[Zeeter]

I'm hearing a lot of talk here about getting rid of the avoid hits rating for pitchers. That is a big mistake in my opinion. I understand the logic - pitchers may only affect walks, K's, Homers, and GB/FB%. Defense is responsible for the rest.

I'm sure that for pitchers like Johnson and Schilling this is fairly accurate. However, look at pitchers like Maddux and Glavine. They do not overwhelm with strikeouts. They don't walk a lot of hitters, but a lot of pitchers are like that. It is their pinpoint control that keeps the hits off of the board. It is their ability to keep the hitters guessing and not allowing them to get good wood on the ball.

Personally, I don't mind if we get rid of the avoid hits rating, but if we do, I suggest that the control rating be used more for avoiding hits. Maybe it is and this has already been discussed and I'm just blowing hot air - this wouldn't be the first time.

Anyway, that's my 2 cents.

[ctorg]

I don't think it would be gotten rid of entirely, just greatly reduced.

[Gastric ReFlux]

Originally posted by ctorg
I don't think it would be gotten rid of entirely, just greatly reduced.

I think that's the thinking.

[blackamus]

I just don't understand it. This guy makes some outlandish, counterintuitive, and entirely bogus argument and it seems that all of you little Rob Neyer's of the world are eager to jump on the bandwagon. What gives here? We all realize that their is a certain amount of "luck" involved in whether or not a batted ball is a hit or not (i.e. every well struck ball is not a hit whilst some poorly struck balls are). But, how sexy does that sound? Not nearly as sexy as saying that a pitcher has "no effect on whether or not a batted ball becomes a hit".
Stats are obviously one important analytical tool. However, when mathematical evidence points to an irrational solution, that solution is usually discarded.
And while I'm on this train...F**k Sabrmetrics. If baseball were run as sabrmetricians wished it would be a boring and aesthetically displeasing game. A walk is quite obviously not as good as hit. Batting average quite obviously does matter seeing as it usually accounts for at least 75% of OBP. But it sounds so much better to make ridiculous counterintuitive arguments such as "batting average doesn't matter" or "a pitcher has no effect on whether a batted ball is a hit" followed by shoddy and selective statistical records to back up said nonsense.

[ATypicalGiantsFan]

Just because people such as Voros McCracken overstate their cases doesn't mean that everything they say is meaningless. Batting Average as a primary tool to measure a players value is not that useful, especially with stats such as OPS. I'm not, or just about anyone else saying Batting Average is useless as it can tell you something about a player. The same goes for pitchers giving up hits. Voros overstated a bit but it is true that pitchers giving up hits is a lot less out of their control than previously thought.

[blackamus]

This guy (McCracken) is supposedly a statistician, right? And yet he makes the fundamental statisitcal error of basing conclusions on an insufficient sample size!

[Johnny Slick]

Hate to bring this argument up again, but... if you're looking for guys whose career DIPS don't line up well with their actual stats, Maddux and Glavine may not be the best choices. Maddux's per-162 averages are 215 hits in 237 IP. He's effective not because he doesn't give up many hits despite his K rate (8.15 H/9IP is right in line with what you'd expect a guy who strikes out 6.33 batters per 9 innings), he's effective because he doesn't walk ANYBODY (under 2 BB/9IP) or give up many homers.

Glavine's a bit tougher of a nut to crack, but even so his H/9IP numbers are not way out of line with his K/9IP numbers. Like Maddux, he doesn't give up many homeruns at all, so that lowers things even further. He walks a few more hitters than Greg, but his control is still impeccable.

Outside of the 1940-60 period, the only guy that I know of who consistently outperformed his DIPS was Jim Palmer. And even there, you've got to admit that there were extenuating circumstances; who else in baseball history could count on 4 Gold Glovers in the field when he pitched?

John Craven

[Gastric ReFlux]

But, how sexy does that sound? Not nearly as sexy as saying that a pitcher has "no effect on whether or not a batted ball becomes a hit".

Please, a lot of young men frequent this board and we don't want them getting all excited.

