Rick Porcello has not been living up to the four-year extension he received in April of this year. He has been charged with a loss in each of his last six starts. Ian York dives into his pitches and finds that location is the key for Rick Porcello.
Rick Porcello joined the Red Sox for the 2015 season and was promptly given an extension through the 2019 season. At 26 years old, he is young for a pitcher in his 7th full year in the majors. Although over his career he has been a league-average pitcher (ERA+ of 95), he has shown a promising trend toward improvement over the past few years. At best, the Sox hoped he would blossom into an ace; more realistically, they probably hoped that he would provide 200 innings of somewhat better than league-average pitching for five seasons.
Those hopes, so far, have not come true. About a third of the way through the 2015 season, Porcello has an ERA+ of just 71. Why is Porcello so much worse this year than last?
Earlier this season, we asked a similar question about Clay Buchholz, and concluded that Buchholz had probably simply been unlucky, and was likely to have much better results in his next games. In the seven games he has pitched since then, Buchholz has had an impressive 2.42 ERA, suggesting that our analysis was probably right. Has Porcello also been unlucky?
Unfortunately, the evidence does not point that way for Porcello. Although his FIP (4.45) suggests he has pitched slightly better than his 5.61 ERA indicates, that is still not much more than mediocre. Just about any way you look at his splits – righty/lefty, home/away, by month, with men on or off, battery mate – they are all bad. The only real sign of some bad luck is his home run per flyball rate of 13% (league average for starters is 11.3%), and even discounting that leaves Porcello as a significantly worse than average pitcher.
Using PITCHf/x, we looked at his pitches this season, comparing them to last year when he was much more successful. Although there is no smoking gun explaining his poor results so far, some of the explanations that have been put forward – in particular, the complaints about his pitch usage – seem unlikely to account for them. It does seem possible that he has lost some control over his four- and two-seam fastballs.
“The Red Sox think they have figured out the why of the results, too. The reason Porcello isn’t getting ground balls as much anymore is rather straightforward: He’s not throwing his two-seam sinker enough.”
At first glance, this explanation makes sense. Porcello’s reputation is as a ground-ball pitcher, one who throws a high proportion of sinkers (or, as PITCHf/x calls them, two-seam fastballs) to induce many easy ground-ball outs. Looking at the year-by-year chart below, we can see that this season, Porcello has thrown fewer two-seam (“FT”) and more four-seam (“FF”) fastballs than his career average.
But when we look more closely, this explanation falls apart. For one thing, Porcello is not throwing many fewer two-seam fastballs this year compared to last; the difference (30.0% vs 28.1%) is insignificant. For another, Porcello’s success is less ground-ball dependent than conventional wisdom suggests: both this year, and over his career, there is little, if any, correlation between his ground ball percentage and his outcomes (measured as WHIP or earned runs per inning).
More importantly, Porcello’s successful outings this year have not required his two-seam. Here are Porcello’s starts so far this year:
His most effective starts – 4-29-15 and 5-5-15 – are those in which he threw the highest percentage of four-seam fastballs and the fewest two-seamers. By contrast, in his last three starts – all disappointingly ineffective – he threw a much higher percentage of two-seam fastballs. In other words, Porcello has done exactly what many of the armchair managers have demanded by reverting to his career two-seam usage, and the results have been ugly; whereas when he did the opposite and used more four-seamers, the results were much better.
The same goes for his change and curve. There is no clear correlation between the frequency with which Porcello uses them, and the outcome of the game.
The one obvious connection, even though it was only a single game, is that in the worst game of Porcello’s season (4-19-15 against Baltimore, when he gave up 8 runs in 5 innings), he threw neither his four- nor his two-seam fastball extensively, instead relying heavily on the slider that is usually a secondary pitch for him. In a way, that game is actually slightly reassuring; even the best pitchers have days when some pitches do not work for them and they have to rely on their less-preferred repertoire. Since Porcello only used that mix on one day, it may have just been the day when he was unhappy with both of his primary pitches.
If Porcello’s pitch mix is not the problem, could the quality of some of his pitches have changed? Since we are not the major-league batters facing his pitches, we can only look at this indirectly, by comparing PITCHf/x data on this year’s compared to last year’s pitches. The three major components of a pitch are speed, break length, and break angle; 2014 is on the top, 2015 on the bottom:
There is little evidence for any significant decline in Porcello’s pitch quality, based on these measurements. His velocity and movement are essentially unchanged from his successful 2014 season to his 2015. The one change that we can see is that his curve is being thrown slightly slower and with slightly more break on it. This gives much more separation between his curve and his other pitches, and conceivably could either make the pitch harder to throw for strikes, or make it easier for batters to identify and either lay off it, or hit it. To make a long story short, though, the numbers on Porcello’s curve do not indicate that it is either much more hittable, or harder to throw for strikes, this year compared to last. Of course, pitches are not thrown in isolation; a more easily recognized curve might make all his other pitches more hittable.
If pitch mix and pitch quality are not the explanations, what is left? We can look at the locations to which Porcello’s pitches are thrown, and here we do see a difference, if only a small one. Here are the locations of Porcello’s 2014 four-seam fastballs (top) vs 2015 (bottom). These are from the umpire’s viewpoint, and the de facto 2014 strike zone is overlaid on the pitch distribution:
First, we see that even in 2014, Porcello didn’t follow the stereotype of a groundball pitcher who pounds the bottom of the strike zone. His four-seam fastball was mainly thrown in the middle to bottom third of the zone. In 2015, though, both to left- and right-handed batters, his four-seam has been coming in to the middle to upper third of the strike zone.
His two-seam fastball is more similar between the years:
The difference in the average height of his four-seam is small, a matter of 2.5 inches, but baseball is a game of inches:
Bringing his fastball into a batter’s wheelhouse, as well as making a more obvious distinction between the sinker and the fastball, could contribute to his reduced success this year.
While this seems to be a potential explanation for Porcello’s difficulties in 2015, looking at fastball location on a game by game basis does make this a little less convincing. For example, looking at his 6-15-15 game against Atlanta, in which he gave up 4 earned runs in 6 innings, it is clear that he was deliberately, and quite successfully, locating his fastball at the corners of the strike zone:
However, the broad spread of Porcello’s fastball over the season as a whole does hint that he is either changing his general approach, or having difficulty with fastball location from game to game. This fits with John Farrell’s comment, as quoted by Timothy Britton:
“The damage is coming on mislocated four-seam fastballs, and that’s an area that’s being addressed continually with the sidework.”
The problem is not that Porcello is overusing his four-seam, or under-using his two-seam. The problem is that Porcello simply hasn’t been throwing his four-seam where it will be most effective.
If we are looking for reasons to be optimistic, there seems to be some in the consistency of Porcello’s pitch quality from last year to this. Individual games show that he can, at least on occasion, locate his pitches well. The best-case scenario is that he is the same pitcher as last year, but needs a different approach, or a catcher who better understands his strengths – as well as some good luck. The worst-case scenario, of course, is that this is what he is: a mediocre pitcher with a single decent season to his name.