Stats Show or Not to Show

I am a very strong believer in keeping good stats and sharing them. It's one of the purest forms of performance review. Not that numbers can never lie, they always tell a good story. A story that begs the question, "What do I need to do from here?" I think along with sharing the stats it's important to have a team discussion (and individual if needed) about kind of attitude they should evoke. They’re only to be used for positive purposes. They also serve a great purpose to explain (to players and parents) why one player is in the line-up over another.

there is only one stat that matters and we all know what that one is.

They also serve a great purpose to explain (to players and parents) why one player is in the line-up over another.

If a kid is striking out or walking a lot of players or not picking the ball I think he already knows that. If you have to show him the box score to tell him he is not playing well what does that tell you? Explaining to parents? do they watch the same game the coaches are watching?

quote:

Originally posted by Will:…
If a kid is striking out or walking a lot of players or not picking the ball I think he already knows that. If you have to show him the box score to tell him he is not playing well what does that tell you? Explaining to parents? do they watch the same game the coaches are watching?

How does a kid know how to measure his performance without any numbers? On one team a kid who strikes out once out of every 3 at bats might be bad, but on another team it might be good, so on 1 team it might be a reason he’s in the lineup, and on another the reason he’s not. Without having numbers to look at, everything’s left up to perceptions and memory, and that’s not a good situation to be in.

As someone who’s job in the game is marking down what happens in it “for the record”, I can say with deep conviction that not only aren’t the parent’s watching the same game as the coaches, the coaches often aren’t watching the same game I do. There’s no doubt about it that perspectives bring biases and prejudices to the table, and those biases and prejudices can be very subtle or very harsh indeed.

That’s why having someone who’s consistent and as unbiased as reasonably possible keeping score and producing the numbers, is important. Chances are, Tommy’s coach, his dad, and the scorekeeper saw his performance on any given day much differently, so its nice that they have a way to look at what happened without many of those biases and prejudices. It’s a place of commonality they can work from to try to see the other perspective.

Wouldn't "quality at bats" be the best stat? Surely batting average can easily be swayed up or down in a short season like high school or lower levels.

As a coach or teammate, who do you want up to bat when the game is on the line - the .350 hitter whose average includes a dozen or so weak bloop popups, seeing-eye grounders and other weak hits, with virtually no chance for an extra-base hit or the hitter whose .200 average includes a dozen or more rocket shots that were hit right at fielders?

quote:

Originally posted by Sandman:
Wouldn't "quality at bats" be the best stat? Surely batting average can easily be swayed up or down in a short season like high school or lower levels.

As a coach or teammate, who do you want up to bat when the game is on the line - the .350 hitter whose average includes a dozen or so weak bloop popups, seeing-eye grounders and other weak hits, with virtually no chance for an extra-base hit or the hitter whose .200 average includes a dozen or more rocket shots that were hit right at fielders?

When you can define a quality at bat that everyone agrees on, a discussion can begin.

Yes, BA at the HS or below levels can be “swayed up or down”. But then again. BA at levels above HS can be too. Its an AVERAGE, and unless a player would never get a hit or make an out, the AVERAGE would be SWAYED by each at bat. The only difference is, the more at bats used in computing the average, the less it would change with each event.

Who would I want up in that situation? I’d want the player with a history of making fewer outs per at bat! I wouldn’t look at BA, OBP, or QABs. I’d look at RBA(Reached Base Average), and if I really wanted to increase the precision, I wouldn’t count ROFCs, even though technically the batter did reach base.

quote:

Chances are, Tommy’s coach, his dad, and the scorekeeper saw his performance on any given day much differently,

Tommy strikes out looking 3 times or makes 3 errors at shortstop. what is there to see differently? On the other hand he goes 3 for 3 hits the ball hard each time and picks up everything that comes his way at shortstop.

`

quote:

Originally posted by Will:
Tommy strikes out looking 3 times or makes 3 errors at shortstop. what is there to see differently? On the other hand he goes 3 for 3 hits the ball hard each time and picks up everything that comes his way at shortstop.

You’re trying to make everything fit one template when things don’t work that way. I know lots of coaches who don’t see all strikeouts as bad, and quite a few who might even see them as a QAB. Say Tommy strikes out looking 3 times, but each time was an 11 pitch at bat?

