Stats, I don't think you were bothered and I'm not squeamish. Hope some of my input helps with what you are trying to do. If you want to use some ideas, fine, if not, that's fine as well. Like I said, I hope you make progress with what you are trying to do. I agree that we have been hearing a lot more about "high stress pitches". I think it is a noble effort you are pursuing. Hopefully, it will do some good to some kids in the future. Good luck.
Originally Posted by bballman:
…If you want to use some ideas, fine, if not, that's fine as well. …
As a matter of fact, I have been thinking about another minor adjustment. I increase the stress number for more than 15 pitches in an inning, but I’m wondering if there ought to be a decrease for less than 15. The thing is, I don’t want to give it during the inning because then everyone would get it. I’m thinking about making it an after-the-fact item as a reward.
FI, a pitcher throws a 9 pitch inning, maybe I subtract the 9 from the 15 and reduce the total for the inning by either that amount, or maybe that amount times the inning number. Heck, I don’t know if its even worth it, but it sure seems like a short pitch inning ought to be at least as valuable in a positive sense as a long pitch inning is in a negative sense.
…I agree that we have been hearing a lot more about "high stress pitches". …
I don’t want to get too far off track, but it sure seems as though the talk about it hasn’t gotten to the baseball forums yet, but rather been Sports talk and broadcast rhetoric. It’s a lot like when some metric has suddenly gained notoriety. I remember when BARISP hadn’t been heard of, and now you can’t watch a sports channel or a game without it being talked about constantly.
The talk I hear about stress innings or pitches is nothing specific, but it is getting mentioned.
The 15 pitches an inning is probably an arbitrary number to start with, but I have been hearing for many years that it is the standard for an average inning. So, sure, I think it would not be a bad idea to award points for pitches under 15. In my mind, I would definitely consider an inning under 15 pitches to be less stressful than an inning over 15 pitches.
And my guess is that announcers are using that term more as a gut feeling, based on experience rather than anything quantative. I don't think that makes it less valid, just not measured with actual numbers.
Well,
After thinking about it some, I decided that less than 15 pitches in an inning should definitely get “rewarded” somehow. Since every pitch in an inning over 15 is worth 1, I don’t see why every pitch under 15 for an inning shouldn't be worth -1.
So what I’ve done is to have the program add 1 for every pitch over 15 and subtract 1 for every pitch more than 3 and less than 15. After all, at least 3 pitches have to be thrown in order to get 3 outs. 
I ran off a few innings and from what I can tell, it looks like its working ok. It does look a bit strange because the stress numbers are so low early in an inning, but it really doesn’t make a lot of difference as long as they’re being computed the same way for everyone.
Unless someone has something else to consider, I’m gonna stop changing the algorithm for a while. Our fall ball season starts on Sept 7th, and I want to make sure I have things rung out and have a good handle on what I’m looking at before then.
Still tinkering with the pitches. Here’s last night’s game I scored.
http://www.infosports.com/scor...images/pressure7.pdf
Even though it really doesn’t make any difference what the numbers are as long as they stay relative to the factors, I have a mental block about seeing negative numbers. The Atlanta starter certainly pitched a “better” game and got more support. But to me the negative numbers make it appear that he was breezing through the game with absolutely no stress at all, and that wasn’t the case.
So, I’m making a little “tweak”. I like the idea of rewarding low pitch counts in an inning, but I’m changing giving negative points for every pitch from 3 to 15, to only pitches 10 thru 15, then I’ll take another look at it. I may well end up giving points for pitches differently, but I think this way isn’t bad, and this tweak should get it closer to being more aesthetically acceptable.
I was thinking about it earlier, but didn't want to bring it up to muddy your waters more, but since you're making a change anyway, here's what I was thinking. Who says 15 is the magic number? It's generally regarded as that, but who's to say pitch 16 is more stressful than 15? So, to give some leeway, why not make pitches 13-17 a normal inning and starting at 12 take away points and starting at 18 add points. I think it gives a little more leeway in what would be considered a normal inning.
