after the trade at home last year, and it had an average horizontal movement of 1.39 and -8.95 vertical movement. In 2013, he has thrown 35 at home by my count, and they have an average horizontal movement of .62 and -8.55 vertical movement. So he is actually getting less movement on his splitter so far this year according to the Pitch F/X data.
So we can’t find real Pitch F/X evidence for why Mujica was successful at the end of 2012 versus being unsuccessful at the beginning of 2012. The only real difference between Mujica as a Marlin and 2013 Mujica is the loss of the slider. The slider was so ineffective that it would be believable that this was a cause of a turnaround, but he continued to throw it in his successful run at the end of 2012. The splitter is more effective, but really no different. Unless there is something I am missing, it would seem that even though Mujica’s “peripheral numbers” are excellent, they will regress. Perhaps this is some kind of small sample size run, though we are at about 4 months of success now. But there is no reason that I can point to for why he has gotten better. Perhaps this speaks to limitations in Pitch F/X, or something else that we are missing, or maybe we just need to wait and see if Mujica can continue his success.