Jon Cryer????

Seriously??

I thought he was hiding out.

Ehh - I said that Voros came across like a total jerk in the BA roundtable, and he did. I didn't call him thoughtless or unreasonable.

*Note - this comment used to say that I had no recollection of ever writing about Voros' personality, because I hadn't RTFA yet. I'd forgotten all about the post he links to.

David, Voros links to a page where you definitely portray him in a negative light. He may or may not be right, but you'll need a different defense.

I like Voros' preemptive strike -- "If I admit over and over that I'm arrogant, then I can go on to be as arrogant as I want." I'll have to remember to use that rhetorical device.

While I share the skepticism Voros has for being able to tell too much from just watching a couple of games, I take issue with this paragraph:

I didn’t see the Fall League game in question, but I’ve seen Fall League games before and they’re little more than glorified batting practices. Only two groups of people are at the games: scouts (or people doing the same thing scouts are doing) and people looking for autographs for future stars they can sell at card shows.

I watched a game last night where Greg Smith was snapping off curveballs and sliders, perfectly spotting a 90 MPH fastball on the corners, and changing speeds like a magician. 6 shutout innings of 2 hit ball. Then Max Scherzer came in and was throwing 94-97 MPH fastballs, mixed in with a changeup in the low 70's. Glorified batting practice my ass.

Now of course the pitching depth isn't always great in the AZFL, and the parks and the environment are not conducive to low scoring games, so offense gets inflated....(not nearly as much this year by the way), but there are still an awful lot of good pitchers coming through there throwing their best stuff.

Also.....the "type" or groups of people that are there has nothing to do with what kind of pitching the batters are facing.

Ehh - I said that Voros came across like a total jerk in the BA roundtable, and he did.

I just read that article. I think you are reading something into it that is not there. If you care to provide and example, fine. But most of what he said seemed reasonable.

I’m sure David Cameron will love this since he knows how thoughtless and unreasonable I am, despite having never met me.

I think Mr. Cameron is saying that you came off as less than pleasant in that one meeting. Having met you before the meeting is irrelevant regarding that one meeting.

But the reality is I remain highly skeptical of the ability of anyone to be able to make long term wide sweeping conclusions about a professional baseball player from 7 or 8 at bats.

"Long term wide sweeping conclusions" seems a bit dramatic of KL's position based on his scouting experience. JC's summary of KL's opinion is, "You once compared him to Grady Sizemore, but after seeing him in the Arizona Fall League you were a bit more pessimistic." A "bit more pessimistic" sounds like KL's opinion of Schafer was slightly lowered rather than fundamentally changed as "long term wide sweeping conclusions" would suggest. KL then said, "I saw him in the Arizona Fall League and was disappointed at how short he fell of the hype. The ball comes off his bat well, but the part of his swing leading from his set point to contact isn’t consistent, and it gets long because he loads so deep." So his opinion changed a bit because KL's assessment of Schafer's mechanics. Even if it's based on 8 ABs, I don't think that's unreasonable. You just treat 8 ABs with the caveat that's its only 8 ABs. Scouting experience, while not quantitative data, is still usable data and I don't have a problem using that as evidentiary source to form an opinion. If KL's opinion of Schafer went from top-notch prospect to dud based on 8 ABs of flawed mechanics, then sure, that's unreasonable. But that's not what happened at all.

Glorified batting practice my ass.

It may be glorified batting practice in the sense that hitters may be experimenting with their mechanics in ways that they're afraid to do during the regular season.

I have gotten the Jon Cryer thing since I was about 12. I don't see it personally, but it's definitely the celebrity to whom I'm most often compared.

I thought Voros really missed the point in about a dozen different ways here, but let me just pick out one bit of silliness here:

If you really could evaluate players based on 8 at bats, what the hell is the point of having four levels of organized minors playing 140 games a season?

Do we even need to ask this question? Players develop. They improve, and they only improve through repetitions - that is, experience in games. The minors are there to develop players, not to provide statistics for evaluations and studies. Those are nice byproducts, but they are hardly the raison d'être of the minors.

It's kinda hard to develop a player without evaluating him at the same time.

Tools. Tools. Tools. You definitely can see bat speed. You can see approach (by and large). 8 AB's may mean nothing statistically, but you can see qualitative measures of skill. You may not be able to make 100% accurate scouting readings, but if a guy truly lacks bat speed or has a terrible approach for 8 AB's, you can make some assumptions that will for the most part hold true.

If you really could evaluate players based on 8 at bats, what the hell is the point of having four levels of organized minors playing 140 games a season?

Gee, I always thought that the point was to have these kids develop and learn. Sure, 8 games doesn't seem very useful to measure someone's future. It still may be a decent snapshot of where they are right now, however.

Dammit. Because of this headline I have the Waitresses "Thinking About Sex Again" stuck in my head which is 1) not a good song, and 2) not a concept I really wanna associate with any kind of Voros.

It may be glorified batting practice in the sense that hitters may be experimenting with their mechanics in ways that they're afraid to do during the regular season.

Nah....thats just rhetoric. You can't dismiss an entire league as glorified batting practice on the basis that the hitters "may" be experimenting. They are trying to hit the damn ball. Sure they are working on things. Just like they do during the regular minor league season.

SPBM,

I'm glad you like it and feel free. For example, you can keep calling yourself a ######### and then you can go on being a ######### for as long as you want.

See how easy it is?

