Skip to main content

Replies sorted oldest to newest

quote:
Originally posted by ofs10:
they do play by play so they know if its earned run!and then u can edit stats at home!somebody has game changer at the game to keep the game live!


I understand how it works. I’m asking how the program determines ERs? Is it completely hands off? If it is I’d really like to know how, since there are times when an ER is very subjective, and impossible to determine whether a run is earned until the third out of the inning.

I ran into this problem when I converted my stat program to a real time scoring program, and never did find a resolution that didn’t require someone to verify whether a run is earned or not in every instance. I managed to eliminate about 90% of all runs automatically, but there are definitely some that get tricky. I ran into the same thing trying to have the computer figure out unnecessary pitches.
quote:
Originally posted by Jess1:
ERs by rule are, as OFS correctly pointed out, are handled by the program. Beyond that, GC allows the scorer to manually change 5(IIRC) different stats, ER included.


I don’t know how to explain this so it can be understood, and I’m grasping here, so please don’t be offended or think I’m taking an unnecessary shot at GC.

There are times when an ER simply isn’t black or white, and the scorer has to do things like “ASSUME” what “MIGHT” have taken place if there were no errors in an inning. There are also times when there is no error, but a run might very well be unearned. There are also times under HS rules, where there’s a “TEAM” ER but not an ER charges to a pitcher.

All I’m saying is, since those things do take place, and as far as I know there’s no app or PC that has AI yet, I don’t know how the ERs GC puts out can possibly be trusted. To the best of my knowledge, I-Score is the same way.

I’m not saying its something that happens a lot, but I’ve been scoring HS games for 11 straight years now, and there’s at least a few times every season where things are cut and dried, and with I ER having such an effect because of the low number innings, ERs can really be affected.
quote:
Originally posted by Stats4Gnats:

There are times when an ER simply isn’t black or white, and the scorer has to do things like “ASSUME” what “MIGHT” have taken place if there were no errors in an inning. There are also times when there is no error, but a run might very well be unearned.


Maybe if you gave an example it would help to explain how it would be handled on GC or iscore.
quote:
Originally posted by Jess1:
I guess I'm not being clear enough - GC allows the scorer to manually change certain data fields, including ERs to ensure correct assigment (team and/or individual error, for example).


I know they do. But I’m assuming they do some form of automatic computation for ERs. If not, there’d be no reason to change anything. Wink

I’m not one for a lot of eyewash stuff like on-line play-by-play or spray charts. Those things are great for some people, but to tell the truth, I’m really only interested in creating valid stats. Now if the only people using a scoring app like that are experienced old farts like myself, then there’s a good chance there’s gonna be someone watching for those little things that need attention. Unfortunately, I’ve seen GC or IScore used in a game about a dozen times now, and every time its been a coach trying to score the game while he’s trying to coach, some player who’s about as interested in scoring the game as he is about spending an extra hour studying history, or some fan who can keep score, but really doesn’t understand the minutia of something like this.

I’m not saying the way I did it was the best way, but what I did way not allow the pitching stats to be done until every run in any inning where there was an error, a PB, or defensive interference has been checked. Some are pretty obvious, like any runner that reached on an error and scores is a UER. Those are simple. But while there’s no E on a PB, it could be the reason a run scored, and if it is, it too is UE.

As I said, I’m not at all saying GC isn’t a great app for 99.99% of the people out there, but most of them aren’t really gonna care about such things. Its just that I have other things I’m interested in, like unnecessary pitches. It’s a metric that doesn’t mean jack really, but I find it very interesting when I see one pitcher being forced to throw a lot of unnecessary pitches while another isn’t. In today’s world where pitch counts are so closely tracked, it means a lot more than it did years ago. And as for ERA, if its not going to be done correctly, it seems stupid to even bother. Wink
quote:
But I’m assuming they do some form of automatic computation for ERs


No offense, but..?

As to your example, GC would score it as UE-as would anyone w/a passing familiarity w/the game. As mentioned previously, GC allows the scorer to edit "third out" runs (E to Pitcher, not to Team). FWIW,one can also pull pitch counts.

All stats are no better than the scorkeeper, regardless of the sport. If that's your complaint, then no scorekeeping, neither automated nor paper & pencil would be suitable.
Stats4gnats - I encountered a similar problem scoring for our HS team....a questionable call where the concept of "replaying" the outcome as if the error had not occurred created a judgement call (and one that was debated amongst the coaches).