[draven085]

Originally posted by blackamus
I just don't understand it. This guy makes some outlandish, counterintuitive, and entirely bogus argument and it seems that all of you little Rob Neyer's of the world are eager to jump on the bandwagon. What gives here? We all realize that their is a certain amount of "luck" involved in whether or not a batted ball is a hit or not (i.e. every well struck ball is not a hit whilst some poorly struck balls are). But, how sexy does that sound? Not nearly as sexy as saying that a pitcher has "no effect on whether or not a batted ball becomes a hit".
Stats are obviously one important analytical tool. However, when mathematical evidence points to an irrational solution, that solution is usually discarded.
And while I'm on this train...F**k Sabrmetrics. If baseball were run as sabrmetricians wished it would be a boring and aesthetically displeasing game. A walk is quite obviously not as good as hit. Batting average quite obviously does matter seeing as it usually accounts for at least 75% of OBP. But it sounds so much better to make ridiculous counterintuitive arguments such as "batting average doesn't matter" or "a pitcher has no effect on whether a batted ball is a hit" followed by shoddy and selective statistical records to back up said nonsense.

Its probably not even worth it, but I'd like to point out that the sabremetric offensive philosophy is not based on taking walks. Its based on hitters swinging at pitches that they can hit hard somewhere. Walks are a byproduct of this philosophy. No one ever said batting average was not important, but it has been traditionally overrated.

I'd like to see an explanation of how McCracken's argument is entirely bogus.

edit: typos

[crackpott]

I'd like to see one too. I've seen a lot of bold statements proclaiming McCracken's work to be meaningless simply because people refuse to accept such a groundbreaking discovery. I'd like to see some proof of this...

[Dwolfson20]

Originally posted by blackamus
I just don't understand it. This guy makes some outlandish, counterintuitive, and entirely bogus argument and it seems that all of you little Rob Neyer's of the world are eager to jump on the bandwagon. What gives here? We all realize that their is a certain amount of "luck" involved in whether or not a batted ball is a hit or not (i.e. every well struck ball is not a hit whilst some poorly struck balls are). But, how sexy does that sound? Not nearly as sexy as saying that a pitcher has "no effect on whether or not a batted ball becomes a hit".
Stats are obviously one important analytical tool. However, when mathematical evidence points to an irrational solution, that solution is usually discarded.
And while I'm on this train...F**k Sabrmetrics. If baseball were run as sabrmetricians wished it would be a boring and aesthetically displeasing game. A walk is quite obviously not as good as hit. Batting average quite obviously does matter seeing as it usually accounts for at least 75% of OBP. But it sounds so much better to make ridiculous counterintuitive arguments such as "batting average doesn't matter" or "a pitcher has no effect on whether a batted ball is a hit" followed by shoddy and selective statistical records to back up said nonsense.

Your overdramatization isn't helping your case. Nobody says that "batting average doesn't matter." It matters a lot less than people think it does. Big difference.

And math is generally considered to be rational, therefore I don't see how it comes to irrational conclusions. Maybe most people think they are irrational, because they aren't thinking the right way, but that doesn't make it so. If something is proven with statistics, it's a lot more plausible than if a bunch of people just say something is true.

[holyroller]

Originally posted by crackpott
I'd like to see one too. I've seen a lot of bold statements proclaiming McCracken's work to be meaningless simply because people refuse to accept such a groundbreaking discovery. I'd like to see some proof of this...

http://www.diamond-mind.com/articles/ipavg2.htm

Now, Tippett doesn't say DIPS is meaningless, on the contrary it reaffirms it's usefulness, but he does prove that McCracken overstated his case which I think a lot of us already suspected.

[crackpott]

I read that, and it by far doesn't put his work at "outlandish, counterintuitive, and entirely bogus" It states that there is likely a slight affect that a pitcher has... which most have already attested to. Still, many of his examples, including his prime example in the argument are knuckleballers, anyway, which takes away from it.

[holyroller]

Originally posted by crackpott
I read that, and it by far doesn't put his work at "outlandish, counterintuitive, and entirely bogus" It states that there is likely a slight affect that a pitcher has... which most have already attested to. Still, many of his examples, including his prime example in the argument are knuckleballers, anyway, which takes away from it.

I agree. Just thought I would point it out w

[Zeeter]

I don't know how this discussion turned to discussing the merits of batting average, but I'll bite.

Batting average has always been the measuring tool for offensive production. That and Home runs. What's the first thing you look for on a player? BA and HR.

Lately, many statisticians and other people who want to look smarter than everyone else have come to the conclusion that BA is far less important than it once was. Well, for all of you old timers out there, I'm going to take one for the team and defend the batting average.

The best measure of a hitter is his batting average. Sure, he can play "lawer ball" and go for a walk, but unless the bases are loaded, it is not as good as a single, and there is no chance for a double, triple, homer, or an error on a walk (unless it is a wild pitch, which could happen whether it's ball four or ball one).

