Batter progression/regression

[scorekeeper]

I got a request from one f the parents to see if I could show how players were progressing/regressing as far as BA went, and to see if there were a way to see if they were getting to see more or less pitches. I took a shot and came up with the report as shown below.

Its pretty simple and shows how BA, pitches per PA and OBP has changed from when these players first started to get data in the database to the present, by games as they’re played. I’m not quite sure what it proves, but it does show the direction a player seems to be going.

I jus thought I’d show it in case someone might think of a way to make it useful.

[Jake Patterson]

I got a request from one f the parents to see if I could show how players were progressing/regressing as far as BA went, and to see if there were a way to see if they were getting to see more or less pitches. I took a shot and came up with the report as shown below.

Its pretty simple and shows how BA, pitches per PA and OBP has changed from when these players first started to get data in the database to the present, by games as they’re played. I’m not quite sure what it proves, but it does show the direction a player seems to be going.

I jus thought I’d show it in case someone might think of a way to make it useful.Hard to see anything I wonder if there is a way to graph it.

[scorekeeper]

Hard to see anything I wonder if there is a way to graph it.

LOL!

Definitely a thought that came to mind! I’ll see what I can do.

[scorekeeper]

This is as close as I can get without a lot more trouble.

[FiveFrameSwing]

This is as close as I can get without a lot more trouble.

Good, but it can be misleading to view a graph with each data point consisting of a batting average for one particular game.

Can you plat a 10-game moving average? That might be more indicative of how the individual is progressing as a hitter.

[scorekeeper]

Good, but it can be misleading to view a graph with each data point consisting of a batting average for one particular game.

Sorry. I should have explained it better. The datapoints aren’t for the individual games. What I did was compute the BA, PPPA, and OBP for everything up to and including that date. So, it really is a progression that shows what’s been going on up to that particular date.

Can you plat a 10-game moving average? That might be more indicative of how the individual is progressing as a hitter.

I could do that, but the truth is the numbers are for 4 seasons, fall 2006-10 games, spring 2007-30 games, summer 2007-25 games and fall 2007-probably 23 games total. I can easily run the numbers for each season and that would get close to what your asking for, but like I said, the numbers you see on the report and the graph are “moving averages”.

As much as I love the numbers, I really have mixed emotions about trying to come to “drop dead” conclusions when many of the numbers aren’t from the regular HS season. That’s because I know the games are played differently depending on not just the season, but the time of the season too.

FI, in the regular spring season, the early games are often a conglomeration of the coach playing different combinations of players trying to find that magic formula. Then, once the league games start, there’s no more playing around until the Easter break, and some non-regular players will get the chance to play. Then things get serious again until the playoffs are over.

I know I’ve really quit trying to say whether any particular season or time during that season is more “important” than any other when it comes to judging player performances. It comes down to, the pitchers are throwing against hitter, and hitter are batting against pitchers. Whether they’re top quality V, average V, or less than V players is really immaterial.

The only thing I worry about is making sure the players are all playing against pretty much the same teams, and so far they are. It’s a pretty complicated thing, and definitely nit something anyone can get a great deal out of with just a quick glance. ;)

[scorekeeper]

I’ve done a bit of thinking and made a couple of refinements on the batting progression/regression thing. I changed the original report to make it a bit easier to se an up or down tick, and I also made another report to attempt to show what happens when the number of pitches a batter sees is above or below what he “normally sees.

By looking at the totals for everyone, in broad terms is looks as though the more pitches one sees, the better the BA and OBP, but it still seems to be very much individual if one looks at the numbers for individual players.

[glovemedic]

I just thought I’d show it in case someone might think of a way to make it useful.

Why don't you try to run the data as a regression equation where BA is the dependent variable and # of pitches is the independent variable. You might also substitute OBP as a dependent variable to see if there is a linear trend as well.

[scorekeeper]

Why don't you try to run the data as a regression equation where BA is the dependent variable and # of pitches is the independent variable. You might also substitute OBP as a dependent variable to see if there is a linear trend as well.

LOL!!!!

You’re talkin’ to an old fart who took his last math class 35 years ago, and wouldn’t know a regression equation from a Mack truck if they both ran me down!

But I’ll give it a try if you can explain it. ;)

[glovemedic]

You’re talkin’ to an old fart who took his last math class 35 years ago....But I’ll give it a try if you can explain it. ;)

Its not that hard anymore and I think Excell will run it for you in the stat functions. However, I read your previous post more closely and saw that the BA is not a descrete datapoint, but is changing over the course of the season. If you are willing, let's try this set up a column for number of pitches in a batter saw in a single game and another column for the BA in that particular game (.000, .300, .500, 1.00). The idea (or hypothesis being tested) is that there is a functional relationship between game BA and the total number of pitches the batter saw in the game so that for each increase in magnitude of BA there is a corresponding increase in pitch count. Intuitively I can see how deeper counts may favor the batter drawing a walk or getting a pitch he can handle, but there may be other circumstances where 1 and 2 pitch counts also result in high BA too. That's the cool thing about the scientific method, you test a hypothesis to see if what you think is true has validity. I am especially intriqued by the relationship of pitchers counts (0-2) and hitters counts (3-0, 3-1, etc.) to BA but I am guessing that data is hard to collect and more difficult to analyse. Maybe a project for next year. Right now I feel like I do more than I really want to by just keeping up with the basic stats without the complex analysis. Send me a spread sheet with the data and I will see what I can do.