GC calculates it's own version of ER by applying as much logic as it can. I sent my scenario to GC Support and they asked me "well, what should it be?". That led me to realize there may be a weakness, although I am not convinced that it is a GC problem as much as it is that you cannot always apply "logic" to scoring. There are some interpretations where you "just had to be there".

With that said, I suggested an enhancement where they would color code the runs scored (assume green means ER and red means unearned run). They seemed to like the idea, but I have not used it since last season, so no way of knowing if it has been incorporated.

For the record, I love GC...took 90% of the work out of it for me, although there are several ways they can enhance it to get closer to 100% accuracy. Editing after the fact (other than changing the stats) should be much easier than it is. But, with that said, we plan to use it again this year since it does a wonderful job mixing the live scoring with the game recap, and the automatic calculation of stats which we than import into MaxPreps. I can't imagine doing it the old way.

So, the final answer is, I can't explain GC's logic being applied, but it does apply logic, however, if you disagree with the final determination of an ER vs unearned, the only way to fix it is to manually adjust the stats (takes about 2 min). Hope that helps....PS: I have heard GC made many enhancements over the 2011 summer, so I am anxious to see the latest and greatest.
quote:
Originally posted by Jess1:
No offense, but..?

As to your example, GC would score it as UE-as would anyone w/a passing familiarity w/the game. As mentioned previously, GC allows the scorer to edit "third out" runs (E to Pitcher, not to Team). FWIW,one can also pull pitch counts.

All stats are no better than the scorkeeper, regardless of the sport. If that's your complaint, then no scorekeeping, neither automated nor paper & pencil would be suitable.


I plead ignorance, so no offense taken. If I knew exactly how GC worked, I wouldn’t have asked.

I assumed there was some way it automatically computed ERs, but I’m trying to ascertain how accurate it is. I know it can’t be 100% accurate, and am wondering what it does to forewarn the scorer that s/he should check the runs to be sure. Although no one but me will use it, I’ve made it possible for my program to tell me about every run that occurred in an inning where there was an error, a PB, or DI, and won’t do the pitching stats until I’m ok every one of them. My guess is though, that GC does no such thing, and would happily include everything in the stats if the scorer/statistician didn’t change something on his/her own.

When you say it can pull pitch counts, are those “real” counts or “umpire” counts? IOW, let’s say the pitches were MBBCFFBFF, would the count be 3-2 or 3-6? The 1st is the umpire’s count, the 2nd is the real count. The 1st doesn’t let the viewer know the next pitch is the 10TH.

You are totally correct that no matter what’s used, the scorer is who makes the biggest difference in the stats. I used to say it was equal between the scorer and whoever entered the stats into whatever software was being used, but the load falls totally on the scorer when they use software that does both the scoring and the data entry.

I really have no “complaint” about GC. As far as I can tell, it’s a fantastic product. But for me, I want lots of validity checks in the software that will allow me to make as few mistakes as possible. Also, I want it to put out all the different metrics I’ve become used to seeing over the years, and I want to be able to add, delete, and modify those metrics at will. So far I haven’t seen that in any “off the shelf” software”, yet.

A few months back, a friend e-mailed me a SS of everything I-Score could export. Not stats, but pure data. If GC can do something similar, I’d love to see it. If someone would do that, let me know and I’ll send a PM with my e-mail address.
Stats,

GC creates a pitch by pitch, play by play. It vstill requires a good scorekeeper in order to be totally accurate. With a good scorekeeper you would be able to create a pitch by pitch account of the game. Then also have the ability to edit anything unusual. In the end it will spit out pretty much any statistiv a person could want.

We partnered with GC last year to help develop a scouting friendly version. This helps create information that is more important from a scouting perspective. ie. Notes about body type, mechanics, tendancies, etc. Pitches in our scouting app. include.... Ball, strike, location, ball in play, outcome of ball in play, foul ball, type of pitch, velocity, quality, etc. The system even produces a blog about the game.

We have had a few problems along the way and they all get addressed. The major cost involved for us was purchasing the IPads. In our case that requires over 100 Ipads.

Pitch counts and everything else is 100% accurate, if the scorekeeper is 100% accurate. The amount of information we gather is amazing. So much so, that several MLB Clubs are actually purchasing that info from us. Of course, that is because of the over all talent involved.