Keith,

My problem Keith is that so often "whether they develop" is usually just a byproduct of his statistical performance. When Chip Ambres becomes a bust, it's because he's not hitting like they thought he would. It has virtually nothing to do with his mechanics. Most often the mechanics are used as ad hoc explanations as to why this guy is hitting .250 instead of .300. That he's hitting .250 still seems to be the driving force behind all the evaluations.

My whole problem with scouting is that a geneticist can explain to me what he's doing and give me references and citations of the previous work that informed his own. I may be too ignorant to understand any of it, but if I wasn't I could do it. Too much of what passes for scouting knowledge is deemed unexplainable and the sorts of things that you just kind have to take on faith.

If I want to know what it means to "show fear" at the plate, having watched a bazillion games and played many years at a lower level, you should at least be able to explain it to me on some admittedly rudimentary level. I have yet to receive such an explanation that differentiates between the sorts of things good major league hitters do all the time. I'm not missing the point, that is my point: if people can't provide me with good evidence on something, then they shouldn't expect me to believe it just because people who know more than me say so.

David,

"There are thoughtful, reasonable, well spoken guys on both sides of the fence. Unfortunately, Voros McCracken and Gary Hughes are not part of that group."

So that's not calling me thoughtless and unreasonable? Fess up, you weren't there, you have no knowledge of the tone of voice and facial expressions that went on there, and to my recollection this is the very first communication we've ever had. Calling me an arrogant jerk based on such scarce information is pretty damn rude, particularly when you know full well I couldn't publicly respond.

Didn't see Hughes after the discussion, but I did run into Eddie Bane and we had a nice conversation and he didn't seem at all off put by anything I said. The purpose of the discussions was to air out these sorts of things (per Alan Schwarz) and I stated my case. I smiled the whole time and everyone seemed to leave in a good mood. For my part I then got good and drunk shortly thereafter and had dinner with Joe Sheehan, Jonah Keri and some other of the BP guys.

I'm perfectly content with people accusing me of being wrong. But I'm not about to lie and claim to agree with or believe things that I don't. If that makes me a jerk, so be it.

I'm surprised feminine hygiene products don't get past the filter. My bad.

I have gotten the Jon Cryer thing since I was about 12. I don't see it personally, but it's definitely the celebrity to whom I'm most often compared.

This is a fun game. We could have a whole thread stating what celebrities people compare us too. It's much more fun reading Law's posts if you picture Jon Cryer saying them as a speech.

After reading the BA article, I thought Voros acted fine. Hughes came off as a bit insecure.

that is my point: if people can't provide me with good evidence on something, then they shouldn't expect me to believe it just because people who know more than me say so.

Out of curiosity, if social phenomena X is unquantifiable, can it be counted as evidence? Or can social phenomena X be recorded in a way that can be used as evidence?

My problem Keith is that so often "whether they develop" is usually just a byproduct of his statistical performance. When Chip Ambres becomes a bust, it's because he's not hitting like they thought he would. It has virtually nothing to do with his mechanics. Most often the mechanics are used as ad hoc explanations as to why this guy is hitting .250 instead of .300. That he's hitting .250 still seems to be the driving force behind all the evaluations.

I don't know what Keith said, but if he saw that the guy had a slow bat or a terrible approach, that's not some mystical scouting lingo of "showing fear". If you look at a guy like Kouzmanoff, every scout had him pegged as grossly overachieving. I doubt it took more than 8 AB's to reach at least a tenable conclusion of this despite his ridiculous minor league line last year. While I give you that a scouting read can attempt to confirm previously held beliefs based on the stats, broadly painting that "it's 8 AB's, he couldn't have seen anything" is straight ignorant.

I'm not talking about unquantifiable, I just want evidence. Something like "Love" is fairly unquantifiable, but the mechanisms behind that concept are becoming more and more well known by psychologists and geneticists by the day.

My problem with scouting is that there seems to be an attitude that the mechanisms and theories behind much of it are beyond investigation. That by merely suggesting that some of it may be right on the money and some of it may be way off, we're therefore torpedoing the whole industry. Shouldn't these things be tested and evaluated?

I've been told by a few people that I look like Brendan Fraser, and also told by a few more that I look like Roger Clemens. Obviously, they are not very similar. If I had to pick between the two, I'd say I do have a resemblance to Clemens.

At least you guys aren't occasionally compared to Christopher Hitchens.

If you look at a guy like Kouzmanoff, every scout had him pegged as grossly overachieving. I doubt it took more than 8 AB's to reach at least a tenable conclusion of this despite his ridiculous minor league line last year.

You don't need a scout to tell you Kouz overachieved in 2006. All you need are multiyear records and knowledge of how to translate minor league stats. I just looked up my projection on him from last January: 272/329/451. Actual line: 275/329/457.

Damn. That might be my best projection.

What do you mean "terrible approach?" Did he stagger drunk to home plate? Did he scream like a girl when an inside pitch was thrown? Did he forget to lower the landing gear? I don't know what that means, and every time I ask people get mad at me.

That's my point? Show, don't tell. OBP is an important stat not because smart stat geeks said so, it's an important stat because it's painfully easy to demonstrate it's value in helping a team to score runs. My say so doesn't enter into it.

I'd like to know what constitutes a terrible approach, and how these sorts of things have been tested and verified. What is it about a "terrible approach" that means a guy can hit anywhere in the world except Major League baseball? If Kouzmanoff hits .330 next year (and that's possible), can we then just say he's now "improved his approach?"

If I walked into an MLB front office and suggested they used a double blind study to evaluate their scouts using a collection of Japanese professionals and semi-professionals of varying quality, I'd get my ass kicked. My problem with scouting is that scouts tend to view such things as insulting, and I'd view them as useful.

