Filter posts by...
Scouting
Friday, December 09, 2011
Matt follows up John Mayne’s article from almost two years ago, with his article.
I’m not really surprised by the results. However, I’d like to correct Matt’s interpretation when he says things like:
Not only do pitchers who throw faster succeed more often, but they improve more as well.
Let’s go back to what ERA and component ERA (like FIP) tells us: it is an INFERENCE of their true skill, based on the OBSERVATIONS available. That’s all these things are. If all I know is his K/PA, then I can infer how good he is, with a certain confidence interval. If I know his BB and HR and BABIP, then I can infer better.
So, if you have two guys who have an identical K, BB, HR, BABIP rates, but one also happens to throw 95mph and the other throws 90mph, then we can INFER that the harder thrower guy has less good luck in his ERA than the 90mph guy. It’s more real for the harder-thrower than the softer-thrower.
The job of the saberist is simply to figure out: how much more real. That’s the only job of the saberist, to infer as best he can, given the observations at hand.
Now, if the softer-thrower actually locates his pitches better, then that’s another variable to consider.
But, the important point is this: ERA is simply an observation, and observation that has a certain amount of good luck and bad luck.
It’s more obvious with things like steals per SB attempt. If one guy can run from 1B to 3B in 6.2 seconds, and another guy can run that in 8.2 seconds, and both guys have an 80% SB success rate, guess which guy got luckier? Now, obviously, you want to measure things like jump and reading-the-pitcher, etc. But, given a large enough group of players, say 20 in each, those kinds of things will likely cancel out. So, you’d bet on the faster guys to have a better success rate (since this will presume “all other things equal").
And, hard-throwers are more likely to perform better than soft-throwers, all other things equal.
Tuesday, November 01, 2011
Below you will find the results of the 26 players who played on more than one team in 2011, and who received at least 5 votes from each of their teams’ fans. Wilson Betemit for example received an identical (overall) score from both the Royals fans and the Tigers fans. You will see broad agreement for most players.
Those who didn’t get much agreement either points to a gap in the methodology, or perhaps captures some dissatisfaction with that player. Colby Rasmus for example probably embodies this the best, with a low score with the Cards fans (in 2011) of 44, while he got a 63 with Jays fans. Last year with the Cards, his fans gave him a 57, and in 2009 he was a 71.
Read More
This is the last week to submit your ballots for the Fans Scouting Report.
http://tangotiger.net/scout/
Saturday, October 08, 2011
Looks more like quantitative analysis at the scouting level. Good stuff.
Helping determine who those players are is a major part of Weisbrod’s job as assistant GM of player personnel. A Harvard University English grad with a fondness for the classics, Weisbrod played hockey, but a shoulder injury ended his minor pro career. He found administrative work in the IHL, where he ran the Orlando Solar Bears. The Bears owner, Amway co-founder Richard DeVos, liked Weisbrod so much he named him president of the NBA’s Orlando Magic.
Unfortunately for Weisbrod, he got off to a bad start with Magic fans. Instead of drafting Emeka Okafor, he went for Dwight Howard fresh out of high school. The result was garbage bags filled with hate mail, a sampling of the hostility that followed when Weisbrod traded Tracy McGrady to the Houston Rockets. For that, Weisbrod received hand-delivered death threats at his home.
“[McGrady] was one of the most talented players in the league, very popular, but I came to the conclusion he didn’t have the internal fortitude to win a championship,” Weisbrod said. “I went to the ownership and said, ‘He can be Robin, not Batman.’ The FBI moved me out of my house to a hotel under an alias [because of the public’s anger].”
Weisbrod resigned in 2005, returned to the NHL and later joined the Bruins, where he was hired as a scout by former Harvard teammate Peter Chiarelli. Impressed by Weisbrod’s résumé, Feaster got permission from the Bruins to talk to him. That was before the 2011 playoffs began. Once the Stanley Cup was secured, Weisbrod listened to Feaster and saw another chance to redefine a franchise.
Thursday, September 29, 2011
A few years ago, MLB was filled with great fielding 3B. Lately, however, things have started to change back. I looked at the Fans Scouting Report, to see how the fans evaluated their team’s 2B and 3B. With only three teams did the fans strongly prefer the fielding talents of their 3B over their 2B:
Cleveland Indians
St. Louis Cardinals
Atlanta Braves
On the flip side however, there were 9 teams that strongly preferred their 2B to their 3B.