I only know about the GC system we use (scouting app.) We helped in the design. I'm not that familair with the normal product, alone, though I'm sure it is included in what we use. I do know that many others say it is outstanding.

I suggest you actually try GC, I think it is free. I would be interested in getting your thoughts once you tried it.
quote:
Originally posted by schwammi:
Stats4gnats - I encountered a similar problem scoring for our HS team....a questionable call where the concept of "replaying" the outcome as if the error had not occurred created a judgement call (and one that was debated amongst the coaches).

GC calculates it's own version of ER by applying as much logic as it can. I sent my scenario to GC Support and they asked me "well, what should it be?". That led me to realize there may be a weakness, although I am not convinced that it is a GC problem as much as it is that you cannot always apply "logic" to scoring. There are some interpretations where you "just had to be there".


ROFL! Great example, and not an unusual one. One of the things that’s nice about my program is, I didn’t design it to be the best or most efficient piece of software out there. Instead, I designed it to be a piece of software that does the scoring and generates stats for someone who’s been scoring games and generating stats for a very long time.

As I was converting it to do the scoring, every day I was amazed at how many things a scorer just takes for granted, that have to be programmed step by bloody step into the computer so they’ll do them over and over as flawlessly as possible.

quote:
With that said, I suggested an enhancement where they would color code the runs scored (assume green means ER and red means unearned run). They seemed to like the idea, but I have not used it since last season, so no way of knowing if it has been incorporated.


I’m not sure I understand. Care to explain it a bit more?

quote:
For the record, I love GC...took 90% of the work out of it for me, although there are several ways they can enhance it to get closer to 100% accuracy. Editing after the fact (other than changing the stats) should be much easier than it is. But, with that said, we plan to use it again this year since it does a wonderful job mixing the live scoring with the game recap, and the automatic calculation of stats which we than import into MaxPreps. I can't imagine doing it the old way.


That editing thing seems to be a common flaw. I-Score uses have the same complaint, and I’ve been constantly working on ways to make it less cumbersome myself. Trouble is, the more data is being tracked, the harder it is to just press a “go back” button. Things need to be reset exactly, then allow the play to be put back in.

Here’s a small example. Runner on 1st and batter hits a gapper to the fence. Runner scores all the way from 1st and batter ends up on 3rd. You’re sitting there pressing all the tight buttons, but when you look up, you see the runner going back to 3rd and the batter to 2nd because the ball had been caught under the fence and ruled a GR double. I don’t know about GC, but my program would be angry. Wink

I’d have to remove the run from the scoring data, change the batting data to show a double and remove the triple and change the RBI for the batter, remove the run scored from the runner, and change where the runners were for the new batter. Then the pitcher’s data would need to be changed, as would the defensive data, and time in the field data.

It could be done manually, but I don’t want to be bothered. So what I did was put in a go-back button of sorts. After each at bat. I store dang near everything imaginable, so I just go back to before the 1st pitch of the at bat, and start again. It actually works pretty well, and all I have to remember is the pitch sequence, which I can easily look up before I press the button.

I don’t get all involved in worrying about live updates so people can watch it on their smart phones. I write a play by play newsletter after every game, and that’s the limit for me. If someone wants that other stuff, they’ll have to take the whole job, and there ain’t a lot of job seekers. Wink

The MaxPreps thing is definitely a consideration though. But I’ve had my stat program kicking out data I can import into MP for 3 years now, so that isn’t an issue for me either.

quote:
So, the final answer is, I can't explain GC's logic being applied, but it does apply logic, however, if you disagree with the final determination of an ER vs unearned, the only way to fix it is to manually adjust the stats (takes about 2 min). Hope that helps....PS: I have heard GC made many enhancements over the 2011 summer, so I am anxious to see the latest and greatest.


You may not know it, but you and the others have definitely done a lot to help me understand how they’re doing it. There’s really on so many ways to do this stuff, and my guesses about how they’ve been doing it are pretty much right on. There are certain problems that just have to be handled by grinding it out. Wink
Interesting PG.

A combination of the scouting version and the standard version seems to be what would most interest me. One of the issues that faced me, was what I would do with the years of data I already had. I wasn’t going to toss it, so I needed to have a way to make the computer make the same data that I’d made by hand.

Since I’ve always done quite a bit more than the standard MaxPreps stuff, the 1st apps that came out weren’t even close. I have a friend over at MP who was a big reason they partnered with GC, and he spent a year beating on me about how wonderful and comprehensive it was. Since he’s a coach on one of our league opponents, last year I got to see it in action. It seemed fine to me, albeit on an IPhone much too cramped and small for my old eyes.

