.
Q. Are MLB talent scouts able to evaluate players based on intuitive judgments?
A. The thought was floated that they aren't. In this article the author implies that the only appropriate tool for making roster decisions is a performance-based analysis.
For example:
.
Every year, though, decisions are made based on how players do in March. The decisions are justified by claiming that it they aren’t based on the results, but on how the players look to experienced coaches and scouts who are paid to evaluate players in an up-close-and-personal atmosphere. The problem is that human beings — even experienced scouts and coaches — are pretty terrible at evaluating the difference between “how a guy looks” and what his results are.
Incidentally an incorrect implication, but the spirit is clear.
.
Q. What makes you think that "experienced scouts and coaches" can evaluate anything based how a guy looks?
A. Oh, just f'r instance, let's consider the amateur draft.
In this Baseball Analysts article, Sky Andrecheck examines the MLB return that franchises have gotten with their draft picks.
He finds a verrrrrrry nice, smooth curve: the higher an amateur player was taken, the better the career he had in the major leagues.
.
Q. Implying what? What's your point?
A. That MLB franchises knew how good that amateurs would turn out to be, up and down the draft, with a remarkable level of accuracy.
.
Q. Implying what? Still don't get your point.
A. Do you think they used a performance analysis paradigm to evaluate anything? Or do you suppose that they evaluated 18-year-olds based on how a guy looked?
If experienced scouts and coaches can't tell anything based on how a guy looks, what is that curve doing there?
.
Q. Oh.
A. Any time we sabermigos get to thinking we know it all, and Eric Wedge knows nothing, we might do well to remember the amateur draft. How large a distance is there between you, gentle reader, and your being able to achieve a curve like the one above, based on watching high school games?
There is a million miles between ME, and me getting a curve like the one above, by watching high school players. I doubt the distance between YOU and there, is much less than that.
.........
Now remember something: the above curve is based on the talents and abilities of fine men who are at the bottom of the MLB employee pool. Jack Zduriencik, and Eric Wedge -- and all GM's and Managers -- are they at the bottom of the MLB employee pool? Or are their skills at all toward the top of the pyramid.
.
Q. One last question. The Mariners did seem to do pretty lousy last March, making decisions off Arizona performance.
A. I think there's a lot of "noise" in their 2-for-7 result last year. :- ) But if you want us to go point-by-point, just let us know.
Supposing the Mariners did do lousy last year, which I don't necessarily suppose, would that mean you'd give up the task? After all, every good organization watches spring training ......
.
Q. What is appeal to authority, by the way?
A. The syllogism thusly:
- Most of what authority A has to say on subject matter S is correct.
- A says P about subject matter S.
- Therefore, P is correct.
If you can find ANY statement in the last few years, locally, with that final line in it, I'll buy you a Good Humor bar.
We are arguing about whether Capt Jack's and Sgt Wedge's intuition should be ONE PIECE OF THE PUZZLE, or whether it shouldn't be.
Nobody says Eric Wedge must be correct because he's the manager. Nobody.
.
Your friend,
Dr D
Comments
Here's what I do.
In terms of weeding out noise...I go game by game and count up the statistics ONLY of players with a legitimate chance of making the big league roster and only when those players are competing against other players who have a legitimate chance of coming home a big leaguer.
To clarify...we see that Justin Smoak went 2 for 3 with a walk, a sharp single to right, a line out to left and a two run homer. Those ABs occurred against:
BB: The starting pitcher - an establishe dbig leaguer
Line Out: The team's primary set-up man
1B: Someone with maybe a 10% chance to make the club
GW HR: An MLB (TM) Mr. Magic Closer
So...in my head, Smoak went 1 for 2 with a walk and an impressive HR.
I've made a habit of doing this...it works better than you;'d think at getting rid of the noise.
I was thinking along the same lines
and I have trouble believing the managers, coaches, and GMs that are watching every play are unduly influenced by stats. The advantage to small samples like this is that it becomes easier to remember hard hit outs and fly balls as well as embarrassing strikeouts and batters being fooled on pitches. Fans rely on stats because we can't see every game, much less the batting practices, the bullpens, and the drills, and we can only guess how a player is feeling physically or mentally. Much as we may not understand from outside, all those things contribute to decision making processes, as well they should.
