Figure 1: The movement of Mike Pelfrey's pitches this year. The square box represents the strike zone. '

Mike Pelfrey has been a confusing pitcher to watch this year. At the all star break, he looked terrific and many Met Fans felt he was snubbed for a spot on the All-Star team. Then he hit a rough patch (well, he hit the rough patch a few starts before the all-star break, but you get the point). Then he seemed to get a few good starts here and there. But all in all, he has not panned out as we thought he would after a great April and May. Pelfrey's xFIP (a measure of the actual skill-level of his performance) for each month has been in order: 4.29, 3.85, 4.15, 5.81, 4.67, and 5.12. You should note that the first 3 of those xFIPs are pretty good, while the last three are mediocre to just plain bad.

Making matters more interesting, Pelfrey's pitch repertoire has changed this year: He added a split-finger fastball in a move that seemed to provide dramatic success in the early season. He also has maintained his slider and curveball from last season and has two fastballs (a four-seam and two-seam fastball) to go along with this new split-finger fastball. Which of these pitches is responsible for his erratic performance this year? Is his poor performance from July onwards the result of him doing something different? Or simply pitches not being as effective as they were earlier?

After the jump, we'll take a look.

Figure 2: Graph of the movement of Pelfrey's pitches in 2009 and 2010.

To Read:
Vertical Spin Deflection/Movement: the amount of inches the ball drops/"rises" as compared to how we would expect gravity to make a pitch drop. So a fastball with positive 10 vertical movement "RISES" 10 inches more than it should if gravity was the only force acting on it and a curve-ball with -10 vertical movement drops 10 inches more than a pitch thrown that is just acted on by gravity.

Horizontal Spin Deflection/Movement: The graph is from the view of a catcher or umpire behind home plate. So a pitch that's on the left side of the graph (and has "negative horizontal movement") moves in on righties and away from lefties. A pitch that's on the right side of the graph moves in on lefties and away from righties.
-----------------------------

As you can see, Pelfrey seems to have five pitches both in 2009 and 2010. There is a clear different in the movement of some of those pitches this year compared to last year, particularly in his fastball and his changeup/split-finger-fastball. I'll discuss this below as I talk about each individual pitch.

Below is Table 1, the average movement and speed of each of Pelfrey's pitches in 2010.

Table 1: The average movement and speed on each of Pelfrey's pitches in 2010.

*It's VERY hard to separate Pelfrey's two fastballs for Pitchf/x analysis. You can see this in the 2010 graph above. So take the differences between these two fastballs lightly.

** It's very possible that Pelfrey hasn't stopped throwing the changeup this year on occasion and that those pitches are being counted with the split-finger fastball. If he is still throwing the changeup, he's not doing so enough as to be noticeable.

I'm now going to into detail about each of these pitches and explain why they have resulted in an inconsistent Pelfrey, who seems to have gotten worse over time.

The Split-Finger Fastball

At the beginning of the year, Pelfrey's split-finger fastball was considered the reason for his success. It was talked about as that final complement to Pelfrey's fastball that would lift him to his true potential. The pitch seems to have replaced Pelfrey's changeup from last year and he uses the pitch far more than he ever used the changeup. Unlike his use of the changeup last year, he DOES throw the pitch against BOTH right and left-handed batters, whereas his changeup was formerly used only against left-handed batters.

If you look at Figure 2, you'll notice that the biggest difference between the graphs in 2009 and 2010 was the addition of a big green cluster below the fastballs in 2010. This is the split-finger fastball. In 2009, the green dots are barely visible at all as they are hidden underneath the clusters belonging to Pelfrey's fastball. What this means is that while in 2009, Pelfrey's changeup had the same movement as the fastball, in 2010 the split-finger (which replaced the changeup) has additional "sink" so that it seems to drop more than Pelfrey's 2009 changeup or his two fastballs.

Figure 3 shows how Pelfrey's split-finger fastball gets more drop than his changeup did:

Figure 3: The vertical path of Pelfrey's changeup in 2009 and his split-finger fastball in 2010, as seen from the point of view of someone looking from the side.