And how about on offense? Well, the shift has been so dramatic that the average 2B is a slightly better hitter than the average 3B! Just a few years back, MLB not only had the better fielder at 3B, but also the better hitter at 3B. Now, in 2011? A better fielder at 2B and just a shade of a better hitter at 2B, too!
***
How about LF/RF? I noted in the past that the better fielder was by far in RF than LF. Is this still true in 2011? Only 1 team had the clearly better fielder in LF than RF (Yankees). On the flip side, there were 13 teams with the better fielder in RF than LF (with 8 of them being in the NL, a league that doesn’t have a DH, and so, may use LF as a DH-like spot).
How about hitting? Not only is the RF a far far far better hitter than the LF (average RF is a slightly worse hitter than the 1B), but EVEN THE CF is a (slightly) better hitter than the LF. The LF in 2011 is clearly the spot where teams “hide” or otherwise “play with” their players.
Never do what others have done in the past, and that is to “zero out” the stats such that the average LF (offense + defense) is considered equal to the average CF or average RF. This is clearly the wrong thing to do.
Congratulations to Troy Tulowitzki, on being voted the best fielder in baseball, by the hardcore fans who visited Tangotiger.net.
Voting will remain in effect until shortly after the World Series.
Tuesday, September 27, 2011
I’m going to make one post like this a day, focusing on a team that requires more participation.
Today, it’s the Dodgers. If you follow the Dodgers, have a Dodgers forum you frequent, know where the Dodgers fans congregate, then send them my way. Or post a link below, and I’ll post in that forum.
http://www.tangotiger.net/scout/
Monday, September 26, 2011
I’m going to make one post like this a day, focusing on a team that requires more participation.
Today, it’s the Nationals. If you follow the Nationals, have a Nationals forum you frequent, know where the Nationals fans congregate, then send them my way. Or post a link below, and I’ll post in that forum.
http://www.tangotiger.net/scout/
Monday, September 19, 2011
While I have no qualms with his basic point, his conclusion misses the larger point. If it is impossible for Jason to evaluate Mark Ellis and Luke Scott’s throwing, then this only invalidates the results of the data if you end up comparing Mark Ellis to Luke Scott.
HOWEVER, and this is important, this does NOT invalidate comparing Ellis to Brandon Phillips. If let’s say everyone is having a hard time following the instructions, then this bias applies to all secondbasemen, to some similar degree. Which is why getting 20 or 30 evaluators for each player is important: if my instructions to insist on position neutral is too complicated, then those instructions are NOISE. (Random noise, within each position.) And how do your counter random noise? With sample size.
So, it works on two levels:
1. If you believe that fans can go the job on a position-neutral sense, then you get great results.
2. If you don’t believe they can do that job on a position-neutral sense, then random noise of the instructions is reduced by sample size, and you get great results (if you stay within position).
Jason ignores the FSR for whatever reason. But, if you actually look at the results, are you left scratching your head thinking “that’s totally off”? No, you don’t. Well, you shouldn’t in most cases. Indeed, I’ve asked a few teams in the past to evaluate their players (and they do follow a position-neutral aspect to it, as they should). And guess what? They always say the same thing: most of it looks really good.
Tuesday, September 13, 2011
By , 03:22 AM
I know there has been similar research published on the web, but I am too lazy to look it up right now. I took the 1998-2010 draft list from BA and looked at how the various pitchers did in the major leagues, breaking the rounds down into various buckets. It was a quick study and I just matched names from the draft lists with my major league databases. I probably missed 3-5% of the players because the names, especially first names, did not match up exactly.
First I looked at the rookie years for all pitchers in each draft round 1-3+, for a total of 4 buckets. Each round included the supplemental rounds, so, the first round actually has 60 picks in most years. The rest of the rounds pretty much have 30 picks each. Perhaps I should consider the first 30 picks of the 1st round as the 1st round and then the next 30 picks in the 1st round supplement as the 2nd round, etc.
Does anyone know if there is anything special about the supplemental round after the first 30 picks or is it just the next 30 best players and then the second round is 61-90 best players?
For each pitcher, I looked at whether their primary role in their rookie major league season was as a starter (S), a reliever (R), or mixed (N).
The currency I used was my normalized, component ERA (nERC), which is an “ERA” based on a pitcher’s raw stats (s, d, t, hr, nibb, hp, outs, and wp) adjusted for park, opponent, and defense, and normalized to his own league, where the average pitcher in each league, weighted by TBF, is 4.00. I weighted the aggregate nERC in each bucket by each pitcher’s IP. If I used the simple average of all the pitchers (weighting each pitcher exactly the same regardless of how many IP they threw), the numbers would be much higher, as the pitchers who had the worst true talent generally had the fewest IP.
