The Bullpen Has Blown 7 Games for Trachsel

[Sports Guy]

You mean to tell me a scout or anyone who pays close attention to watching a player bat four times a game is not able to tell if he is a Millar or Cust type or a wild hacker like Tejada?

Sure but this has absolutely nothing to do with you saying you can tell those things from a scout. I can look at a stats sheet and see 100 walks and 120 k's and know what type of player i have. I

can watch Bobby Abreu for example and tell you the guy gets a lot of walks just like Giambi because they take borderline pitches all the time. I don't need stats to see that.

Of course you don't need a stat for that because well, there isn't a stat for that. You are just thinking of examples that make no sense.

I think you and some here operate the other way around

Don't think.

[Fan4Life]

Yea and he has been a statisical anomoly this year.

Betting on that is a bet you lose 9 times out 10.

Agreed. I'm not advising anyone to take that kind of bet, merely pointing out as you have just stated, stats help predict what the outcome likely will be, but there are anomalies.

[Sports Guy]

So what people play the lottery with a million times worse odds yet there is a winner isn't there?

There are a lot more people who play every week spends hundreds of thousands of dollars in their lifetime and never win.

Your examples keep getting worse and worse.

[Old#5fan]

Sure but this has absolutely nothing to do with you saying you can tell those things from a scout. I can look at a stats sheet and see 100 walks and 120 k's and know what type of player i have. IOf course you don't need a stat for that because well, there isn't a stat for that. You are just thinking of examples that make no sense. Don't think.

There isn't a stat that shows Abreu is among the league leaders in walks? Why do you think that is the case? Uh, the fact he takes a lot of borderline pitches? Unbelieveable. It makes perfect sense if you watch each player and study how they approach an at bat!!

[RShack]

Well, it apparently has never happened before in the history of baseball. Is it smart to bank on it happening for the first time??

That's *exactly* the question I asked you about the wisdom and plausibility of your scheme for turning over 80% of the starting line-up... and your response was basically "Don't know... who cares?... why does it matter?"

[Old#5fan]

There are a lot more people who play every week spends hundreds of thousands of dollars in their lifetime and never win.

Your examples keep getting worse and worse.

Oh, I agree with you. I think it is a stupid waste of money. But don't tell that to people who have won the lottery, including some who have won it twice!

[Sports Guy]

Agreed. I'm not advising anyone to take that kind of bet, merely pointing out as you have just stated, stats help predict what the outcome likely will be, but there are anomalies.

Well of course...No one has ever said different.

You are trying to figure out the most likely outcome.

[Old#5fan]

Agreed. I'm not advising anyone to take that kind of bet, merely pointing out as you have just stated, stats help predict what the outcome likely will be, but there are anomalies.

And there are more bad predictions than good when using stats as the sole or primary basis.

I would love for all these stats gurus to predict what Aubrey Huff and Melvin Mora will put up stats wise next year?

I would also like to know how many of them predicted Baez would be so horrible this season?

[ScottieBaseball]

No, they would not be. Every single time someone is asked here to solely use stats to predict anything all they do is hedge and give possibilities. So in reality just how useful are these stats? They are not much better than guessing when it comes to "predicting anything." As a measurement of what has already happened they have great value but other than that they do not stand on their own as much value at all. A observation by a scout would be of more benefit in most cases.

I will give you an example. Suppose a guy is hitting 300 in the high minors but in reality he is hitting 400 on fast balls and 200 on curves. So if you look at his stats he gets promoted to the majors. Once he gets to the majors scouts have already told other teams the guy has trouble with the curve. Now he gets a steady diet of even better quality (i.e. major league curves) and hits below the mendoza line and gets sent back down. Now how reliable were the stats in this case? I will tell you, they are not even presenting an accurate picture period, or the kid wouldn't even be called up until he learns how to hit a curve ball much better. So relying on a scout is far better than any stats.

No one, not one single "stat head" in this thread said stats and stats alone are the sole predictor of the future. What has been repeated over and over again is that they are an effective tool widely used and accepted throughout baseball, from Little League all the way up to Major League Baseball. Scouts, managers, coaches, and fans alike all put them to use.

I challenge you to find one baseball professional, whether it's a scout or coach, that will unequivocally agree that statistics are the equivalent of a crapshoot. You might find an oddball or two that will they give stats very little thought, but I can assure the gross majority put them to use in analyzing a player's past as well as his future potential.

