Privious pitch effects


#1

Didn’t know quite where to ask this question, so I stuck it here.

Does anyone have any thoughts about if whether a pitch is a strike or a ball affects what “success” the batters have on the next pitch? IOW, if we define success as simple BA, will the BA be better after a strike or a ball?


#2

This is a good stats question … Scorekeeper, what do you have?? :slight_smile:


#3

Well, since I don’t mess with ML or Mil stats, I can only comment on what I have for HS. And, since I’ve only been looking at this for a few days, I don’t really have all that much. Take a look at http://www.infosports.com/scorekeeper/images/pripit1.pdf and I’ll do my best to explain what you’re looking at.

The 1st column of course is who’s pitchers the corresponding numbers are for. This 1st report is a accumulation of all of our pitchers for the last 4 seasons, and all of our opponents.

In order to actually look at things in a way that made sense to me, since I was only interested in seeing what happened after a previous pitch in an at bat was a strike or a ball, I had to filter out certain things.

The 1st were at bats that weren’t “official at bats”, and weren’t either walks or HBP’s. That would be things like Sac Bunts, Sac Flies, or when an at bat wasn’t completed, such as when the 3rd out was made or the game ended on something like a SB or a CS.

The next Major column, “ABs ending on the 1st pitch” is something else I didn’t want contaminating the final data. The reason is, since the AB finished on 1 pitch, there couldn’t have been a ball or strike thrown before in the AB. Just to keep things consistent, I showed the ABs, hits, and the BA for that condition. Just so everyone understands, contrary to popular belief, hacking at that 1st pitch doesn’t necessarily mean it always works out well for the pitcher. :wink:

I didn’t have to because it really didn’t add a lot to the report, but I did show the K’s, BB’s’ and HBP’s after both strikes and balls.

The next Major column, “ABs after strikes” just shows what happens when BA is computed using only ABs where the pitch before the one that ended the AB was a strike, and the “ABs after balls” column s of course the one where that next to the last pitch was a ball. The final column “Total At Bats” is of course the two possibilities added together.

When I ran that report against all 4 years, the 1st thing I noticed was a .062 point increase for their batters against our pitchers from the next to the last pitch being a ball rather than a strike, an .084 point increase for our hitters, and a .081 increase together. That really caught my attention, and when I ran it against other data sets, like our JV and Fall Ball teams, my son’s HS team, and his JUCO teams, that’s pretty much the same thing that happened every time.

Of course there were exceptions, and that’s what prompted me to take the next step, which you’ll see on the next 3 pages. What I had to do to try to explain why there were situations where things reversed, was to break everything out a bit more. Our team is broken out by individual pitchers and our opponents by team.

At this point in time, to be honest, the only think I can say for sure is, throw as few balls as possible! lol! But seriously, early on in something like this, its really difficult to come to many valid conclusions, because every question that gets answered, gives rise to 2 new ones. FI, does there being runners on have a bearing, how about the pitcher’s strike percentage, or maybe being ahead or behind in the count affects it, and there’s about a gozillion more questions that come up.

When I do things like this, what I’m really doing is attempting to find reasons for what takes place on the field, hoping to give solid reasons for working on something rather than just doing it, mostly because I feel it helps me better understand the game, and because its fun too. :wink:


#4

I don’t see how you could ignore all other variables and conclude that whether the previous pitch was a ball or strike was the lone contributor to success on the next pitch. There’s just too many other variables at play at any point in a game.


#5

I would never suggest to anyone that they should only look at one or two factors in anything, let alone the outcome of a pitch. As I suggested, the process of investigation isn’t a one step thing. It often takes a great deal of time to work through the variables, and that’s what I’m in the process of doing. But, that doesn’t mean when what looks like a direct correlation is seen, it should be tossed off as nothing.

How would you explain the apparent differentiation between what happens after strikes as opposed to what happens after balls?


#6

Does it matter that four balls end an AB but only three strikes? It seems there would be more overall chances to hit after a ball.


#7

I‘ve never thought about it much, so I don’t know for sure. I suspect that if balls and strikes were 50-50, that would have a huge impact, but they’re not. Even weak pitchers are about 55-45% strikes to balls. And, foul balls enter into somehow I’m sure.

If you have some way to take a look at it, I’ll sure be glad to see if I can prove it or not.:wink:


#8

I’ve taken a step to see if I can show more of a correlation of what prior pitches were to how an AB ended. Please see http://www.infosports.com/scorekeeper/images/pripit3.pdf

Basically everything was done the same, but instead of looking back only 1 pitch, I looked back at the last 2, and then only looked at the records where those two pitches were the same.

Again, this data in no way represents ALL of baseball at every level, but rather just what I’ve been witness to for the last 4 years. However, when I looked at the data for my son’s HSV team, the numbers were very close to the same, and when I looked at the JUCO data I have, the correlation is even much stronger.

I suspect that at the levels get higher and the hitters and pitchers get better, the disparity widens between whether the previous pitches were strikes or balls, and as the levels get lower, the difference gets smaller.

I’m wondering if it might be reasonable to assume that this might be a “reasonable” way to tell if a pitcher is beginning to “lose his stuff”.

I’m also considering looking at whether or not there being runners makes any significant difference. If anyone has any thoughts about any other ways to try to look at it, please don’t be afraid to say so.