Didn't see Hughes after the discussion, but I did run into Eddie Bane and we had a nice conversation and he didn't seem at all off put by anything I said. The purpose of the discussions was to air out these sorts of things (per Alan Schwarz) and I stated my case. I smiled the whole time and everyone seemed to leave in a good mood. For my part I then got good and drunk shortly thereafter and had dinner with Joe Sheehan, Jonah Keri and some other of the BP guys.

Could someone post a link to this Stats vs. Scouts roundtable?

27 - Well-said, Voros. I'm not sure why all these guys aren't diggin' your scene.

Also, I saw the Furs in 1985 at SPAC. Opening were the Blow Monkeys.

Stats vs. Scouts roundtable

Also....Pedroia.

/runs

I'm not talking about unquantifiable, I just want evidence. Something like "Love" is fairly unquantifiable, but the mechanisms behind that concept are becoming more and more well known by psychologists and geneticists by the day.

Can something like "love" be made quantifiable then? For example, perhaps you can create certain measures of "love": the amount of time spent with person X, heart rate changes, the number of love letters written to person X. But I don't know if these quantitative measures is really the best way to measure "love" (nor is it completely invalid). Or with "fear at the plate", perhaps a scout could create certain quantifiable measures, I don't know what they would be, maybe the level of head movement during the AB, but whatever these measures are, I'm not sure that's best way to go about studying it. I guess I kind of see scouting notes as something similar to an anthropologist's field notes. There's valuable evidence in there but it's not quantifiable or replicable. And there are certain things you can do to improve the quality of those notes (e.g. more time spent with the subject is better, or in our case, more ABs is better) and thus enhance its usage of evidence. I just hope people don't reduce evidence as something that can only be expressed numerically. Luckily in the case of baseball, the investigator's(or scout) evidence can checked for validity through video. And if each person agrees with the investigator's assessment, I think that enhances the validity of the claim. There be certain indicators to help person X search for whatever you're studying (e.g. urban decay to fear at the plate) but those things may not be quantifiable.

Oi. Rough crowd. Some woman at work yesterday said I looked just like Jesse James. Is this a good thing?

I do believe there will exist until time eternal a rift between the "numbers" crowd and the "scout" crowd. Some people prefer things to be broken down and explained and shown to them in little parts. Some people can look at something in a whole and, having seen it before, dismiss it or fall in love with it. As someone who both loves to tear things apart and put them back together and is a romantic I can understand how both camps feel. I was estatic to find out about people like Bill James. I thought, 'There are others?? Others who want answers? Others who don't view player analysis as baseball creationism??' But a big part of me just enjoys soaking in watching players play and the flags gently blow on a summer night at the ballpark without trying to figure out what Jeff Weaver's FIP is now.

That is all.

This is getting ridiculous. You don't know what a terrible approach is??? I really am a stats guy at heart so this does hurt me and I do hate so stoop to this level, but honestly if you can't look at a guy 10-15 times and tell what his general plan is at the plate (let alone whether his bat speed is good or not which is generally the primary opinion of concern here) and what that general plan typically leads to...I don't know what to say.

There's really no point in continuing this discussion if you can't give scouts anything so I'm just going to stop now before I end up going to bed at 5 AM after 25 posts of inconsequential substance (though I will if you really want me to).

I don't care that Voros has an ego.

I do care that the statistical methodology ("Correlation between sequential time series points") behind his biggest achievement ("Pitchers don't control BABIP") is spectacularly, unreasonably and ridiculously unsound, and then the results of this flawed methodology were then massively overstated.

That's not arrogant. It's just wrong.

"But a big part of me just enjoys soaking in watching players play and the flags gently blow on a summer night at the ballpark without trying to figure out what Jeff Weaver's FIP is now."

So do I, which is why I don't like Fall League games very much. I prefer actual cheering at baseball games, whether little league, big league or all points in between. Fall league games are baseball's version of a colonoscopy. Players are only human and it's awfully hard to give a damn who wins the game when no one else in the park does including the folks in control of your career.

Want to have some fun, go sit behind home plate at a fall league game with a high powered magnetin a backpack. Between the palm pilots, laptops and the radar guns, you could cause untold hav-ock.

Even spring training has people cheering.

"That's not arrogant. It's just wrong."

So is your belief that such was the only study I did or your general beliefs as to what it is I've said. But hell, knock yourself out sparky.

I know what his plan is at the plate. His plan is to hit the damn ball, preferrably hard, and if he's any good, his plan is to avoid swinging at pitches out of the zone.

That goes for Nick Punto as much as it goes for Albert Pujols. But the reality is that Ryne Sandberg is in the Hall of Fame and his reaction to breaking pitches throughout his career often bordered on comical.

Furthermore, I notice that you didn't actually explain what a "bad approach" is, you just expressed shock that I didn't know. Like I'm a four year old, what's a "bad approach?" Is it swinging at bad pitches and taking good ones? Why not just say that? And isn't that the sort of thing easily quantifiable without scouts? And if it's not that, what is it? Is it going to the plate wondering what you're going to have for dinner? I'm betting sometimes that helps. I once homered during at bat in which I periodically stepped out to check and see if the 1st base coach had noticed the practical joke we played on him. When I pitched all I could think about was the next pitch and couldn't find the strike zone with a Sherpa guide. Obsessing about what I was going to do seemed to be a lousy approach to my pitching. There's no way I could find the link, but I do remember a fairly accomplished Major League Hitter saying he did very little thinking up there: it was mostly see ball, hit ball.