I'll agree that OBP is very important, and has been neglected for years, but let's not get so carried away with OBP that we ridicule the BA. Me? Unless the guy is leading off, I'd rather have a guy hit .310 with a .380 OBP than have a guy hit .290 with a .400 OBP.

I just did a search on ESPN stats to prove my point and found arguments for both sides. Frank Thomas - low average, high OBP. He has 25 homers though which has produced a ton of RBI's. Edgar Renteria - high batting average, relatively low OBP (when compare to his average). He is right up there in RBI's.

I guess this is why we came up with OPS, to get a better look at the player's production. OPS, however, is not a good representation in all cases as it tends to reward power hitters over lead off/speedy players.
One thing that struck me when looking at the stats was that everyone in the top ten for On Base Percentage has an OBP of almost 100 points or higher than their average except for one - Albert Pujols. His is only 65 points higher.
In other words, while Todd Helton and Brian Giles' OBP are greatly helped by walks, Pujols OBP is not. Pujols is also four homers away from the triple crown right now.
I'm also looking at a guy like Vladimir Guerrero. This guy has steadily increased the difference between his BA and OBP from fifty-something to eighty-something. Yet his runs scored have remained relatively the same. He had more hits and more walks than any time in his career last year and had his best OBP. Yet his RBI's were not as high as he had in previous years (although higher than the last year by 3). It looks to me like Vlad's better OBP has not necessarily made him a more productive batter.
In conclusion, let me just say that I'm not trying to discredit OBP. I am trying to credit BA.

[Eckstein 4 Prez]

Originally posted by Gastric ReFlux
Please, a lot of young men frequent this board and we don't want them getting all excited.

The sheep's on the board..... I'm sitting here at my desk trying not to make audible laughing sounds.

[Eckstein 4 Prez]

Originally posted by blackamus
I just don't understand it.

You could have stopped your post right there and been correct.

[Gastric ReFlux]

oy oy oy

What's the job of a hitter?

Answer: To create runs.

Do we know of something that can, in a sense, remove what a hitter has done from the context of the players batting around him?

Answer: Yes

What is it?

Answer: There's a basic formula known as runs created. The simplest version is written like this ((H + BB)*(TB))/(AB+BB)

What do you mean?

Answer: I mean if you take that formula, and plug in a team's batting stats, you'll see it in almost all cases give a result that is quite close to what that team actually scored.

Gee, that's nice, but why would that work for individual players?

Answer: If you go through the players individually, and calculate their runs created, and add those up, it will again work out close to what the team scored. Furthermore, you can oftentimes compare that runs created result with the simple (Runs + RBI)/2 and again those two numbers will be close. The variance is often greater than at the team level, because the sample size has gone down and how many runs a player scores and drives in is dependent upon his position in the batting order and the players around him in the order.

So are you saying batting average is unimportant?

Answer: no, just saying that one should recognize that is a component of how runs are created. The runs created formula has three distinct areas: measuring how often the player got on base (H+BB), how far he moves people with his hitting (Total Bases), and how often he does those things with the divsion into his plate appearances (AB+W)

So wait a second, if I look at that formula, a walk isn't quite as good as a hit, because walks aren't part of total bases, are they?

Answer: correct

But still, I guess walks would be important then, why's that?

Answer: Because if you get on base you have a chance to score. Walks are positive contributions. Also, drawing a walk doesn't give up an out. Outs are a precious commodity in baseball, you only get so many of them and you don't get extras.

So maybe the statheads are just saying we should try to figure what's important to creating runs?

Answer: Yes, batting average is part of that. OBP is a bigger part of that. Slugging is part of that.

[Eckstein 4 Prez]

Originally posted by Zeeter
I don't know how this discussion turned to discussing the merits of batting average, but I'll bite.

Batting average has always been the measuring tool for offensive production. That and Home runs. What's the first thing you look for on a player? BA and HR.

Lately, many statisticians and other people who want to look smarter than everyone else have come to the conclusion that BA is far less important than it once was. Well, for all of you old timers out there, I'm going to take one for the team and defend the batting average.

The best measure of a hitter is his batting average. Sure, he can play "lawer ball" and go for a walk, but unless the bases are loaded, it is not as good as a single, and there is no chance for a double, triple, homer, or an error on a walk (unless it is a wild pitch, which could happen whether it's ball four or ball one).

I'll agree that OBP is very important, and has been neglected for years, but let's not get so carried away with OBP that we ridicule the BA. Me? Unless the guy is leading off, I'd rather have a guy hit .310 with a .380 OBP than have a guy hit .290 with a .400 OBP.