And from the last spring training we have:
Saunders wins vs. Wells, good call
Kawasaki wins vs. Rodriguez, my guess is the Mariners didn't believe in Rodriguez at Short
Liddi wins vs. Peguero, push
Beavan wins vs. Ramirez, fail, but tucking Ramirez in the bullpen was a hedge that didn't work when hr was injured, it should be pointed out that Beavan didn't exactly implode.
Noesi wins vs. Iwakuma, again a fail, and again, the Mariners tucked him into the bullpen and had him work on strengthening his arm according to Baker.
Kawasaki came out of ST stuck deeeeeep on the bench
He hit .455 (or whatever) in spring and Wedge shocked us all by announcing, loud and clear, that he was a utility player who wouldn't be used much.
Very odd for other blogs to hold that one up as a case-in-point. He was precisely an example of the club SEEING THROUGH spring training stats.
You and Baker are riding in the same cab there Matty
He routinely gives you the report as to whether a good game occurred against premium competition and, if so, whether the pitcher was just out there working on something.
It's a great point. And, yes, you still might wind up with only an 8-for-17 batting line against premium pitching but ... suppose it were 0-for-17? Would people hesitate to bust Smoak's (or Figgins', or Peguero's) chops for a terrible result?
................
Pitcher J: 3 strikeouts, 10 walks in 20 innings
Pitcher J2: 21 strikeouts, 2 walks in 20 innings
One of those is, on March 27th, Jon Garland and one is Jeremy Bonderman. The Enlightened GM wouldn't even want to know which had been which?
Exactly
The same was something I noticed about Bays HR off Tyson Ross in the 1st. In general the Mariners have been scoring early which means that our most likely MLBers were scoring against their most likely. Nobody can convince me that this is not positive.
I dont appeal that completely. More like since A said P, P may be true. The Zunino quote early this morning is a pretty good example where since Zunino said that Maurer has a good changeup I explore the possible implications of him saying so. I've said the same recently about pressure in certain lineup spots and veteran leadership. Outside of that appeal in that situation, since the players and coaches believe in those things I believe there's a possibility in their effect. In that case you are talking about faith in abilities which is a key to life. If the players themselves have more faith in their success while wearing tie dyed armbands, give them the armbands. They will produce better if they have faith that they will.
I read
I read the article in question on the Big Blog and left with similar issues with it. ST stats are, in aggregate, less than optimal. Still, the author uses Iwakuma as an example of ST performance bias. Baker (who is with the team, coaches, trainers and players every single day) has indicated that in ST and thorugh the first half of the season, Iwakuma literally took 3+ days to recover from throwing a couple of innings. It wasn't his ST stats that kept him out of the starting rotation - it was the fact that he wasn't ready for the SP grind. I think it's to the club's credit that they kept him around until he was.
Too, they don't get credit for the correct decisions. Catricala had a very good ST and the M's had no incumbent (beyond Figgins) at 3B. But they made the call to send him to AAA - a decision that his performance in 2012 says was the right call.
Long way of saying that these things are a lot more nuanced than a simple "teams pay too much attention to ST stats". Fans probably do but I doubt the front office does. (Much).
The big event last ST, was Z sitting behind home plate one game
Hearing the concerns, Zduriencik went down to watch some AB's, watched the hitters stand on the plate and dare Iwakuma to throw something, watched them line shots around like BP, and shook his head "fuhgeddabdouit."
It wasn't stats at all. Iwakuma's stats were fine - as I recall he had a 3:1 control ratio or something.
looks vs production
I pretty much agree with everything you've said in this article.
But, it does make me wonder about one thing. What is the 'normal' skew between "looks" and "performance"?
My thinking is in a VERY high % of cases, the looks and production agree. In all of those cases, the question of "tastes great" or "less filling" is moot.
Let's face it ... in baseball it is HARD to look horrible and be very productive. We remember those exceptions (and likely overestimate their total number), for the simple reason they ARE exceptions.