The same thing happens with errors. 1st of all, he’s playing short because chances are the coach thought he was the best infielder on the team, with the widest range, strongest arm, and the best leader, so 1 game where he goes in the tank won’t mean jack. But even so, its possible that 1 error came when he made an amazing play getting to a ball no one else did, but throw it into the dugout for a hit and an error, the 2nd one came when he ran into short left foul territory, snagged a foul ball, then made a bad throw trying to get a runner going home, and the 3rd error when he was in the outfield as the cut-off man, then fired a great throw to get a runner trying to score, but the ball hit the rubber and bounded into the dugout. All 3 are legitimate errors, but I don’t know many coaches who’d see any of them as heinous offenses.

On the other hand if he goes 3 for 3, what difference does it make if he hit the ball hard or not? His job was to not make an out or cause an out to be made, and he did that whether it was 3 HRs or 3 duck farts.

No matter what happens on the field, chances are there will be at least 2 people who “perceive” them differently. I watched my son get benched for hitting a home run. The sign supposedly was a suicide squeeze, but he smoked the ball at least 40’ over the left field fence. The poor kid coming in from 3rd almost had a heart attack, his mother couldn’t stop screaming she was so frightened, and his dad wanted to kick my a$$ for raising my kid to be so selfish. Turns out after some investigation, the coach mistakenly took the sign off, but he’ already replaced the boy and embarrassed him for no reason.

The bottom line is, there’s almost always some difference between perception and reality.

at the end of the day after all the discussion it comes down to one thing. You either got the job done or you did not.

quote:

Originally posted by Will:
at the end of the day after all the discussion it comes down to one thing. You either got the job done or you did not.

But unless the end of the day was the last game of the season, somebody has to set a defense, a lineup, and pick a pitcher for the next game. If you’re satisfied with whoever doing that using only his gut and his perceptions about what’s gone on in the past, that’s fine. I’d prefer whoever’s job it is, actually makes an attempt to measure what’s going on.

If you’re satisfied with whoever doing that using only his gut and his perceptions about what’s gone on in the past,[/quote]

gut and perception? Believe it or not coaches have eyes. They SEE what a player has done.

quote:

Originally posted by Will:
gut and perception? Believe it or not coaches have eyes. They SEE what a player has done.

Of course they do, but that’s not the issue. The issue is whether or not they accurately remember every detail, or at least as many details as possible. Why does one player start over another, or one player bat 3rd rather than 8th? Its fairly simple to see during a game, but who do you know that can accurately remember all the details 2 days, a week, a month, or 6 months later, when all the games, at bats, pitches, and fielding chances blur together, to where the best you can do is recall the outstanding events, and then you wouldn’t bet your life on them.

Why is it that so many people believe using numbers to manage a business makes sense, but using them to manage a sports team is unnecessary? You sound like the guy who sat next to me Saturday. I had no roster, no lineup card, and half the team was made up of kids from last year’s JV team that I’d never seen before other than in passing. There was a pitching change in about the 5th, and I didn’t have a clue about the kid who came in. So I hollered out at some of our fans and asked what the boy’s name was. While they were trying to figure it out, the fellow next to me says, “Just put down ‘Lefty’.”

When I told him I don’t keep score that way, he went into a short dissertation about how the score wouldn’t be in the paper, and the game didn’t matter. I explained that if he was the scorekeeper an wanted to keep score like it was tee-ball, that was fine, but when I take the time out of my day to drive 20 miles one way to spend 6 hours scoring a double header, I’ll keep score the same way I would if it was the final game of the state playoffs.

Then I asked him the REAL question. If it were his son who’s name I was trying to find out that was trying to impress the coach and make next season’s varsity team and who had a great game, would he want the coach remembering that Joey Smith threw 3 innings, struck out 3 swinging and 2 looking, had a strike percentage of almost 70%, only gave up one bloop hit, no runs, picked a kid who’d reached on an error, didn’t go to 3 balls on any of the batters, …. Or would he rather have him remember Lefty had an pretty good game.
I ended with, “Of course I suppose that if your kid wasn’t very good you wouldn’t want anyone remembering the details.”

quote:

The issue is whether or not they accurately remember every detail, or at least as many details as possible.