Read More
Monday, September 12, 2011
Great piece by Keith Isley.
Sabermetrics evaluates a player’s “true” abilities as statistical regularities found in large samples. Defensive evaluation is the one aspect of the game that has been resistant to the statistical approach, at least so far. The customary sabermetric solution to this problem is to get more data and treat it with stronger mathematics, which is exactly what the advanced PBP-based defensive metrics do. Sometimes, though, progress requires not simply more data and better processing, but new methods that produce new observations that lead to new abstractions.
Fortunately, there is an established (if not well known) framework for the objective and rigorous analysis of subjective data. It’s called Q methodology, and it’s one of my favorite tools for understanding a phenomenon from a different perspective. Q provides instrumentation (the “Q sort”) for the quantification of subjectivity and a technology (factor analysis) for data reduction and interpretation.
A Q sort ranks orders subjects by a series of relatively subjective variables (such as those typically provided by baseball scouts) and the factor analysis uncovers the commonalities among those subjective elements. The result is a better understanding of what it is that makes a fielder great. Let me run through an example, and I’ll put the more technical details in a footnote.
Note: that article was from Feb, 2006. I took a two year blogging hiatus between Apr, 2004 (my old blog at Primer) and Jun 2006 (when I started blogging here), between which, I worked on The Book. I confined my comments to threads at the old Fanhome/Scout otherwise (for all intents and purposes, inaccessible now).
So, I happen to run into the above article, and, since there’s no active thread for it, I figured it would be a good one to link to. So, if you happen to see me link to more old pieces between those two dates, will, just treat it as something new.
Tuesday, August 23, 2011
The ninth annual ballot is now ready!
This is my single favorite project that I’m involved in, and its success is completely dependent on your participation. Your help benefits everyone out there.
http://www.tangotiger.net/scout/
And help spread the word on your blog!
Wednesday, May 11, 2011
Perfect, just perfect.
Glove-slap: Neil.
Saturday, March 12, 2011
This is why we need scouts.
Glove-slap: Mike Fast.
Monday, March 07, 2011
Works for me:
“Jerry DiPoto probably made the best analogy I’ve heard as far as how to describe the balance between traditional scouting and statistics. He said that it’s almost like you want to imagine a seesaw. On one end of the seesaw is your pure statistics, your pure analytics, and on the other end is your pure traditional scouting where you’re using your gut and what you see to make your call. As evaluators, we decide where we want to stand on that seesaw.
Thursday, February 24, 2011
Buster, wanting to forecast the past, must have talked to a dozen scouts, until a minority of two said what Buster was hypothesizing, and then presented the view of those 2 (and to the reader it’s 2 out of 2, as Buster is obviously not going to report on the 10 that disputes him ):
A couple of scouts say this: they saw wainwright’s arm angle dropping down the stretch last year, a sign of trouble.
This is called confirmation bias. But, this is 2011. And last year is 2010. The data is so easily available. And Jeff looked at it. And Mike looked at it. And they report that the data shows nothing of the sort.
Why not hypothesize that his speed was down, his break was less, his pitch selection less varied, and the time between pitches is longer (all of which are legitimate hypothesis), and then… ask a couple of scouts? Why would you need to ask scouts about counting numbers? You ask the number counters instead. You test your hypothesis. You don’t assert it with selectively sampling the scouts who happen to agree with you.
I love scouts. I’d hire at least 10 scouts for every one saber person. It’s guys like Buster Olney that gives scouts a bad name.
Thursday, February 10, 2011
Carson does something that I’ve been hoping to see someone do for the longest time. What he does is figure out the surplus value of a player drafted, and link that surplus to the scout who signed him. So, if you signed Tulowitzki, you look like a genius. But this is a sample. It still is not something true, something real.
After all, I presume all thirty teams had Tulo ranked somewhere between #1 and #10. How much does it help us to give the 50MM$ surplus (or whatever it was) to the Rockies’ scout who drafted him #7, and 0$ surplus to the other 29 scouts who also ranked him quite high? In that same draft, Alex Gordon was drafted #2. I don’t follow the college scene, but presumably all the teams had him ranked pretty high as well. Do we slap down only the Royals for drafting him, because they are the ones that bought the Alex Gordon lottery ticket, even though the other 29 teams ALSO wanted to buy that ticket?