Looking exclusively at stats, I pegged Carp as a better bet than Smoak, (I tagged Carp before he took his first AAA swing, and shrugged off Smoak after his initial stint with Texas in the Majors - just to be fully forthcoming).
Eyes on - the "initial" take I heard on Carp was "meh" - while the eyes on take for Smoak was - (apologies to Molly Shannon), SUPERSTAR!
At this point, Smoak is a career .683 hitter (1421 PAs), while Carp is a career .740 hitter (608 PAs) ... but Carp is gone and Smoak remains ... (clearly NOT based on production).
Saunders, on the other hand, was viewed as having major "look" issues, while being extremely productive all the way up until he hit the majors.
Can looks override production? yes.
Can production override looks? yes.
But, both of those are within lmits. A very productive, but awful looking AAA player rarely will get an MLB chance. A great looking, but dreadfully producing player won't always get the call up and will eventually get the axe.
Honestly, I think most of the bad lookers end up failing due to injury. Major league effort with bad mechanics is a naturally bad mix. To this day, the best looking SS I have seen in my life was BJ Upton. He "looked" (to my non-scout eye), smoother than even Ozzie. But, 50 errors a year doesn't lie, so today he is an OF.
In the end, what I think "look" vs. "stats" comes down to is there is no way for a scout to see inside any players head. The productivity is what shines a light on the cerebral part of the equation. Doc can look at all the hours of footage available, but this will not tell him that Jose Lopez is basically satisfied with himself and really just doesn't want to put in the effort to get better - or that Michael Saunders is willing to do anything, up to going into a meatlocker and punching dead cows for a couple of hours each day, if the club thinks doing so will give him a legit shot at the bigs.
Veteran vs Unpolished
The spring idea is different from the scouting a young players tools idea. One seems to be a determination of what they are now while the other is a determination of what they could be. What they could be has more variables and wasn't really the discussion. Coming out of spring you have to decide who's better right now and this year in general. Long term analysis of ceiling and likelihood may have an effect on those decisions, but it shouldn't be the determining factor in most cases.
That doesn't mean that scouting doesn't work in determining what they are now, though.
The case's of Michael Saunders, Mike Carp, and Dustin Ackley
Going into last season I had pretty much given up on Michael Saunders. His "look" looked pretty much like it had in the years prior, but then something happened. His new approach started to yield some results, and with renewed confidence in what he was doing, his look changed. The deer in the headlights look was gradually being replaced by a look of challenge. Still though, just as I do not have complete faith in Saunders going into this season, I am sure that he too feels a bit of trepidation. Players in Saunder's current condition need to have success in order to become complete confident in what they are doing. I've seen it before in other ventures.
Mike Carp, on the other hand, never had a confidence issue. Carp was supremely confident in himself to the point of cockiness. On top of that, Carp also has the work ethic to more finely hone what is able to do. But Mike Carp has an issue of "look" that most of us will never get to see, but some of us have. Mike Carp has a bit of an attitude problem that raises it's ugly head every now and then when he is placed in a position of stress. I heard about it in camp, and I've seen it first hand. Gordon and I both saw it first hand when in his first at bat after being sent down from the majors to Tacoma, Mike Carp got into an altercation in his very first at bat with an umpire and got tossed. Carp was then reprimanded by Darren Brown for a couple of games. When things are going his way he is a sweatheart (or so I am told) and a great team player. When things aren't going his way he turns into a something less desireable.
In a Carp versus Smoak steel-caged death match, Carp got tossed to the curb because of his occasional "look".
Now we have Dustin Ackley who bore a striking resemblance to Jeremy Reed last year. That "look" is going to be hard for him to lose, IMHO. In 2011 when he was finally brought up he "looked" like the eager rook who had never faced a level of competition that he couldn't compete in, and had a measure of success. Last year, once the pitchers figured him out, he found himself unable to compete and "looked" lost. I'm willing to bet a ton of cPoints that Wedgie has Ackley under a constant observation from afar to track how he looks during the ST. Ackley's bona fides gives him some rope, but how much depends on what Zduriencik thinks.