So I honestly don't know what you mean by "bad approach." OBP = (H+BB+HBP)/(AB+BB+HBP+SF). Water = H2O. Confirmation Bias = the tendency for human beings to see and remember those pockets of information that match their own beliefs and to discard and forget pieces of information they don't.

Why is a definition such a hard thing to come by for this stuff? Is scouting that much more complicated than everything else?

There's really no point in continuing this discussion if you can't give scouts anything so I'm just going to stop now before I end up going to bed at 5 AM after 25 posts of inconsequential substance (though I will if you really want me to).

I think he's willing to give scout's something as long they give quantifiable parameters. If it's not quantifiable, it's telling but not showing.

Want to have some fun, go sit behind home plate at a fall league game with a high powered magnetin a backpack. Between the palm pilots, laptops and the radar guns, you could cause untold hav-ock.

That sounds an awful like 'screwing with the man'. Count me in.

People have compared me to Matt Damon and Tom Brady. It angers my friends greatly. It angers me greatly that it's not true.

And this seems like the perfect thread to ask what happened to "Jack who resembles Voros"?

Someone who met me at the winter meetings called me Santa Claus. I've been working hard to try and fix that with some success but a long way to go.

I get Jim from the Office from time to time.

John Malkovich

I get compared to Hugh Grant all the time by women (and my current girlfriend as well). Hair, eyes, face, rakish smile, inexplicable 'awww-shucks' British demeanor, you name it.

At first I was hideously upset about this - I am sui generis!, I would proclaim - until I realized that all the women making the comparison meant it as the most sort of softbatch "I have a crush on you" compliment imaginable. Then I began to play it up madly. BWAHAHAHAHAHAHA. From awkward muso/baseball dork to ladykiller in one easy, utterly inadvertent step.

33 - I have no idea what a bad approach is.

35 - If I pretend that Voros' original DIPS work was perfect and you pretend it didn't exist, my ability to forecast pitcher performance will greatly exceed yours. I think that makes it a valuable and important study even if some of the conclusions were overstated.

I used to look kind of like the guy who played Miles on Murphy Brown. Now, I'm fatter.

Trying to veer this back onto point a bit...

While it's undeniable that certain things may be obvious from a small sample - like bat speed or velocity (although even these I'm skeptical of) - it's completely unclear the degree to which scouts, or anyone, has a good idea as to the importance of these factors, or how they play into other less obvious skills.

I'm not a scouting doubter in terms of projection, but in terms of stuff like analyzing existing skill sets? I've yet to see a lot of evidence that there is a significant level of accuracy there. It may be there, but I haven't seen it.

I don't have a ton to add, but this has been one of the most fascinating discussion threads I've seen in awhile, and Voros' points are at least worth considering. Personally, I would love to see a major league team--one like the Pirates comes to mind--try an experiment where they get rid of their entire scouting department and replace it with three college students making $15 an hour armed with Baseball America and related books/magazines. See whether the team's drafting/trading is really horrendously worse than everybody else's.

You might see some slippage, but then, you might not. I doubt it would be a total disaster.

I get mistaken for both Billy Dee Williams and Meg Ryan, post cosmetic surgery. Go figure.

Also, I'm not sure I get the reisistance to what Voros is trying to say here. Right now we have stats and we scouting. He's just pointing out that there's no reason the two can blend together and is suggesting ways it might be done. The use of video, bio-mechanical models, etc. etc. are available and it's not hard to see how they can be incorporated into the evaluation of baseball players. We all admired Carlos Gomez' work here and it seems he's working towards the same goal except instead of coming from the stat side of the fence, he's moving towards it from the scouting side of the fence. This is the undiscovered country for us and I'm not sure why we're so resistant to exploring it. The statheads supplementing the seamheads yet remaining intrinsically separate dichotomy is dinosaur thinking.

one of the most fascinating discussion threads I've seen in awhile, and Voros' points are at least worth considering. Personally, I would love to see a major league team--one like the Pirates comes to mind--try an experiment where they get rid of their entire scouting department and replace it with three college students making $15 an hour armed with Baseball America and related books/magazines.

This seems to be an overstatement of what Voros is recommending (again). What Voros is saying is that scouts refuse to try to quantify and then VERIFY their analytical methods. Qualitative research is done all the time. What Voros is arguing against is the complete reluctance to VALIDATE the research. You can validate quantitative methods, you can validate qualitative methods. But what certain things are essential in research (and let's be honest, that's what player prediction is, no matter what methods you use) : validity, transparency, reproducibility, reliability. I don't think that Voros is saying that scouts couldn't be organized in such a way that we could look at what they're doing. He's saying that they're irrationally defensive to prevent such a thing from ever happening AND that they irrationally criticize quant-only methods with the extremely weak (and completely condescending) "I just know" defense, which is just not admissable in any reasonable discussion about methods.


And BvA's suggestion isn't exactly replacing scouts, it's using a mixture of experts model where you're using the information from EVERYONE'S scouts (BA's and other books/magazines) to make decisions. And I would be shocked if everyone isn't already doing this to some sort of informal extent already (doesn't someone on the Interweb already have a meta-list of prospect rankings based on everyone else's rankings every year?). You'd have to be functionally retarded as a GM to not have some sort of meta-view of every single prospect ranking when evaluating your own scouts' reports. Teams should have their own scouts because they want MORE detailed information on certain players that can't be obtained just from a coarse overview gleaned from the mixture of experts.