Detail? What detail do you have to know if for the last 4 games your shortstop has not made the plays or a certain pitcher can not find the plate. What detail is there to errors and walks. On the other hand your shortstop is making all the plays and a pitcher is getting people out.

Numbers can be made to say anything you want them to. Although I am a big fan of keeping stats, in a short season the result is often in the eye of the beholder.

And also, in a short season, the coach's knowledge of his team, and that of the opponent, is much more critical than individual stats.

But the OP asked whether or not to show the kids their stats. I say, why not? But with a measure of instruction as to how easily they can change up or down when the denominator is low.

2

quote:

Originally posted by Will:
Detail? What detail do you have to know if for the last 4 games your shortstop has not made the plays or a certain pitcher can not find the plate. What detail is there to errors and walks. On the other hand your shortstop is making all the plays and a pitcher is getting people out.

Look, I get where you’re coming from, but you’re trying to put conditions on their use which only cover limited situations. Is it better for everyone if as many factors are considered before making decisions for the vast majority of situations? If you believe all there is to making the best decision possible are only the few things you mentioned, then by all means don’t bother to look at the REAL numbers. But I’m telling you that I’ve been doing this for a long time now, and there aren’t many people I’ve every run across who’s perceptions are either as good or bad as what actually took place.

quote:

Originally posted by RedSoxFan21:
Numbers can be made to say anything you want them to.

Of course they can. But that kind of thinking presupposes that someone is going to purposely twist them in order to validate or further their own purposes. Think about that. Is it likely that would happen in the business world? ABSOLUTELY! But what would the purpose be to do it on an amateur baseball team?

quote:

Although I am a big fan of keeping stats, in a short season the result is often in the eye of the beholder.

It totally depends on what you’re trying to do with them. If you want to get a good idea about how the players on the team are performing against each other, what difference does it make whether the season is 10 games or 100 games long? Now if you’re trying to decide whether or not pay for some kid’s hotel room 3 weeks after the 1st game based on the 1st game’s stats, that’s something else again.

quote:

And also, in a short season, the coach's knowledge of his team, and that of the opponent, is much more critical than individual stats.

That’s quite different than the numbers mean nothing and serve no useful purpose.

quote:

But the OP asked whether or not to show the kids their stats. I say, why not? But with a measure of instruction as to how easily they can change up or down when the denominator is low.

Its interesting that you should add that last caveat. I happen to agree, and spend a lot of time during the season making sure I explain that. But who’s job is it? I’m the SK/statistician not the coach. I don’t have to explain why the coaching decisions that take place are made. All I do is my best to keep score accurately, manipulate the numbers using proper math, and present the results. I don’t interpret ANYTHING!

But when there are “problems”, it’s almost always the interpretation that causes them, not the numbers nor the math. If the coach says I’m going to put my 9 best players on the field, but doesn’t bother to explain how he’ll determine who those players are, then each individual is on his/her own as to how to define that. If I happen to believe BA is the best indicator, that’s what I’m looking at. But you might believe its at FPct + OPB and that's what you'd be looking at. But if the coach believes its about QABs, then we’re both wrong and have no way to make a valid judgment, and that’s the kind of thing that leads to “issues”.

quote:

Originally posted by Stats4Gnats:
If you want to get a good idea about how the players on the team are performing against each other, what difference does it make whether the season is 10 games or 100 games long?

Hey Stats, I've seen how passionate you are on this topic throughout HSBBW and I respect that. But on this point I must risk falling into the abyss to say that with a small sample the statistics are absolutely not a good decision driver. How those outs or hits or steals or ERAs or what else are achieved can be heavily influenced by the fine lines of a particular play. Over time those lines get erased because the extremes even out, the luck gets balanced. But in a small sample there can be wild variations. You still need a human to filter the results, particularly during the short season that is New England high school baseball.

quote:

Originally posted by RedSoxFan21:
Hey Stats, I've seen how passionate you are on this topic throughout HSBBW and I respect that. But on this point I must risk falling into the abyss to say that with a small sample the statistics are absolutely not a good decision driver. How those outs or hits or steals or ERAs or what else are achieved can be heavily influenced by the fine lines of a particular play. Over time those lines get erased because the extremes even out, the luck gets balanced. But in a small sample there can be wild variations. You still need a human to filter the results, particularly during the short season that is New England high school baseball.

