# Baseball by the book:Plate Discipline

• 06-14-2007, 11:23 AM
texasdave
Baseball by the book:Plate Discipline
One of the tenets of baseball is the belief that working the count is a good thing. A number of positives are said to come out of working the count. First, by working the count a team gets to see all the pitches of the opposing pitcher. Second, by showing good plate discipline you force the opposing starter into throwing a greater number of pitches. This leads to getting into their bullpen sooner. Third, the more pitches you make a pitcher throw the more likely he is to make a mistake and leave a pitch over the middle of the plate. Fourth, taking a lot of pitches leads to walks and walks lead to runs. These are the major benefits touted; but they all add up to one thing - scoring more runs. This is a myth. At least for the 2007 Reds it is.

I looked the number of pitches and the number of pitches by the opposing team in two categories - overall and by starter. The average number of pitches seen by a Reds' hitter in 2007 is 3.77. Using this information the starts were grouped as follows: 4.00 or more pitches is Group 1, 3.54 - 3.99 pitches is Group 2 and 3.53 or fewer pitches is Group 3. Here is the result of that grouping both overall and by starter.

Code:

```OVERALL        RUNS        GAMES        RPG GRP 1        66        18        3.67 GRP 2        144        28        5.14 GRP 3        86        20        4.30 STARTER        RUNS        GAMES        RPG GRP 1        66        18        3.67 GRP 2        157        32        4.91 GRP 3        73        16        4.56```
In both instances, overall and by starter, the fewest runs were scored in the group that had seen the most pitches per batter. Perhaps 66 games is too small a sample size. But it does seem to indicate that seeing more pitches doesn't lead to scoring more runs. If more pitches lead to less runs being scored, then why doesn't the group that sees the least pitches score the most runs? I think one reason is because the group that sees the least number of pitches is probably seeing a greater percentage of strikes, and as a consequence, is falling behind a lot.
This puts the pitcher at the advantage. This study would seem to indicate one thing - take a hack at the first good pitch you see.

Since I had all of the starters' pitches per start available for Cincinnati and their opponents, I took a quick look at Pitcher Abuse Points. Here is a chart that shows the number of starts and the number of PAPs for each. Not suprisingly, JN cracks the whip early and often in that regard.

Code:

```O-PAP        21753                CIN-PAP        102811 O-PAPG        20                CIN-PAPG 37```
Cincinnati starters have gone over the 100 pitch mark in 37 of 66 games.
(56&#37;) The opposition starters only 20 of 66 games. (30%) In these games Cincinnati has compiled roughly 5 times as many Pitcher Abuse Points. :(
• 06-14-2007, 11:29 AM
flyer85
Re: Baseball by the book:Plate Discipline
Quote:

Originally Posted by texasdave
In both instances, overall and by starter, the fewest runs were scored in the group that had seen the most pitches per batter. Perhaps 66 games is too small a sample size. But it does seem to indicate that seeing more pitches doesn't lead to scoring more runs.

the argument I have seen posited is seeing more pitches gets you into the middle relief of the opposing team. For teams with decent starters and little depth in the pen that would be a good thing. Against other teams with marginal starters and a good relief corps it would not seem to help.
• 06-14-2007, 11:56 AM
oneupper
Re: Baseball by the book:Plate Discipline
I don't see the connection between the two issues raised (Pitcher Abuse and Pitches per AB).

As for the first point. Does a higher P/PA lead to a higher RS/G?

I'd start by correlating those two variables DIRECTLY. Plot P/PA vs. RS and run a regression on it.

See if there is a correlation (positive, negative, etc.) and how good it is.

It would probably help to have a larger sample size.

If you do get a correlation that may be significant, it would help to get corroboration from another data set (i.e. another team).

This is an interesting example of how individual stats may (or may not) not project into team stats.

It is apparent that -on a team basis- that RS correlates pretty well with OBP.
On an individual (player) basis, OBP correlates pretty well with P/PA.
It would seem logical that RS (team) should correlate with P/PA (team).

Your preliminary data shows the opposite.

Interesting.
• 06-14-2007, 12:11 PM
RedsManRick
Re: Baseball by the book:Plate Discipline
We are treated summarized P/PA in very different ways. In one, the context is P/PA by pitcher in a single game against multiple batters. In the other, the context is P/PA by hitter in many games against multiple pitchers. So while they are measuring the same thing, we cannot compare the insights gained from these two very different aggregations.

I would run two regressions:

1.) Individual hitters -- P/PA and RC (or VORP or OPS)
2.) Pitchers -- P/PA and game ERA (or R/PA)
• 06-14-2007, 12:28 PM
oneupper
Re: Baseball by the book:Plate Discipline
Here's an article from the Hardball times on this issue.

http://www.hardballtimes.com/main/ar...res-the-pitch/

Their conclusion is that seeing more pitches is great, if what you are taking are balls.