Very nicely said, Russ. Yeah, I know I was sort of changing the subject. I would just really like to see that experiment done, because I'm curious about how much you really get for the tens of millions of dollars teams presently spend on their own scouting departments, as opposed to spending like $50,000 on three twentysomethings and a few books/magazines.

In other words, I personally agree with Voros on the general point that the structure of modern scouting departments is somewhat broken.

Doing it with the Pirates would skew the results. It's better to do it with both the Pirates and the Braves. Then we'll see.

It could be worse.

When my hair was shorter and I lived in the SF Bay Area, people told me I looked like Rich Reuschel.

This seems to be an overstatement of what Voros is recommending (again). What Voros is saying is that scouts refuse to try to quantify and then VERIFY their analytical methods. Qualitative research is done all the time. What Voros is arguing against is the complete reluctance to VALIDATE the research.

Eddie Epstein pointed out -- it must be 15 years ago now -- that teams didn't even track how scouts did. He wasn't talking about how they did what they did -- but simply how they did. How successful were they?

It wasn't that there was no accountability such that scouts that did a bad job could keep their jobs, like a Pirates GM; it was that nobody even knew who the good and bad scouts were. (Or, at least, that "good" and "bad" were defined by something other than the correlation between their evaluations of players and the performance of players.)

I wonder if that's changed at all.

And this seems like the perfect thread to ask what happened to "Jack who resembles Voros"?

His friend Weekly Journalist still hangs around. Last I heard, Jack was sweating out the NBA finals. He's a Cavs fan.

Are we still doing the celebrity thing? Some people tell me I look like the scientist in "short circuit," others tell me I look like the robot.

Are we still doing the celebrity thing? Some people tell me I look like the scientist in "short circuit," others tell me I look like the robot.

I call BS on this. Nobody will admit to having seen Short Circuit. You sir, are a dirty liar!

I have nothing to contribute of substance, but I did want to say that this has been my favorite thread here in a long, long time. I tend to sympathize with Voros' frustration. Even when doing qualitative analysis, accountability is absolutely essential. I have no idea what the state of scouting is in baseball, but feel comfortable saying that if there aren't methods for verification and accountability, then there is much room for improvement.

Is scouting that much more complicated than everything else?

Ask a music producer or songwriter, "Sit down right now and write a hit." They can't do it. But write hits they do. Some things are unquantifiable. Is the "ability" to write a hit indistinguishable from luck? Maybe, or maybe for the most part. Lotta songs get written, there are lots of one-hit guys, no-hit guys like stars in the sky.

I get compared to Hugh Grant all the time by women

Me, too. "Hugh Grant he ain't," they all say.

I have gone from being told I look like a less muscular Tom Selleck in Magnum P.I. (when I had the big mustache) to a not unreasonable facsimile of Higgins...if he had grey hair.

To the point, I'd be surprised if one of the more stat-friendly front offices isn't doing something to codify scouts' observations. The real problem for the teams, it seems to me, isn't whether Voros or I are convinced that the observations are verifiable or valid, but whether the front office can get scouts to learn from one another. In many other careers, from stockbroking to Amish furniture-making, much of the knowledge is tacit and best learned through experience with access to multiple examples (mentors or templates). I imagine the same is true in scouting for any sport. But there's a real selection problem in that teams may not be able to figure out who really knows how to do the equivalent of sound ethnographic research and who just has the personality to sit alone outside for hours looking at other people do stuff. Don't get me wrong, I'm an inveterate loner, and would love the job. But that doesn't mean I would do it as well as someone who resents all the hours on the road, but actually learned the trade from a great mentor, or even someone who just has an amazing eye for detail and concentration. There are tradeoffs between getting people who can do the job well, and getting people who want the job.

Also, it's probably a lot like outside sales--nobody wants to talk too much about how to really get the job done because some of it is stuff the front office would rather not know. You put a sales guy (or gal) in charge of the other sales guys because he "gets it," both the need to depart from SOP sometimes and the emotions of the lifestyle. That doesn't mean it can't be made more organized and effective. But you're a heckuva lot more likely to spread good sales ideas through the company by bringing all the salespeople to a fun conference and putting them in the same room with an open bar than you are by having the accountants interview them and give them an electric shock every they say "I just know, that's all."

60 - We can't quantify how they write a hit, but we sure as !@#$% can quantify whether the songs they write are hits or not.

Jim McLaughlin used to ber a quantitatively oriented scouting director, but that was before I was born. I think that he had some lingering influence over the Orioles scouting department through his hires, but that ended years ago. There was a good chapter on him in Dollar Sign on the Muscle.

FWIW, I also read Prophet of the Sandlots recently about Tony Lucadello of the Cubs and later of the Phillies. Interesting book, if you ever get a chance to read it. He was a very successful scout, but I'm not sure that all hist theories made sense.

If I walked into an MLB front office and suggested they used a double blind study to evaluate their scouts using a collection of Japanese professionals and semi-professionals of varying quality, I'd get my ass kicked. My problem with scouting is that scouts tend to view such things as insulting, and I'd view them as useful.

This is a fantastic idea. Wouldn't it be even better to take video of actual MLB players, and use CGI to obscure their appearance (make them a little bit like generic video game players, all 6'2" 200 lbs, showing only the swing/motion difference) and then see if scouts could tell good from bad?

That can be carried too far, too. You can find two guys who are both 6'2" 200, and watch each of them bat ten times, and have both of them go 0 for 10, and if one has the talent to play professional baseball and one does not, any fool is going to be able to tell the difference.