I understand your thought process, but I never tried to imply stats of any sample size were good decision drivers, and never will.

Yes, things do balance themselves to some degree, but that’s not at all my point. 1 hit in 4 at bats is .250, just as 200 hits in 800 at bats is .250. That means the batter averaged 1 hit every 4th at bat, not that he got 1 hit ever 4th at bat. He could go 200 for 200, then go hitless in the next 600 at bats and he’d still average .250, the same as if he went 0 for 600 then got 200 hits in a row. The thing is, no matter what the sample size, the math is exactly the same.

Of course any decision made by that number has to be taken in context. FI, if it’s the bottom of the 7th with 2 down in the 7th game of the WS and you had 2 guys left on the bench and both were hitting .250, would you send up the guy who was 0 for his last 600 or the guy who had the 200 straight hits? The decision drive isn’t the number or the sample size, it’s the perspective and the context.

I have a head coach who is very much “old school” and believes in going with the guy who has the hot hand. Its very much along the lines of what have you done for me lately. After every game I send him a package of several things that took place in the game. The batting stats, runner disruptions, the pitching stats, the fielding stats, and one other thing. If you go to http://www.infosports.com/scor...images/batting12.pdf and do a find on “hot”, you’ll see what it looked like after the last game of the season.

Of all the things available, I can take that one and see a direct link to the batting order. There was only 1 player who was “locked” into a batting spot for the entire season, but the other kids moved up and down in the order in pretty close relation to that metric. I don’t mean a kid will move from the 8th spot to the 3 because he got hits in the last 5 games, but you can pretty much bet your patoo that he won’t be batting in the 8th spot anymore. I also send something else in the package. Do another find on “recent”. What that one is, is the stats for only the last half of each players games. Its another one of those what have you done lately things.

Those two things are definitely intended to do one thing. Focus attention on what the hitters are doing now, not what they might have done 6 weeks ago. But the trick is, they can’t be looked at by themselves alone. Those things are only a report of what took place on the field during games. There’s nothing there about grades, attitudes, being on time, work ethic, leadership, hustle, or any of the other factors the go into the total picture of a player.

Like I say over and over. I don’t interpret anything! I just score it and present it, and sample size hasn’t got a dang thing to do with what’s printed on all those pieces of paper.

Stats,

In Statistics the science (not sports) the size of the sample does matter from the standpoint of the liklihood that the results of the sample have validity.

It is why Polling companies can poll 1,500 people and have a margin of error of +/- 3%. If they polled a larger sample the margin for error declines just as a smaller sample increases the margin of error. Conceptually over time everything regresses to the mean. There is always some variance but it continues to move toward the mean over time.

As you state the numbers are what the numbers are and the point of the thread is how much value should be placed on them. The other factor that complicates things is the disparity of talent in HS. A 50 plate appearance sample could be skewed dramatically two 3 hit games against a poor pitching opponent. It adds at least 120 points to the batting average which turns a .280 hitter into a .400 hitter. If that player can't get around on a 86 MPH fastball and he's now facing a stud throwng 89 he won't hit .400 over 50 plate appearances.

I'd suggest that his numbers would need to be heavily discounted or disregarded altogether in that situation. I would not bat him leadoff if I had faster bats available to me.

quote:

Originally posted by luv baseball:
In Statistics the science (not sports) the size of the sample does matter from the standpoint of the liklihood that the results of the sample have validity.

It is why Polling companies can poll 1,500 people and have a margin of error of +/- 3%. If they polled a larger sample the margin for error declines just as a smaller sample increases the margin of error. Conceptually over time everything regresses to the mean. There is always some variance but it continues to move toward the mean over time.

Never said anything different.

quote:

As you state the numbers are what the numbers are and the point of the thread is how much value should be placed on them.