The split-finger fastball in fact drops an additional three inches more than the changeup does from release until it reaches home plate. Meanwhile, the pitch is thrown at the same basic speed (84.5MPH) as the changeup (83.3 MPH). But this additional drop could be significant.

Before we look upon the results of Pelfrey's splitter this year, lets look at where Pelfrey aims the pitch. As Figure 4 below shows, Pelfrey aims the splitter low and away from hitters, regardless of whether they're left or right handed batters:

Figure 4: The location of Pelfrey's split-finger fastballs in the strike zone this year against left and right handed batters. The graph is shown from a catcher's point of view and the square box represents the strike zone.
NOTE FOR READING: The more negative (more-left on the graph) a dot is on the x-axis, the farther in the pitch is on right-handed batters and away from left-handed batters. The more positive (more-right on the graph) a dot is on the x-axis, the farther in the pitch is on left-handed batters and away from right-handed batters.
-----

Okay, so now that we've talked about the new movement of the pitch and the way Pelfrey uses it, lets look upon the results:

Table 2: The Results of Pelfrey's Split-Finger Fastballs this year. Explanations of each of the columns are as follows:
GB %: % of balls hit into play by batters that result in ground balls.
Whiff Rate: (# of Swinging Strikes)/(# of Pitches Swung at by Batters)
Swing Rate: (# of Pitches Swung at by Batters)/(Total Pitches Thrown)
Swinging Strike %: (# of Swinging Strikes)/(Total Pitches Thrown)
% In-Strike-Zone: % of Pitches in a wide (2 feet wide) strike zone.

These numbers aren't particularly encouraging. The swinging strike rate is quite a bit below average for a split-finger fastball; this is because when batters do swing at the pitch, they're whiffing at a rather low rate of roughly 20% (compared to the average for this type of pitch of roughly 30%). Moreover, Pelfrey gets this pitch in the strike zone at a higher rate than the average pitcher, to a near-fastball-like rate of 59.49% against right-handed batters. This may not be a positive thing, as most change-ups/split-fingers fastballs are thrown more frequently out of the strike zone. On a positive note, the groundball % of the split-finger fastball is fairly good.

(It's odd to note that against the performance of this pitch is worse more or less across the board against Left-Handed Batters, though having a lower strike zone rate may not be a total negative thing.)

But these results do not tell the full story. Let's look at Pelfrey's results with the pitch per month:

Table 3: The results of Pelfrey's split-finger fastball during each month.

This table is more disconcerting:
You can see how in April, the split-finger was a very effective pitch especially against right-handed batters. This was mainly caused by an absurd swing rate of 73.68%. If I had to guess, this was caused by batters facing the pitch without having seen it in their scouting reports of Mike Pelfrey. Left-handed batters weren't taken as much off guard, because they'd faced the change-up in previous years, and thus were expecting a similar pitch. But right-handed batters in 2009 had never seen such a pitch, and thus were unprepared. You'll note that the swing rate on the pitch dropped to its most frequent rate of 50% by May. Moreover, in April, the pitch was extremely effective at getting ground balls. All-in-all it was an excellent pitch, particularly against right-handed batters.

But in every month since, the pitch has gotten worse and worse results. The pitch did have a resurgence in June, but it's slightly misleading: Pelfrey earned most of those results in 17 innings pitched in back to back starts against the Padres. This appears to be an outlier. Meanwhile, even then in June, the GB% of the pitch was way down from its April and May results. In July, right-handed batters never whiffed on the pitch once.

Meanwhile, as the pitch's result declined, so has Pelfrey's use of the pitch, dropping to an all time low of 5.35% usage against right-handed batters in August.

NOTE: Aside from the aforementioned decrease in use of the pitch, there really doesn't seem to be anything significantly different in how he's thrown the pitch in the later months to cause these results.

NOW ITS IMPORTANT TO REMEMBER ONE THING: Just because the pitch is no longer great on its own, it doesn't necessarily make it a "bad pitch." It's very possible that the presence of the split-fingered fastball in his repertoire made the rest of his pitches, particularly his normal fastballs, more effective. So we should look at his other pitches before we reach conclusions.

-In Part 2: We'll look at Pelfrey's other pitches in depth to see if the split-finger's loss of effectiveness has possibly been counterbalanced by an improvement in them.