By the same token, when you have two guys who are both talented enough to play professional baseball, individual scouts' opinions can vary greatly on which one is more likely to be a successful major leaguer. That's where I don't know whether or not there's any coherent system for accountability for scouts.

Are we still doing the celebrity thing? Some people tell me I look like the scientist in "short circuit," others tell me I look like the robot.

Twenty years (&, alas, well over 100 lbs) ago, I looked exactly like the next-from-last band member on the jacket of the Poison Girls' Where's the Pleasure. C'mon, you know you all have a copy. (Well, OK, maybe Repoz. Maybe.)

I've heard John Gruden for me in the past.

I look like Tom Green. From the side, especially. From the front, you can see the resemblance but it's not as striking.

I've been compared to Christopher Eigeman, Richard Benjamin and Ron Livingston.

I look like Jackie Chan!

Actually I don't, but not like you F!@3ers can tell.

I am constantly compared to Kevin Smith. Sigh.

Actually I don't, but not like you F!@3ers can tell.

On the Internet, nobody know you're a Pekinese.

Russell Crowe

I don't know for certain, but it does seems like teams rely on scouting information a great deal when making trades and promoting players internally. I don't think it's accurate to say that nobody in baseball acts like you can evaluate a player based on watching 8 ABs. It's the fact that they do that is the problem.

I wonder if that's changed at all.

Probably for some teams.

I know one team that a few years ago tried to create a uniform scouting form to have all their scouts use - so the team could take the form and have some intern intern them all into a spreadsheet. The scouts essentially rebelled, some refused to use the form, those who did made up their own personal nomenclature etc. As an exercise it was useless, as data collection it was useless.

I have no doubt that most scouts are aware at some level, that they could be "graded", someone could look at all their reports over a period of time- you though this guy was an A prospect, he was a bust, you thought this guy was a D- he's starting for someone else now... And the thought terrifies them, the average scout, like the average worker in every profession thinks he's above average, being graded by someone else, especially by someone the scout looks down on (a college geek with a calculator) is both insulting and terrifying.

I don't think it's accurate to say that nobody in baseball acts like you can evaluate a player based on watching 8 ABs. It's the fact that they do that is the problem.

What always amazes me, is that whenever you mention that some teams do something really stupid, like base personal decisions on 20 ABs in spring training, or see someone start the year 9 for 50 (.180) say they'll give him a week to see if he can turn it around, player goes 6/20 (.300) in that week, but he's still hitting 15/70 (.214) on the year, so the team will bench the guy saying, "we gave him time to turn it around but he never started hitting)- someone will respond and say that no teams do that, because people running MLB teams are professionals and no one does something that stupid.

You see really stupid stuff like this EVERY YEAR.
Every year you will see teams break spring training having handed a roster spot (and sometimes a starting job) to someone who has no business being in the majors, while someone else is either exiled to the PCL or released. You will see someone lose his job or role due to a poor start EVEN after said player started playing better, never to really regain the role even if his team has no better options- it happened to Matt Murton for instance.

I'm a Met fan, in 2005 I saw them inexplicably break camp with Manny Aybar and Mike Matthews and Felix Heredia occupying critical roles in the pen. Why? Because Willie/Omar liked the way they THREW in ST.

Why does stupid stuff like this happen?, because there are baseball people who really believe they can evaluate someone based upon 8 at bats or 3 spring training innings- and given the opportunity they will act on those beliefs

First, the important thing. People use to always say that I look, and sound, like Vince Vaughn. It used to happen almost every time I met someone new but it hasn't happened in a while.

Second, you guys are giving Voros way too much credit. He's not saying that scouts need to be held accountable. He's saying that the only thing that counts in accountability is quantitative measurement. Those are completely different views.

Third, consider this analogy that Voros made in defense of himself:

I'm not talking about unquantifiable, I just want evidence. Something like "Love" is fairly unquantifiable, but the mechanisms behind that concept are becoming more and more well known by psychologists and geneticists by the day.

Only a technocrat could say something this wrong. The debate here isn't between scouts and stats it's between philosophy and "empirical" science. Voros the "scientist" thinks that we are close to discovering the truth about love every day the more we learn about genetics. If it weren't true that many nihilists in scientist clothing didn't actually believe this nonsense. Bill James was a great writer, and a great thinker about baseball, because he was a philosopher first, and a scientist second.

I'll put things in an equation to make things intelligible to the illiterati. Science - Philosophy = DIPS = Profound Error

My whole problem with scouting is that a geneticist can explain to me what he's doing and give me references and citations of the previous work that informed his own. I may be too ignorant to understand any of it, but if I wasn't I could do it. Too much of what passes for scouting knowledge is deemed unexplainable and the sorts of things that you just kind have to take on faith.

This to may sets the bar too high---it assumes there's nothing we cannot quantify that has value.

I also agree generally with Voros' skepticism about scouts/scouting analysis as a whole. But I do not think that means any particular scout is without value. It is just that across hundreds of scouts there are huge variations in quality, ability, etc. So, when someone cites an unnamed scout that has essentially no value---but there are absolutely scouts out there (imo) and scouting opinions which DO have value.

Ugh, I get "Rob Schneider" a lot. I got Dean Cain once, but I think she didn't have her glasses on.

Robert Downey Jr. Usually by middle aged women.

Daid beat me in with the Epstein reference. Still worth reading though:

[...] an Oriole scout watched Ron Gant and David Justice play at Triple-A Richmond in 1989. In his reports, he said that *neither one* (emphasis his) was a major league prospect. [...] In most industries, a judgement error of that magnitude is grounds for dismissal. This scout was not dismissed - in fact, it was never even considered - because he was a friend of the general manager.

