Stat Notes - Aug 16
A note on taking pitches, and updates on Rollins, Ibanez, and Hamels.
Selectivity and Scoring
In the beginning of the season, when the Phillies were scoring lots of runs, the announcers made a big deal of how the hitters were working the pitch count and "making the pitcher work". In reality the Phillies were at the league average of ~3.84 pitches per plate appearance. Maybe it was the fact that Jimmy Rollins was taking more pitches that was causing this misperception (and they didn't notice that Howard, for one, was taking fewer pitches).
In any case, this topic has been a thread throughout this season. When the Phillies are hitting well, the announcers may point to their ability to work the pitcher. When they're not, someone will comment on a 10-pitch offensive inning with a tone of disgust.
I've felt the same way. Pedro Feliz and Rollins drove me crazy last year with first-pitch swinging, and Polanco, Rollins, and Victorino (not to mention Dobbs and others) are doing the same this year.
So I looked at each game, the number of pitches the Phillies saw per PA, and matched that against the runs scored and other stats for the game. I then grouped the games by pitches per PA: games with less than 3.6 P/PA, games with 3.6-3.8, and so on. The results are below:
| P/PA | G | BB% | K% | AVG | OBP | SLG | OPS | R/G | |
| <3.6 | 28 | 6.4% | 14.0% | .282 | .336 | .474 | .810 | 5.82 | |
| 3.6-3.8 | 25 | 9.6% | 17.9% | .250 | .323 | .400 | .723 | 4.52 | |
| 3.8-4.0 | 26 | 8.0% | 19.9% | .257 | .322 | .402 | .724 | 4.35 | |
| 4.0-4.2 | 26 | 10.6% | 22.2% | .261 | .342 | .390 | .732 | 4.46 | |
| 4.2+ | 12 | 9.0% | 22.9% | .232 | .308 | .372 | .680 | 3.92 | |
| Overall | 117 | 8.6% | 18.8% | .260 | .329 | .413 | .742 | 4.72 | |
This is not the clear trend I expected to see. In fact, OPS and R/G are highest when they see the fewest pitches (<3.6 per PA). OPS and R/G are worst when they see 4.2 or more pitches per PA.
The only explanation I can think of for the <3.6 stats is that some nights they are just teeing off on the pitcher. I don't have any explanation for the 4.2+ data.
Jimmy Rollins
A few things on JRoll:
- His walk rate this year equates to 92 BB per 162 games
- Since coming off the DL the 2nd time on 6/22, his line is .221/.307/.300 (.607 OPS)
- OPS this year is down to .696, compared to .719 in his "down year" last year
- However his wOBA is higher, .323 to .316, due to the better walk rate
- His BABIP this year is .252, compared to .293 through 2009. At a BABIP of .293, his OPS would be .761, but his struggles may not just be due to bad luck -- his line drive rate is also the lowest of his career at 17.6%
Raul Ibanez
- After the 18th game in his hit streak, Ibanez' OPS was over .800 for the first time this year, at .804.
- Since then, including Sunday night's game, he's hasn't been on base in 4 games (0-18 with 0 walks)
Cole Hamels
- 2.85 ERA since May 16th.
27 comments
|
0 recs |
Do you like this story?
Comments
Interesting results
I agree, the results for Runs per game are a bit backward of your initial thinking. However, it makes some sense when you think about it (hitters are seeing less pitches because they are hitting them more, hence, scoring more runs). Nice job.
pitch data
I think the K% column is the most important one there. If the team as a whole is seeing less than 3.6 pitches per at bat, that means that it’s not striking out much at all (as a strikeout needs at least 3 pitches, usually more). With 4.2+ per at bat, that means more strikeouts.
So, on the one hand, your data is counterintuitive. On the other, thinking about longer at-bats and what good they are for the pitcher, I’m not so surprised.
But, I wonder how this would play out league wide and over multiple seasons. Is this always the case?
by David S. Cohen on Aug 15, 2010 11:29 PM EDT reply actions
Oh, and I second FM here – nice post!
by David S. Cohen on Aug 15, 2010 11:30 PM EDT up reply actions
On the other hand, the OBP is highest for the 4.0-4.2 range, so you’d think they’d score more runs that way. But more pitches might also mean more offspeed/breaking pitches which could effect the quality of contact or if there is any contact at all.
Pessimism of the intellect, optimism of the will.
by FuquaManuel on Aug 15, 2010 11:37 PM EDT up reply actions
Beat me by seconds!
Remember the Phitans
by RememberthePhitans on Aug 15, 2010 11:38 PM EDT up reply actions
I wonder what would happen if you took Werth out of the equation and see what happens. Dude is just having a crazy train of a season, and he is Mr PPA.
by Wet Luzinski on Aug 16, 2010 12:37 AM EDT up reply actions
But that makes sense too, as you can’t walk if you don’t get at least 4 pitches.
by David S. Cohen on Aug 16, 2010 10:25 AM EDT up reply actions
The “teeing off” scenario is row 1. Thereafter, rows 2, 3, 4 show increasing success with more pitches faced — slugging drops a bit, but the on-base rises. On-base is worth 1.5 — 3x more than slugging per point, so losing .010 slg and gaining .019 on-base is a “win” for the hitters.