That’s not the idea I got from the OP at all. My understand was where or not to show the numbers to the players during the season, not at all about their validity nor their value.

quote:

The other factor that complicates things is the disparity of talent in HS. A 50 plate appearance sample could be skewed dramatically two 3 hit games against a poor pitching opponent. It adds at least 120 points to the batting average which turns a .280 hitter into a .400 hitter. If that player can't get around on a 86 MPH fastball and he's now facing a stud throwng 89 he won't hit .400 over 50 plate appearances.

But if you take your example the other way, 2 oh-fer games against top caliber pitching negates whatever advantage there was on the other side. That’s why they’re averages, where everything is considered.

But in HS just as at the ML level, the only pitchers that “stink” are when the manager has no choice but to put in some kid with little or no experience. Like when a ML manager has to put in the Right Fielder because he’s out of pitchers and the guy threw when he was in HS. Not only that, but there simply aren’t a whole lot of HS players gonna be able to get more than a few hits off a true 89 cruiser over 50 PA’s. I’d be surprised if there was an average of 3 on a team who’d even hit the ball more than 50’ into the OF on any kind of regular basis.

quote:

I'd suggest that his numbers would need to be heavily discounted or disregarded altogether in that situation. I would not bat him leadoff if I had faster bats available to me.

And that’s exactly what coaches do, whether they do it with math or not. Through the years I’ve tried to work in a “factor” to adjust pitching and hitting, but there really is no such thing that can be consistently counted on. What I’ve always done, was to use the opposition’s Winning Percentage as the factor. IOW, if a team has a .500 record, any hits against them are only worth half of what they would if the team had won all of its games. The opposite would be true for the offense. If a player went 2-4 against a .500 team, his average for that game would be .250 not .500 because his hits would be less “valuable”.

Sadly though. If you’ve seen much HS ball, depending on how the schedule’s created, there’s just as much as chance of playing the 18-2 team but seeing the Soph the coach just promoted, as either of the studs that were 8-0 with over 100 Ks and only 5 walks each. So the kid who gets 3 hits off that Soph is doing you-know-what in tall cotton, while the kid who went 0-5 against the kid on the 6-14 team who’s gonna be a 1st round draftee goes right into the dumper, and no one understands why. I once tried to do it using the opposing pitcher’s record, and even trying things like the opposing pitcher’s WHIP. Trouble is, that information isn’t readily available like it is for ML pitchers.

But the thing to understand is, as long as the numbers are only used to evaluate the evaluator’s team, it doesn’t matter! In general the starting hitters all face the same pitchers and the starting pitchers all throw to the same hitters, so there’s no reason the numbers wouldn’t be valid for making team decisions. But what people do is think about HS numbers the same way people think about ML numbers. There is no Fantasy HS baseball, and no one is getting traded or paid base on their performance relative to the rest of HS baseball.

quote:

But the thing to understand is, as long as the numbers are only used to evaluate the evaluator’s team, it doesn’t matter!

True, but only in a meaningful sample size - which 50 ABd wouldn't be.

quote:

Originally posted by Sandman:
True, but only in a meaningful sample size - which 50 ABd wouldn't be.

It all depends on what one is trying to do with the information. After the 1st 20 games of the ML season, even with all of the historical info they have available, do you think no manager uses what had taken place as input into their decision making?

Again, it all depends on what one is doing with the information. Would I use the information from 50 at bats to judge whether to give a player a $220M contract for 10 years? Heck no! But I might use it to move my 7th hitter to 6th, or move my 2b back to get more range. But even at that, you have to remember that there’s no choice in the matter. A HS coach doesn’t often get hitters with 100+ ABs or pitchers with over 70 IPs. Unless he’s some kind of megalomaniac fool who believes himself to be infallible, he has to have some basis for making decisions, so he has to use what’s available.

"Stats"
have you applied for for a "stats" job with a minor league radio announcer.
When I visited the Dodgers radio booth with the Vince Sully, he had a "stats" person in the booth to provide "back up" info.

When I was the radio announcer for the Redwood Pioneers in Rohnert Park,
1st year California League. Ken Korick, now Oakland A's announcer was #1.

When the SS came to the booth, after a error charged, his criticism was directed to the scorer. "how dare you charge an error, I cannot advance w/ a low fielding %. This was Class A California League.

"Always Learning"
Bob

PS: the best stats are wins and losses