Many scouts have jobs for no other reason than crontism. [...]

One point I am trying to make is that scouts are not mystic seers of baseball truth. The scouting profession is like most other professions. Some are very good at what they do, some are awful, and most are somewhere in between.

And it's worth noting that Epstein also mentioned one scout (identified only as the father of a current major leaguer -- this was written in 1995) whose judgment as to who was or was not ready for any given level was consistently better than any methodology that Epstein could devise.

I have little doubt that a scout who really knows what he's doing -- even if he can't really explain what he sees in a player -- brings great value to an organization.

On the other hand, There was a joke going around the scouting community when Bill Wight signed Joe Morgan. Morgan was the second player Wight signed. The first was Walt "No Neck" Williams, 5'6" 180 pounds. "First Williams and now a midget." They're *laughing* at a guy whose first two signings make the majors and one of whom is an inner circle HOFer.

So much of scouting is dealing with images (read Dollar Sign on the Muscle for instance). Generally speaking you need to be exceptional to even get a look if you don't fit the conventional look.

And even then you can get the Washington Senators concluding that they have no need for Joe Morgan. Craig Wright includes summaries of more than a decade's scouting reports on Morgan.

After winning two straight MVPs the Ranger scouting report on Morgan includes the gem, "Somewhat limited because of small body".

As late as 1970 the Senators scouting reports were dismissive of Morgan. Swear to God they were saying that they didn't need him. This from the team whose 2B job was in the hands of a Tim Cullen/Bernie Allen platoon.

Though my favorite example of scouting and images is in the scouting report on Harold Baines. The scout matter of factly
says he graded Baines down because he didn't like Baines' hair. (If you can believe it, Baines wore his hair in corn rows)

I was mistaken at an sf convention for the horror/fantasy artist JK Potter. Not sure if that's a good thing or not, since I've never seen a photo of him.

Only a technocrat could say something this wrong.

Ever read the Skeptical Inquirer? There are more people out there that believe this than you think.

Though my favorite example of scouting and images is in the scouting report on Harold Baines. The scout matter of factly
says he graded Baines down because he didn't like Baines' hair. (If you can believe it, Baines wore his hair in corn rows)

I don't know if scouts still do that, but writers in New York were down on Milledge because of his hair.

They believe that his hair said something about his personality/intestinal fortitude etc. I have no doubt that many old white guys scouts in the 70s and 80s upon seeing an African American (or god forbid a white boy) with a distinctive clothing/hairstyle that was culturally alien to such scout- would automatically assume something terrible about the player's personality or work ethic.

A hockey fan ex-girlfriend of mine used to say I look like Pat LaFontaine.

79 - My wife is Filipina, I'm Jewish. When we first learned that she was pregnant, she searched the internet to see if there were any Filipino/Jewish celebrities, to see what our kids might look like. She only found one. Thankfully neither of our daughters looks anything like Rob Schneider!

For what its worth, most other professions don't do the kind of analysis we're debating here either.

Daid beat me in with the Epstein reference. Still worth reading though:

[...] an Oriole scout watched Ron Gant and David Justice play at Triple-A Richmond in 1989. In his reports, he said that *neither one* (emphasis his) was a major league prospect.

Ron, how is this assessment even possible? 1989 was Gant's awful season in the wilderness, but in '88 he had hit 19 homers as a big league rookie.

(I could see a scout saying something like that about Justice during his 1988 stint at Richmond; he was pretty much overmatched starting the season there.)

Actually I don't, but not like you F!@3ers can tell.

You're of Chinese extraction, right? I'm sure if we actually met you, we'd say you look more than enough like Jackie Chan.

I was mistaken at an sf convention for the horror/fantasy artist JK Potter.

Isn't she the one who wrote those Hogwart's books?

60 - We can't quantify how they write a hit, but we sure as !@#$% can quantify whether the songs they write are hits or not.

Yup, and we can't tell who's going to be a ballplayer, but we can ID one when he makes it.

You can find two guys who are both 6'2" 200, and watch each of them bat ten times, and have both of them go 0 for 10, and if one has the talent to play professional baseball and one does not, any fool is going to be able to tell the difference.

Dude, that's nonsense. There are plenty of such guys who don't have the talent yet still get signed. And Mike Piazza didn't have any apparent talent at all as far as the scouts were concerned.

Yeah, that's a dead ringer for Jackie Chan alright.

So is your belief that such was the only study I did or your general beliefs as to what it is I've said. But hell, knock yourself out sparky.

Yeah, but now you're telling, not showing. I think that's where the impression of arrogance comes from. He says your statistical methodology is "spectacularly, unreasonably and ridiculously unsound", and you counter by saying, essentially, "Maybe I've done other studies." If they're better than what you've shared to date, it seems only fair to expect that you'd pass that along - given your dissatisfaction with responses that don't enlighten.

That said, I understand where you're coming from: you've shown something behind your conclusions, while scouts generally share none of their supporting work, at least not with us.

I believe that, to do scouting well, one needs to devote a lot of time and potentially $ to the craft. As a result nobody freelances at it well (except CBW for a time), and everything is proprietary. If you don't like that professional scouts won't share the proprietary methods behind their conclusions, do the damned work yourself. Stop complaining about how nobody brings you fish; learn to fish.