Give me 15 minutes with someone and I can pretty well tell you what chance they have of reaching their potential. I'm not bragging, I'm just confident...
Will Sonnett: "No Brag, Just Fact."
Those of you who are old enough and caught the TV Western, "The Guns Of Will Sonnett" will recognize the allusion in the title of my reply. It's a freebie.
Great post, Lonnie, one that gives insight into the human side of GM'ing, coaching, and playing baseball. For most of us all we ever see is games on TV or at the stadium. We see players through a filter that prevents us from seeing only a little, the tip of the iceberg so to speak, of the human side of the game. And yet for the players and coaches the games, which command all our attention, are in a brief, intense environment the application to those games of each man as a person, all his skills, preparation, personality development, social tendencies, etc.
Anybody who hires knows that, though the accomplishments of a person are key to hiring, in the end you hire a person, not a list of accomplishments. Interviews are done not only to explore the substance of a resume, but to get a feel for the person being interviewed. Their track record is a window on that person, but it is not the whole story.
Thanks for sharing you personal insight into Carp and Ackley. Ackley has a TON to prove this year, and it will make a HUGE difference to the M's whether or not he is able to reach his potential.
The value in this?
Tajuan threw 10 pitches to get his 6 outs today....all of them strikes. Two 3-pitch punchouts, and 4 1-pitch outs.
He walked one guy on 4 pitches.
Hultzen had one 4-pitch walk and two hits inthe 1st, but got out of the jam and was fine in the 2nd. K'ed 4.
Beavan needs to be sharp.
moe
Reminder from last year
Pitch f/x does not always work correctly in ST games. It is unlikely that these pitch counts are correct.
I wasn't really paying attention to the pitch counts
But while I was listening on the radio, Walker's walk was 4 pitches and at least 1 of his strike outs were 3 pitches. Other than that I wasn't paying attention, but whenever PitchF/X isn't working, Gameday shows the minimum number of pitches required, 1 for a hit or out, 3 for a K, 4 for a BB, and ball if there is a wild pitch or passed ball.
That said, both Hultzen and Walker pretty well shredded the batters, the only note on Walker is that 1 batter hit the ball pretty hard on his ground out.
Today's traffic report...
Reminds me of Northern Exposure.
Anyone ever watch that show? :)
The traffic report here in Nome - a car just drove down main street!
Some batter got solid wood on one of Walker's pitches...that was the only bad news. :) Love it.
Oh the irony
Is there any argument made on USSM that ISN'T essentially an "appeal to authority"? Lol
Why the guy seems to love endlessly beating up this strawman version of "The Mariners" that is supposedly acting based on some caricature of simplistic jock thinking...is beyond my understanding.
I'd assume the basic thought process employed by all teams in evaluating spring training is Bayesian and hypothesis based...you have a strong "prior" expectation for each player going in, set by past performance and projected development trajectory, and it takes a LOT of ST evidence to override that prior and change your mind on a player. And you have some hypotheses that can help you decide whether to do so as well. Eg, Smoak put in serious hours on a (quite tedious and methodical) offseason improvement regimen. If he's mashing in ST off real pitchers, you've now got a supporting reason to think it could be real progress.
Sure, some teams are better at this process than others. That's called skill (and luck!). But the process itself is decision making 101 in any field...no idea why a smart guy like Cameron thinks a professional sports team would be doing otherwise.
Well said
Why Cameron, a very smart guy, thinks that GMZ and his team are making poor decisions by utilizing a poor process is a mystery to me, particularly when it is GMZ's job that is on the line. Cameron rarely makes an effort to understand the multiple variables a real life GM must consider; e.g., 1) Cameron has reduced the M 's treatment of Iwakuma last year to a simple function of velocity only and has ignored the facts that Iwakuma had a poor spring and got lit up in the final exhibition game in Japan, and 2) reducing the trade of Jaso for Morse to a simple WAR calculation, ignoring the difficulty of finding true MOTO bats & ignoring the free availability of competent back-up catchers like Shoppach. I know that I should argue these points on USSM and not here but I don't have the energy to butt heads with DC.