Row 5 is rough. Must be the K factor?
Remember the Phitans
by RememberthePhitans on Aug 15, 2010 11:37 PM EDT reply actions
http://www.retrosheet.org/Research/PankinM/Pankin-sabr36.pdf
on base v. slugging relative values
Remember the Phitans
by RememberthePhitans on Aug 15, 2010 11:39 PM EDT up reply actions
Thirding the “strong first post” plaudits.
Splits for the Phils on 0-0 count here. As a whole, the team has a .302/.310/.529 line (I guess there were some HBPs in there?); Werth, Ruiz and Howard have been pretty potent swinging first-pitch, while Rollins, Victorino and Exxon have been pretty weak when they go up hacking.
clarifying...
I was thinking maybe there have been games in which a bunch of guys went up looking to attack first-pitch, possibly explaining in part the <3.6 P/PA scoring.
Teams always have great lines when the count doesn’t involve a possible strikeout. After all, a strike with an 0-0 count doesn’t “count” in the stat line you just mentioned, as it just goes to 0-1. But, a strike with an 0-2 count “counts” against the stat line for that count. That’s why stat lines for non 2 strike counts are always inflated.
by David S. Cohen on Aug 16, 2010 10:27 AM EDT up reply actions
Interesting that two guys who see the most pitches (Werth and Ruiz) would be the most successful on the first pitch. Also, two first-pitch swingers are the least successful. I would guess the pitcher is throwing a more hittable pitch to Werth and Ruiz because they are more likely to let it go, but when they do swing (going counter to scouting reports), they have a good pitch to hit. And on the other hand Rollins and Victorino are chasing pitchers’ pitches.
Interesting
You went and did the legwork on something that I have been thinking about. Over and over the virtues of taking pitches and working counts gets extolled. Yet working the count does not necessarily lead to success. I wonder if some of it has to do with the pitcher faced, instead of the batters involved. With a pitcher who throws strikes, working the count is less likely to draw the walk— so it may make sense to swing earlier in the at bat (provided that the pitch is a strike). Conversely, when the pitcher is wild, it makes more sense to try to tire them out— try to draw out the at bat looking for a great pitch to drive or the walk.
SP vs RPs?
Excellent post. Perhaps greater than 4.2 PPA indicates significantly more foul balls, and therefore generally weaker contact, since the hitter has presumably taken more swings and yet has not managed to put the ball in play.
Instead of averaging each game, it would be interesting to see the direct link between number of pitches seen and the results. How do those 6+ pitch at-bats affect the data?
and then how do these results look when comparing SP to RPs, since presumably taking more pitches means the opposing starting pitcher doesnt last as long.
Yes, great post, schmenk.
Charlie has got to rest Raul. His timing has clearly been all messed up since the Kershaw game.
I wasn’t upset when he failed to rest him vs. Misch, but I will be if he repeats the same mistake vs. Zito tomorrow. Zito has owned Ralph for his entire career, and they faced each other plenty in the AL West. Unfortunately, Charlie has a tendency to run his position players into the ground. He only rests them when it has a vague air of being some sort of punishment. (Sort of like a less intense, less offensive version of what Bowa used to do to his pitchers.)
Great Post Agreed
I am not a huge ‘stats’ follower because each game is so dynamic. But over long periods I have to agree with the correlating trends.
I happy the “just take pitches” strategy (often supported by stats) is slightly disproven. I agree walks are valuable but hits are better. Hits allow for additional bases. My favorite is the man on second, single, advance to second on the throw, merry-go-round. In that specific (and somewhat rare) case each single produces a run and puts a player in scoring position. A walk in that situation increases the chance for a double play but certainly allows for more runs to be scored (more base runners.).
My opinion is that when team batting is ‘hot’ they continue to swing and get good production. However, I would expect ‘wild’ pitchers to giveup plenty of runs as well. Not sure if this is true, but I like to see the team take more pitches against better pitchers in an effort to get them out of the game sooner. Relates the q’ about SP and RP.
I can see how this data can look like an argument for swinging away, but for the reasons stated by others, I don’t think it’s accurate to interpret it this way.
Also, I don’t think anyone would argue that a walk is really as good as a hit. With the bases empty it’s close, but even then a hit has the advantage of allowing the possibility of additional bases on errors. But a walk is (essentially) always better than making an out, and I think there is a benefit to making the pitcher work, and getting to the bullpen earlier rather than later.

by 