What Voros is arguing against is the complete reluctance to VALIDATE the research. You can validate quantitative methods, you can validate qualitative methods. But what certain things are essential in research (and let's be honest, that's what player prediction is, no matter what methods you use) : validity, transparency, reproducibility, reliability.

Absolutely, the only thing I would add is that there's different standards, not lesser standards mind you, for validity, transparency, reproducibility, and reliability with qualitative research. Qual people aren't going to give quant people exactly what they want in regards to these areas. For example, perfect replicability isn't possible with qual research. You either accept that there are different standards or you don't.

"Second, you guys are giving Voros way too much credit. He's not saying that scouts need to be held accountable. He's saying that the only thing that counts in accountability is quantitative measurement."

Yes! Exacty my take as well.

My father, girlfriend, and anyone else who's seen me and an A's game say I look exactly like Dan Haren, but with longer hair. Eh, I'll take it as a compliment.

Yeah, but now you're telling, not showing. I think that's where the impression of arrogance comes from. He says your statistical methodology is "spectacularly, unreasonably and ridiculously unsound", and you counter by saying, essentially, "Maybe I've done other studies." If they're better than what you've shared to date, it seems only fair to expect that you'd pass that along - given your dissatisfaction with responses that don't enlighten.

Except that he has. I think the "maybe I have" was supposed to be snarky since I'm thinking most people around here fully realize that DIPS related research has been published since the initial study yet the detractors either choose to ignore the new stuff or are unwilling to actually look for it. Practically every DIPS debate on this site has almost always boiled down to one side bashing DIPS because of the initial study while the other side keeps trying to point out the new research, rinse and repeat.

Voros said this, this, and this.
Okay he misspoke a bit after more research here is what is going on
Yeah but Voros said this, this, and this.
Yes, yes I know he wasn't totally right on that but here if you look at the data you can see he has a valid point.
But Voros said this, this, and this.
Sigh.

That said, I understand where you're coming from: you've shown something behind your conclusions, while scouts generally share none of their supporting work, at least not with us.

Nor can they fairly be expected to do so - their teams wouldn't be too happy to see internal, confidential reports splashed over the Internet.

Back to an earlier comment:

I'm not talking about unquantifiable, I just want evidence.

On your terms, that is - IOW, YOU get to decide what counts as "evidence", and what doesn't. In my experience, that kind of discussion goes nowhere.

-- MWE

I've shared them. Hell, I've shared them on here: I mean it wasn't like DIPS 2.0 was behind a pay service or anything. You have to remember, most of this stuff was hashed out 6 or 7 years ago on usenet. Similarity studies (which were especially enlightening), studies involving the hits per balls in play rate for non-pitchers. Studies involving career numbers of active starters. Studies from 1901-present. Studies involving only pitchers who switched teams. Studies looking for cross correlations between hits per balls in play and stats like homers, strikeouts and walks. Then the knuckleballer stuff.

I presented all this stuff from 1999-2001, but this was all long before the Red Sox hired me, and so it was scattered about various usenet and message boards. Eventually I was planning on trying to catalog all of it, but it's an ambitious project and I think some of the message boards (like the old Fanhome one) aren't even around anymore. It was all just a hobby that got out of hand for me at the time, I had no idea folks would care enough that I'd need a bibliography seven years later. All this stuff said the same thing, the skill is probably real, but not very large and in the vast majority of the cases not very meaningful. Regardless of what folks think now, that opinion was a major sea change from conventional wisdom in 1999 even among stat geeks.

Now it would be one thing if people simply asked me if it (that I only did one study) was true, but when folks come on here and get snarky thinking they're certain that's all I ever did, well then you'll excuse me if I get snarky in return. Furthermore, from day one I theorized on trick delivery pitchers being exceptions and flyballers and groundballers having different profiles on hits per balls in play. Plus most of that stuff was in the BP article or the followup soon thereafter. People tend not to argue with what I've said, but instead argue with these gross exaggerations of what I've said and needless to say have an easy time knocking down those.

I really don't want to turn this into yet another DIPS thread, but while there's plenty of room to argue on the substance of the various theories, I'll be damned if I let folks accuse me of running some sort of con game when they don't know the first thing about the subject. I was as thorough on the subject as my skills and the available data would allow me to be, and publicly presented everything I found out. Had what I said in November 1999 had the same effect as Neyer mentioning me in 2001 did, this stuff would be easier to find.

That people with more skills and better data came along later to expand what we know on the subject is something I've never once argued with.

"On your terms, that is - IOW, YOU get to decide what counts as "evidence", and what doesn't. In my experience, that kind of discussion goes nowhere."

That's BS. Any actual evidence will do. "Shut up" and "I know more about than you" which are the responses I generally get, don't count. Are you saying they ought to?

I think the "maybe I have" was supposed to be snarky since I'm thinking most people around here fully realize that DIPS related research has been published since the initial study yet the detractors either choose to ignore the new stuff or are unwilling to actually look for it.

Au contraire. Most of the detractors have read the new stuff, too, and realize that the new stuff doesn't address the single most basic issue, that the method by which the problem has been analyzed is wrong.

GWO's comment, with emphasis added:

I do care that the statistical methodology ("Correlation between sequential time series points") behind his biggest achievement ("Pitchers don't control BABIP") is spectacularly, unreasonably and ridiculously unsound

nails it.

-- MWE

Any actual evidence will do.

Well, I think he's saying what you'll accept as evidence is what only meets your terms (i.e. quantifiable evidence). Thus there's no room for discussion. That's fine, you have standards for what counts and you're not budging.

I give up. It's like I'm speaking Klingon.