Just a thought.
bballman,
Well its nice to know I’m not all alone in my line of thinking if not my exact thoughts.
No one says 15 is the magic number, but it is generally accepted as a good rule of thumb. I’m really squeamish about using a number as “normal” that seems to favor the wild guys. Those 2 extra pitches an inning works out to 14 extra in a 7 inning game, and that bothers me a bit.
In the back of my mind I’m thinking I’m gonna end up dropping computing the points on the fly and do it after the inning is over. The trouble with that is, it makes it very very difficult for me with the way my program works now.
Right now when I change pitchers, the pitch count for the inning resets to zero, and the math can be done from there, But in reality it should be pitches per 3 outs, where if a pitcher doesn’t get 3 outs in an inning, his pitches shouldn’t even count. Hope that’s not confusing.
What I’d really be comfortable with would be calculating pitches per batter for every pitcher, which would be pretty simple, then use that number to determine +- points. FI, if Billy has a PPB of 3.9 and throws 6 pitches to end an at bat, he should be “penalized” 2.1 points. If he ends the AB in 2 pitches, he should be credited with 1.9. That way its controlled by him, not some arbitrary number like 15.
Another option would be to use the average number of PPB for either the team or all the pitchers. Last year out pitchers averaged 3.7 PPB, where us and all of our opponents averaged 3.6. to be honest, that would be by far be a more precise number. Since I also have a pitches per inning number I can easily compute, I likely could also use that number as well.
But its not as though this has to be done tomorrow. I have until next March before it has to be etched in stone, but I’d rather have it done ASAP in order to test it more.
Thanx for the thoughts though. This stuff isn’t as easy as just throwing some numbers out there is it? I think a lot of people don’t really realize how much thought goes into something like this in order to try to make the final result a good number.
I’ve decided on how I want to try to go forward with the pitching. When the program starts, I compute the pitches per batter for all the pitchers for the season. As I’ve been thinking about it, I’m liking that more and more because while it will be a variable number that will change from game to game, its not going to change all that much, and it will be something that reflects real numbers.
Then, when the at bats over, I’ll subtract the number of pitches from that number. If the result is a negative number, it will mean the number of pitches was lower than the average and it will reduce the total stress. If it’s a positive number it will mean it was higher and will increase the stress. Once that’s done, next comes figgerin’ out how much to increase or decrease the number.
Here’s where things get a little difficult. Let’s assume the number is 3.76 and the number of pitches is 1. That’ll only take away 2.76 total points for the entire at bat. If it took 7 pitches, it will add only 4.24 points. That’s not much when you consider that if I was still using 3 pitches, it would have been a negative 50 for the same thing. Sooooo, I’m thinkin’ that once I get that number I might do something to make it more valuable.
Trying to keep it simple again, if I multiply the result by the number of pitches, in that example the total would be -2.76 + 29.68 for 26.92. It may not be the most elegant way to show value, but it does recognize both good and bad for the pitcher.
Oh well, chances are by the time I get it hooked up and workin’, someone will either come up with something that makes more sense or I’ll change my mind again! 
Now that I’ve tinkered this thing as much as I feel comfortable with, at least for the moment, I’ve come up with a report that shows some of the things I can now compute.
http://www.infosports.com/scor...images/pressure9.pdf
Its pretty simple, but if stress can be measured, and if I’m measuring it in a valid way, its pretty easy to spot the pitcher who’s throwing under the most stress, Now does that mean anything? I dunno. Should the pitchers be segregated by starters and relievers? I dunno.
But, I’ve at least given myself something concrete to look at, and as I score a few more games I’ll be able to get rid of the data from games where I was flip flopping all over the place.
Thanx to anyone who made any comments, because good or bad they all helped and I appreciate them. And if anyone would like to make further comments, I'll appreciate them as well.