I just did a search on ESPN stats to prove my point and found arguments for both sides. Frank Thomas - low average, high OBP. He has 25 homers though which has produced a ton of RBI's. Edgar Renteria - high batting average, relatively low OBP (when compare to his average). He is right up there in RBI's.

I guess this is why we came up with OPS, to get a better look at the player's production. OPS, however, is not a good representation in all cases as it tends to reward power hitters over lead off/speedy players.
One thing that struck me when looking at the stats was that everyone in the top ten for On Base Percentage has an OBP of almost 100 points or higher than their average except for one - Albert Pujols. His is only 65 points higher.
In other words, while Todd Helton and Brian Giles' OBP are greatly helped by walks, Pujols OBP is not. Pujols is also four homers away from the triple crown right now.
I'm also looking at a guy like Vladimir Guerrero. This guy has steadily increased the difference between his BA and OBP from fifty-something to eighty-something. Yet his runs scored have remained relatively the same. He had more hits and more walks than any time in his career last year and had his best OBP. Yet his RBI's were not as high as he had in previous years (although higher than the last year by 3). It looks to me like Vlad's better OBP has not necessarily made him a more productive batter.
In conclusion, let me just say that I'm not trying to discredit OBP. I am trying to credit BA.

You make some decent points, but there are a few things I'd like to point out:

1. No one is arguing that BA is not important, just that it's been overrated. To put it in Moneyball terms, there's no opportunity to be made with BA, because everyone looks at BA to determine a player's worth. You have to look at the stats that not everyone is looking at "first," as you put it, and that's where OBP becomes so valuable.

2. Your Frank Thomas/Edgar Renteria/Vlad Guerrero point doesn't help much, because while BA is important, RBIs (which you use as evidence) are not. They are merely a function of how many guys are on base when a hitter comes up.

3. Your point about the .310/.380 hitter vs. the .290/.400 hitter is difficult to argue against without looking at more. However, assuming power and other attributes were the same, you'd be foolish to take the .380 OBP guy over the .400 OBP guy. Statistical analysis shows that outs are scarce - you need to conserve them.

4. The best measure of a hitter is certainly not batting average. Here's a fairly simple test you can run yourself: Go back and look at team performances through baseball history. Sort teams by batting average, as most books do, then by slugging percentage, and then by on-base percentage. You'll find that on the team level, on-base percentage and slugging correlates to runs scored (which is, after all, the point) much more closely than batting average. Sample size will be key here - but this is such an obvious point that I think even looking a one or two seasons will show you.

5. Once again - everyone agrees that hitting is valuable, and that a hit is better than a walk. But look at it this way: Let's use your .310/.380 example. Assume two players, each of whom have 62% of their hits as singles, 20% as doubles, 3% as triples and 15% as home runs. If both guys got 600 plate appearances, your .310 hitter would have 7 singles, 2 doubles, 1 triple and 2 home runs more than the .290 hitter. However, the .290 guy would have 24 more walks/hit by pitches than the .310 guy. This is a really close call between the two guys - I don't have in front of me the estimates about how many estimated runs each of these is worth, but it looks to me like your .310 guy creates slightly more. However, this is balanced by the fact that he also creates 12 outs more than the .290 guy - in RC/27, I imagine the .290 guy would come out a little ahead.

In addition, think about how the marketplace would value these guys. One would be a 28 HR, .310 hitter, the other a 26 HR, .290 hitter. Looking just at that, the market would likely put a slightly higher value on the slightly worse player. Thus, you'd be competing with more teams for the worse player. That is why sabermetrics is successful as a philosophy and it's how Oakland has achieved success beyond their market slot.

Edit: the sheep already made some of these points. Oh well....

[dougaiton]

I posted this earlier in another thread, but here it is revised:

Would we sacrifice realistic results for knuckleballers and the Paul Quantrill's of this world to make a game that allowed us to scout better with sabermetric methods for 'normal' pitchers? I agree that most pitchers results are decided by BB/HR/K rather than balls put into play ratings, but I would also suggest that OOTP changing to this system as a rule behind its stats would mean one of two things:

1) a loss of realism in generating knuckleballers stats
2) a new system to determine the way in which certain pitchers work.

Nobody responded last time; can anyone see light at the end of the tunnel and a way around this?

I am all for a K/BB based game engine, but not if it ruins the worth of a small but significant selection of pitchers