What can be expected when a pitcher is ahead or behind in the count.

Depends on the pitcher, the opponent, the situation, etc. Tom Tango covered this pretty extensively in "The Book: Playing the Percentages in Baseball".

J H,

 

I know its dependent on many things, not just the pitcher, and I know what the “percentages” say, but I’m not looking at it only as its better/worse to be in situation “X”. Remember, I don’t look at MLB data, and I’m looking at it in the sense of trying to identify which pitchers on say a HS team might be needing some work, or perhaps even identifying if/when it might be time to change pitchers.

 

Things would be really neat and clean if comparing ML data to any other level were a 1:1 proposition, but it isn’t.

I think it's an interesting project that could be really beneficial but I don't know how practical it'd be because of how case specific the results would be. As you said, high school baseball isn't 1:1- even amongst those playing together at that level. If you do find some sort of correlation/projection, please share because I think it's intriguing.

Here is a study by Coach Stotz of Stanford that might shed some light:

 

http://faculty.engr.utexas.edu...ers/avg_by_count.pdf

Good ole "Stotzy!".

Amazing dude, isn't he!

Thanks for sharing.   Very good article.   I enjoyed reading that.

 

 

Originally Posted by J H:

I think it's an interesting project that could be really beneficial but I don't know how practical it'd be because of how case specific the results would be. As you said, high school baseball isn't 1:1- even amongst those playing together at that level. If you do find some sort of correlation/projection, please share because I think it's intriguing.

 

Yeah, whenever one gets into uncharted territory, things get difficult very quickly and its hard to make much progress. And it is true that at any level below MLB, things are even more muddled because the quality of the players is much more varied.

 

Actually, I did find “some sort of correlation”, but I really didn’t want to share it without at least a little discussion. If it makes any sense, it seems as though the pitcher throwing pitches when he was behind in the count yielded more “bad” things than throwing pitches when ahead in the count yielded “good” things.

 

http://www.infosports.com/scor...ages/pitsncntsp1.pdf

 

The 1^st page of the report is each of the pitchers on our summer team of incoming Fr and Sophs. The 2^nd page is their opponents. The next 4 pages is broken out by game for each pitcher. The next page is our last season’s HSV team, the next their opponents, and the final pages their individual games.

 

Its very easy to see the difference between the younger kids and the older ones, just by looking at the ahead and behind percentages. Unfortunately, after that it gets really difficult to draw many conclusions because its so difficult to define good and bad performances.

 

Its really frustrating to have the numbers but not be able to take into account enough factors, or to give those factors the necessary weight to draw useful conclusions. But its fun to keep looking.

Stats,

I am wondering if you read that article by Coach Stotz and what did you think?

Thanks for sharing the piece by Coach Stotz, I had never seen that before. I did see a piece about his analytics that detailed Mark Appel's sequencing and attributes over at FanGraphs, but this is also outstanding. The statistic that stood out the most to me was the information that noted hitters have a better BA on strikes when in a 2-strike count than in a non-2 strike count. This is something I wish I would've been cognizant of during my pitching days. Coaches always said "get ahead" and "make them swing at your pitch with two strikes", etc., but contextualizing the information would've made it a whole lot easier for me.

 

Stats- I'm pretty confused when looking at your database, to be honest. From what I gathered, you state that hitters are more successful in hitters counts than pitchers are in pitcher's counts. If that's true, how do you measure "success" in this sense? Also, how do you define a "pitcher's count" and a "hitter's count"? For example, a 2-2 count is not even. It's a pitcher's count. Just because the numbers are the same does not mean that the batter is in favor...the pitcher has more leverage with true outcomes than the hitter does.

 

Originally Posted by TPM:

Stats,

I am wondering if you read that article by Coach Stotz and what did you think?

 

Actually, that article was written by Dr. Bickel. Coach Stotz is good and a very knowledgeable guy, but most of the nuts and bolts work was part of a project working toward a doctorate. I ran into Dr. Bickel way back when he was still at Stanford and still involved in Competitive Edge with the Chartmine program. He’s the guy who got me more interested in looking at HS data because the things I hear about HS data are many of the same things he had to fight about with people over college data. They were lucky enough to have the many many years of Stanford data to look at and use, and that gave them lots of credibility. I may be wrong, but my guess is that even now there are still organizations using ChartMine, even though the company’s been defunct a long time. It was so far ahead of its time, its unbelievable.

 

My 1^st involvement with Dr. Bickel was over that exact article. It was written I think in 2002 in a time when he was literally the only guy who took into account that there were reasons why good pitchers’ counts led to lower BA’s. I’d come to much the same conclusions but only in a haphazard way, and been pummeled on various baseball boards for saying literally the same thing. I read that article and couldn’t believe it. I managed to contact him and have enjoyed a being able to communicate with someone of his stature, and occasionally run some of my “stuff” by him.

 

Unfortunately I have yet able to “hook up” personally with coach Stotz, but we have played a little phone and e-mail tag over the years. So the answer to you question is, I really liked the article, and am privileged to have had some of the things in it explained in more detail.

 

So what did you think of it?

J H,

 

You’re not looking at my database. You’re looking at just a very few data points presented in only one of a myriad of ways.

 

If it came across as me saying anything about hitters, that’s not what I intended at all. I can look at this stuff from a hitter’s perspective, but right now I’m sticking strictly with the pitcher’s perspective.

 

But just to be sure we’re speaking the same language, a pitcher’s count to me is 0-1, 0-2, 1-2. Actually pitcher’s counts should include even counts, 0-0, 1-1, and 2-2 as well, but that would get way too difficult to explain to most people who look at my stuff. The problem is, almost everyone has been indoctrinated to think of one of those 2 counts as even.

 

To make it easier to understand, when I talk about it in much depth, I’ll generally define a pitcher’s count as one where a K is close than a walk if the batter doesn’t swing. So in an 0-0 count a K is only 3 pitches away while a walk is 4.

 

I could easily redo those particular metrics to make the even counts 1-0, 2-1, and 3-2, but I’m telling you it would be confusing to most people. When I generate metrics, I need it to be able to be understood by those who frankly are not the sharpest tools in the toolbox, so I’m trying to see if what most people believe should be happening can be shown, and so far it can’t.