You do realize, by the way, you just used statistics in an attempt to support your position, but instead just proved their value, don't you? Your curveball-to-fastball ratio? That, my man, is a statistic. More often than not, a guy with a weakness like the one you described gets promoted with the hopes that he goes against the numbers and overcomes his shortcomings.

[Sports Guy]

There isn't a stat that shows Abreu is among the league leaders in walks? Why do you think that is the case? Uh, the fact he takes a lot of borderline pitches? Unbelieveable. It makes perfect sense if you watch each player and study how they approach an at bat!!

No, there is no stat that says how many borderline pitches someone takes.

[Baltimoron]

No, I don't care to sift thru 11 textbooks on stats and human performance and create a summary.

You don't have too. You wrote this:

[*]Previous performance is the best predictor of future performance, but it is nonetheless an unreliable predictor.

[*]For a given individual, the previous performance of that individual is a better indicator than normative data for a group, but even that is still an unreliable predictor (i.e., low r-value)

What do you mean by unreliable? (an r value?)

What "low r-value" are you thinking off?

--------------------------------------

Long story short, we know the theoretical limits on any projection system's accuracy for an offensive stat like OBP or OBA is an r in the .72-.73 (for 550 PAs) range:

What does this mean? You can take several hundred ballplayers, give them 550 PA one year, give them 550 PA another year, make sure that these guys’ true talent level in OBP does not change, make sure they play in the same parks, make sure they face the same quality of pitchers, and their year-to-year correlation will be 0.72. That is, the absolute maximum year-to-year r you can hope for, given a large number of ballplayers is .72.

How about instead of OBP, we look at wOBA (which is analogous to OPS)? Here, our var(true) is .036^2, var(random) is .022^2, and var(observed) is .042^2. Our r is .73.

So, when looking at forecasters, and you look at their correlation coefficient of their forecast to the actual results, anything close to .73 means that they did as good a job as possible. (They could actually go over that level, since the number of players in their sample is still small enough that the level of uncertainty of that r will be a bit high. But, given thousands of players over several years, that uncertainty level will drop quite a bit.)

link

Here are some r value for various projection systems from 2006:

Shandler .702

James .685

ZIPs .684

Chone .677

Marcel .664

link

Perhaps its semantics, but I don't consider these to be "unreliable predictors".

-----------------------------------------------------

Obviously, its much worse for pitchers because of all the noise in ERA, but less so if you use component ERA.

Here are some r-value for various systems projecting 2006 ERA numbers:

.459 ZIPs

.445 Baseball Info Solutions (Bill James has nothing to do with this although its in his Handbook, he claims it can't be done)

.442 Marcel

.423 Shandler

ERA itself has an r of .29

link

[Sports Guy]

And there are more bad predictions than good when using stats as the sole or primary basis.

I would love for all these stats gurus to predict what Aubrey Huff and Melvin Mora will put up stats wise next year?

I would also like to know how many of them predicted Baez would be so horrible this season?

Many of us predicted Baez would be average at best....No one can really predict someone to be as bad as he has been but you can predict him to be subpar and many did.

[RShack]

Long story short, we know the theoretical limits on any projection system's accuracy for an offensive stat like OBP or OBA is an r in the .72-.73 (for 550 PAs) range:

Right. And .72-.73 is a generally a lousy predictor. The fact that it's the best we can do, and is less bad than others, doesn't change the fact that it's lousy.

[osfan83]

And there are more bad predictions than good when using stats as the sole or primary basis.

I would love for all these stats gurus to predict what Aubrey Huff and Melvin Mora will put up stats wise next year?

I would also like to know how many of them predicted Baez would be so horrible this season?

I predict Huff will have between 15-30 Hr's next year.

I predict Mora will hit between .250 and .290 next year.

If you want we can give the likleyhood of Huff hitting 15-18, 19-22, 22-25, 25-30 and 30+ HR's.

You guys are looking for guarentees, and there arn't any...but there are odds. I can't guerantee you the O's won't make the playoffs this year, they are still mathmatically alive, but I can give you the odds of it happening.

ST might be a Cy Young award winner next year, but the odds point to him being one of the worst 20 starting pitchers in the league next year.

[Baltimoron]

We have imperfect memories and are biased by stuff like primacy and recency. Stats are actual data. Think of stats as a proxy for a perfect memory.

Without any stats, how accurately could you assess the offensive performance of the Orioles this year? Even assuming you watched every game, without recorded data, how good do you think your estimates would be of who did what?

Stats aren't the problem. They are just data. Its when this data is misused or misinterpreted or mis-cited that the problem arises.