Then for one game his scorer didn’t show, and he asked me if I’d do his teams numbers as well as mine. I did, but my program generated about 5 times as much as what he needed, so I had to pare it down. That took longer than anything. But in the end, it showed me that there wasn’t going to be anything out there that suited my needs for at least a few years, and that’s why after the season ended, I took the plunge.

At one time or another over the years, I’ve had my program do everything you said your version does, other than the automatic blog, and I’d say a bit more as well. I take my newsletter a little too seriously to allow a computer to generate it. Needs that personal touch. Wink The trouble is, with only one person trying to tack all the different things by hand, the job quickly becomes impossible because of the huge amount of data entry time, and limited time between pitches to write everything down.

The thing that’s been the most helpful to me, is the thing that makes the most sense. I don’t score the games the way its been done forever. Once I converted my thinking to the way the boys doing the ML data collection score the games, everything became a lot more simple. Breaking each AB down into the play itself, i.e. what happened strictly with the batter, what happened before the play, i.e. SBs, WPs, etc., and what happened after the play, i.e. runner on 2nd went to 3rd, runner on 3rd scored, batter went to 2nd on throw, etc., made things a lot easier for me and the computer.

Of course it could be done more efficiently and slicker, but I’m only one soon to be 65YO by myself with very limited resources. Wink

As for trying GC, I honestly have no desire to do that, since even if I did think it was the greatest thing since sliced bread, it wouldn’t integrate with my old data, and that’s very important to me. But I will say again as I’ve said in the past, from what I know about it, GC is a great product and will definitely fill the needs of 99.99% of the people out there.
stats4gnats, so a follow-up to my original post and maybe answers to your questions:

1. GC is awesome. My suggestion is to download it, and spend a day on MLB network..try to score one of their replayed games. It takes about 3 games to get good, but it is fun to toy with.
2. Pitch counts, it tracks every pitch, not the umpire's count, and it can reproduce every pitch as it happened (sorry, no spray chart of foul balls...just fair ones)
3. It handles virtually every stat, including spray charts for your hitters.
4. In your scenario of a triple that is changed back to GR2B, GC has a "go back" button - although it is too small for my fat fingers, but it will step you backwards. One time I went backwards more than one batter, made a correction, and then caught up in about 3 pitches.
5. GC produces a replication of a scorebook, and on a run scored it colors in the "block" just like most people do manually. Right now, that block is black ink. I suggested that they color code it so we could see whether they considered it earned (green) or unearned (red). I thought that would give me an easy way to validate their computerized interpretation.
6. GC applies the easy logic (man on 2B, no outs, single to right, run scores = earned). At the same time (man on 3rd, 2 outs, e6, run scores = unearned). It is the intepretive issues that are difficult. You know that because you tried to program for it.

So, hope that helps. Like anything else - GC handles 98% of all scenarios at every age level perfectly. It is the 2% that you relish that presents a challenge to even the best programmers. And then you balance that with the need to keep it simple enough for the dad of the 9yr old who just wants to help. In that respect, it is just about perfect!
Schwammi,

I’ll say it once more to hopefully be sure that no one gets the wrong idea. I think GC is a great product, and I’ve recommended it literally dozens of times in person, and in at least 50 different threads on 8 different baseball discussion boards.

The main problem I have with it, is it will not incorporate the mass of historical data I’ve accumulated. Another big problem I have with it, is that there are many many metrics I’ve designed myself that require certain kinds of data that wouldn’t be available to me anymore. I’m also not going to go out and by an IPad or smart phone just to score a ballgame, when I already have 3 laptops that work wonderfully well.

The final thing for me is, I really enjoy the programming, and since no one but me is ever going to use it, I can make it do EXACTLY what I want, without having to worry about whether anyone else likes it or not. IOW, it’s a totally piece of custom software that meets my needs like no off-the-shelf product could ever do.

However, I’m still waiting for someone to hook me up with whatever data GC puts out that could be “mined” by anyone who wanted to do so. Perhaps I’m wrong and everything I’d need is there, and if that were the case, I’m not above reconsidering my position.

Sometimes folks don’t quite understand my perspective. While most people have only seen apps like GC for a year or two, and are flabbergasted at the vast amount of information suddenly available to them, I’ve had all that information and more available to me for over a decade, and that gives me a much different perspective on the entire matter. Not a better perspective, just a different one.