If you look at it as a sample, then you can say, yeah, the scout won the lottery ticket, that’s the money he made. But, you have to look at it from a true talent perspective. Perhaps you need to regress 99% of what you see from Carson’s process. Scout Bill Buck is credited with 74MM$ in surplus. Perhaps his true value is 740,000$, and the rest was his good fortune for having exclusive dibs on players.
The problem with the process is the exclusivity of it, the binary outcome: did he, or did he not draft that player. In many respects, it’s like looking at a player’s single game, where he puts the ball in play 4 times, and we’re trying to figure out what the true talent of the player is based on the observations of these binary outcome based on whether the batter was safe or out. Because, in this case, you would also regress what the batter did in 4 contacted PA 99% toward the league mean. On the other hand, if you had say his launch parameters (how hard he hit the ball, the spray and vertical angles, the spin imparted), then those 4 PA might regress 95% toward the league mean.
Today is sample-is-not-true day!
Wednesday, February 09, 2011
Ken speaks.
Wednesday, February 02, 2011
This is why scouting is critical:
Strengths: Hernandez has scary upside. He’ll open this season as a 17-year-old and he doesn’t need to develop any more stuff. The only guy in the organization with a comparable arm is big leaguer Rafael Soriano. Hernandez has the best fastball in the system and commands his mid-90s heat well. He regularly touches 97 and could reach triple digits as his skinny frame fills out. Hernandez’ curveball is also unparalleled among Mariners farmhands and gives him the possibility for two 70 pitches on the 20-80 scouting scale. Though he’s young and can easily overpower hitters at the lower levels, he understands the value of a changeup and is developing a good one. He can pitch down in the strike zone or blow the ball by hitters upstairs. He has poise and mound presence beyond his years.
Weaknesses: Hernandez just has to learn how to pitch. He needs to tweak his command and refine his pitches. Typical of a teenager with a lightning arm, he’ll overthrow at times but should grow out of that. Arm problems would appear to be the only thing that could derail him from stardom, and Hernandez has been perfectly healthy so far. The Mariners will go to great lengths to make sure he isn’t overworked in the minors.
The Future: Seattle wants to move Hernandez slowly, but he may not let that happen. He’s not going to need to spend a full season at each level and might need just two more years in the minors. He’ll probably start 2004 back at low Class A Wisconsin—the Mariners concede he could have spent all of last season there—and could be bucking for a promotion to high Class A Inland Empire by midseason. It’s easy to get overexcited about young pitchers, but Hernandez has the legitimate potential to become the best pitcher ever developed by the Mariners.
Put simply: if you only look at performance outcome results, you can’t get a report to read like that. That’s because with limited performance data and against competition that is unlike MLB, you are going to heavily regress. That’s why PITCHf/x is the goldmine, as it’s the one thing that can merge scouting observations with performance results.
At the same time, I can’t just look after-the-fact of this fantastically glowing report, see that Felix exceeed expectations, and determine that scouting “is all that”. You have to look at ALL the glowing reports of 17yr old pitchers and see how often they hit and how often they missed. Then we can say how much impact scouting can have.
This is similar to a statistical-only analysis of saying “hey look, I nailed [whoever… Latos, Strasburg, Weaver, Gooden, etc]”, but then ignoring all the others you ranked highly but didn’t nail. You can get great forecasts for these pitchers by regressing only a little, but then that simply means you are going to include alot of pitchers that shouldn’t be there to begin with (and you’ll be saved by having the MLB managers not pitch those guys and thereby removing them from the sample!).
That’s why Marcel does so well by simply saying: “all minor leaguers will perform at league average”. Because historically, this is close to true (90% of league average or so). But, that’s an after-the-fact view, because managers have selected which players they think will perform the best based on in part, you guessed it, scouting reports.
PITCHf/x, FIELDf/x, HITf/x will eventually remove all this doubt, and finally put us in a position where scouting and statistics can converge to a single common point. Just a matter of time until MLB will create its own academy league so that they can put in the f/x system in parks all across the country.
Wednesday, November 17, 2010
Excellent article, especially for PSYC 101 readers. However, I think this can be worded better:
The human brain just doesn’t do this naturally, especially over a large number of events and/or a long period of time. So even if he could observe a prospects performance through a completely rational framework, he’d be unable to properly couch the prospect’s level of performance in the context of his peers without the use of quantitative methods.
I don’t know what “quantitative methods” actually means. Is it simply to assign a numeric value to a subjective impression? Because all scouts do that, as evidenced by their 20-80 scale. Words accompanies those numbers, but those numbers are a “shortcut” for what they want to say.
Or is he trying to say something else?
from Amazon
Recent comments
Older comments
Page 1 of 320 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date