Have fun and play hard!

[CoachB25]

OK so I'm not the brightest bulb in the package. Wouldn't this also drastically be affected by the quality of pitching?

I know one child in particular that tore it up this fall. However, upon playing a very highly competitive level tournament where the pitching was drastically above league play, her BA took a hit. Why? Well to be honest, please meet the outstanding change! Just a thought!

[scorekeeper]

Its not that hard anymore and I think Excell will run it for you in the stat functions. However, I read your previous post more closely and saw that the BA is not a descrete datapoint, but is changing over the course of the season. If you are willing, let's try this set up a column for number of pitches in a batter saw in a single game and another column for the BA in that particular game (.000, .300, .500, 1.00). The idea (or hypothesis being tested) is that there is a functional relationship between game BA and the total number of pitches the batter saw in the game so that for each increase in magnitude of BA there is a corresponding increase in pitch count.

Intuitively I can see how deeper counts may favor the batter drawing a walk or getting a pitch he can handle, but there may be other circumstances where 1 and 2 pitch counts also result in high BA too. That's the cool thing about the scientific method, you test a hypothesis to see if what you think is true has validity. I am especially intriqued by the relationship of pitchers counts (0-2) and hitters counts (3-0, 3-1, etc.) to BA but I am guessing that data is hard to collect and more difficult to analyse. Maybe a project for next year. Right now I feel like I do more than I really want to by just keeping up with the basic stats without the complex analysis. Send me a spread sheet with the data and I will see what I can do.

OK, I’ve got the table in a SS and sent it to ya. Have at it.

I can see more than 1 reason to see more pitches, but I can also see negative possibilities, and there’s always that murky question about how much of what’s intuitive is really true. ;)

The deeper counts do favor the likelihood of OBP going up, if for no other reason than it takes at least 4 pitches to draw a walk. Its also more likely that a batter will get hit, the more pitches he sees. But, it also increases the chances of striking out too because it takes at least 3 pitches for that.

I have that data to a limited extent, and I can generally come up with some pretty neat ways to look at it. You can go to http://www.infosports.com/scorekeeper/images/cpitching.pdf and pages 85&86 to see the gross numbers for our team and our opponents.

Also, pages 93&94 will tell you about the pitchers and what happens on their 3 ball counts. Also, 101-102 and 106-109 will give you a glimpse of when hits happen. 117-120 and 121-126 will give you an idea how pitchers do with batters on as opposed to from the full windup. Pages 73-74 and 75-77 will let you see some things about 0-2 counts. The thing about all the numbers is, as one looks at them, one realizes that things are a lot more complex than hitting in a certain count. :think:

[scorekeeper]

OK so I'm not the brightest bulb in the package. Wouldn't this also drastically be affected by the quality of pitching?

I know one child in particular that tore it up this fall. However, upon playing a very highly competitive level tournament where the pitching was drastically above league play, her BA took a hit. Why? Well to be honest, please meet the outstanding change! Just a thought!

It has nothing to do with how smart or dumb someone is. Of course the quality of the pitching CAN drastically affect the “success” of the hitters, and that the tendency will be for hitters to not do as well against higher quality pitching.

But, even though it’s a given that higher quality pitching will have more success, even the best hitters will not be able to maintain their high success rate. The good hitters will likely still have more success than the poor hitters, but both groups will suffer.

I’ve sure had my share of discussions on this topic, but one thing always comes out. when one is talking about AVERAGES, i.e BA or OBP, you can put the data into all kinds of categories, but in the end, its still an average. No one only bats against the “best” pitchers, and nothing guarantees the “best” pitchers will have they’re “A” game every game.

In the long run, the quality of the hitters and pitchers has to be compared to all of the players in the league, level, or whatever. FI, there are about 360 pitchers and 260 regular players in the ML. In theory they’re the best pitchers and hitters in the world, but when only compared to the ML pitchers and hitters, someone is the best, someone is the worst, and everyone else is in between. That’s why most stats are computed for an average and not against the “best” or “worst”

And how accurate are the numbers when some moron broadcaster says Joe Hitter bats .315 against the “best pitchers? Who the heck are the “best” pitchers? How is that defined? Does it change throughout the season? There’s a lot of things that need to be considered! ;)