Tell ya what though. You go take a look at http://infosports.com/scorekeeper/ and tell me GC can reproduce all the stats you find under the headings on the left, “Comparing This Season With Previous Seasons”, “Varsity Records thru 2011”, “Combined Stats Thru 28 May 2011”, and “Final 2011 Stats”, and I’ll definitely be much more interested in trying it out. Wink

Thanx for the info. I appreciate it, and couldn’t be happier that more and more people are getting to see and appreciate the things I’ve seen for so long. It really makes the game better, because more people are getting much more informed, and therefore don’t have to believe a lot of the old “stories” and “myths” just because some old guy told them. Smile
I went and visited your site and it is an impressive array of reports. I hope someone from GC reaches out to you. Your reports would certainly add to the array of reports they have.

Many of your reports focus on career statistics and records. GC captures the data you need, but I have not seen anything that compiles it for a career (although I have only used it for one year).

And the reports that you produce for "single season" stats are interesting. BA by position in the batting order was nice to look at. GC has the data to produce this, but even their export would not give you what you need until they expose more data. My sense is that they capture everything you would need, but don't give you a way (yet) to get to it. For example, as long as you properly record the line-up changes, it knows who was in what position on every at bat, and where that person batted in the line-up. But at this point, I don't think you could export that to feed your program.

Like I said, I hope someone at GC contacts you. You have probably already developed what they hope to some day achieve. At the same time, they are probably working full-time to improve the daily "game experience" more than they are focused on the analytics.

You have some neat stuff there. Well done.
quote:
Originally posted by schwammi:
I went and visited your site and it is an impressive array of reports. I hope someone from GC reaches out to you. Your reports would certainly add to the array of reports they have.


Thanx for the compliment. But it isn’t really fair to GC. I’ve been at this for about 20 years total, and they’ve only been at it for a couple years. Wink

quote:
Many of your reports focus on career statistics and records. GC captures the data you need, but I have not seen anything that compiles it for a career (although I have only used it for one year).


Actually, though it doesn’t seem like it, the available reports are exactly the same. I just don’t print them all for both.

quote:
And the reports that you produce for "single season" stats are interesting. BA by position in the batting order was nice to look at. GC has the data to produce this, but even their export would not give you what you need until they expose more data. My sense is that they capture everything you would need, but don't give you a way (yet) to get to it. For example, as long as you properly record the line-up changes, it knows who was in what position on every at bat, and where that person batted in the line-up. But at this point, I don't think you could export that to feed your program.


If you look closely, the “By Position” reports are really only extensions of some other report.

A lot of times things like that are done to prove what someone says is true or not true. FI, if you believed the team’s best hitter should be in the #3 position, how would you prove that what was really happening. If you believed OBP was the best definition of a hitter’s skill, you’d pull all the records, but you’d group them by BPOS and compute the OPB that way. But in order to do that, you have to store what the batting position was for each player in each game.

What it ends up being, is the same data being used over and over again, but being grouped and sorted differently. It really isn’t hard at all.

quote:
Like I said, I hope someone at GC contacts you. You have probably already developed what they hope to some day achieve. At the same time, they are probably working full-time to improve the daily "game experience" more than they are focused on the analytics.

You have some neat stuff there. Well done.


Here’s what’s funny. As I read board like this and talk to people about the game, I come up with some new thing I want to look at very often. My guess is, I experiment on something new at least once every week of the year, and actually put something new into the system at least once a month. I have at least another hundred things or so that I’ve had in the system, but for whatever reason removed them.

The thing is, most people up til now have pretty much been restricted like me. But the amount of data one could track on a scoresheet, and then get into some piece of software to analyze it. One guy can only do so much. When I was putting in everything I could, it was taking me anywhere up to an hour and a half for data entry for every game. Most people won’t do that.

But now, with these new apps, the data entry is a moot point. Heck, I’ve already got a list of about 70 new metrics I’m just itchin’ to develop. Many of them weren’t possible before because the data just wasn’t there. Now I’m finding I’m adding new things already, because it takes no more time to get it the way it once did.

I really am excited about what’s coming down the road for people at the lower levels who never had access to this stuff before. All of a sudden many things become very plain that weren’t easy to explain.

Add Reply

Post
×
×
×
×
Link copied to your clipboard.
×