Of couse, that is the very definition of Plate Discipline. (and oh..so hard to do).
• 06-14-2007, 12:51 PM
RedsManRick
Re: Baseball by the book:Plate Discipline
Interesting thought:

Pitchers who are wild actually maintain medium P/PA levels because they're leaving balls over the plate often enough to give up hits early in at bats and balance out their walks. Their high pitch count from from lots of PA, not from lots of pitchers per PA. Pitchers who have better control and miss lots of bats are the ones who have P/PA, but are successful by limiting their PA. The PA part of the equation is variable and has a very strong relationship with runs allowed.

Hitters meanwhile have a "fixed" number of PA. Thus, in any given at bat, getting more pitches can still tend to be a good thing. I love that line oneupper. I think Vlad is a great example. Getting a high P/PA isn't about getting a walk. It's about getting a good pitch to hit and walking if you don't get one. The "problem" with Adam Dunn is that he doesn't make enough contact so he often works himself in to a 2 strike count and then can't connect with the pitch he gets to hit. It's why a guy like Pujols or Sheffield sees a lot of pitches but doesn't strike out very much.

Good pitchers can go deeper in to a given AB without ever giving the hitter that pitch to hit (or make them miss the one they get), bad ones can't. So for pitchers, more pitchers per AB, but fewer ABs.
• 06-14-2007, 04:38 PM
Spitball
Re: Baseball by the book:Plate Discipline
Quote:

Originally Posted by flyer85
the argument I have seen posited is seeing more pitches gets you into the middle relief of the opposing team. For teams with decent starters and little depth in the pen that would be a good thing. Against other teams with marginal starters and a good relief corps it would not seem to help.

Outside of the last game of a series, it is a good strategy to get to an opponent's bullpen early in order to limit the the availability of arms for the next game.
• 06-14-2007, 04:46 PM
Rojo
Re: Baseball by the book:Plate Discipline
Working the count isnt' much work at all.
• 06-15-2007, 12:17 PM
texasdave
Re: Baseball by the book:Plate Discipline
Quote:

Originally Posted by RedsManRick
Interesting thought:

Pitchers who are wild actually maintain medium P/PA levels because they're leaving balls over the plate often enough to give up hits early in at bats and balance out their walks. Their high pitch count from from lots of PA, not from lots of pitchers per PA. Pitchers who have better control and miss lots of bats are the ones who have P/PA, but are successful by limiting their PA. The PA part of the equation is variable and has a very strong relationship with runs allowed.

Hitters meanwhile have a "fixed" number of PA. Thus, in any given at bat, getting more pitches can still tend to be a good thing. I love that line oneupper. I think Vlad is a great example. Getting a high P/PA isn't about getting a walk. It's about getting a good pitch to hit and walking if you don't get one. The "problem" with Adam Dunn is that he doesn't make enough contact so he often works himself in to a 2 strike count and then can't connect with the pitch he gets to hit. It's why a guy like Pujols or Sheffield sees a lot of pitches but doesn't strike out very much.

Good pitchers can go deeper in to a given AB without ever giving the hitter that pitch to hit (or make them miss the one they get), bad ones can't. So for pitchers, more pitchers per AB, but fewer ABs.

Here is a look at all NL pitchers that have made more than 7 starts. They are divided into four groups:
G1 - good stuff, good control. higher than ave k&#37;; lower than ave bb%.
G2 - good stuff, poor control. higher than ave K%; higher than ave bb%.
G3 - poor stuff, good control. lower than ave k%; lower than ave bb%.
G4 - poor stuff, poor control. lower than ave k%; higher than ave bb%.

As far as the Reds' staff goes. Harang is in G1. Belisle and Lohse are in G#3. (Belisle just missed G2 - a couple of good k starts would move him up) Arroyo is in G4.

This chart shows how the four groups with respective to both P/PA and OBP.
(Only hits and walks were included in the OBP.) HBP seemed like it wouldn't have made an appreciable difference.

Code:

```Group        K%        BB%        P/PA        OB% G1        21.7%        6.2%        3.75        0.285 G2      18.8%  9.7%    3.84        0.314 G3        12.7%        6.3%        3.55        0.312 G4        13.5%        10.0%        3.76        0.337 Average        16.2%        7.9%        3.71        0.313```
Three of the four groups have pretty similar P/PA numbers. G3 is lower. This must be the 'pitch to contact' group. If you know that you are unlikely to strike many batters out, I suppose you just let them hit it and let your fielders do the work.
No surprises with the OBP%. If you strike a good percentage of batters out and don't walk many then your OBP will be low. If you walk a lot and can't keep the ball out of play then your OBP will be high.