Filter posts by...
Talent_Distribution
Thursday, April 26, 2007
Some fascinating work by David.
The chart of WAR against “post-season appearances added” is a fairly straight line. In short, each win adds about .06 playoff appearances. The average playoff appearance is about 5 home games per team, so 1 win adds 0.3 home games. Assuming each home game takes in around 5 million$ that goes directly to the the teams, of which most is profit, 1 win should cost 1 million$, over and above what a win should cost in the regular season. (These are just back-of-the-envelope calculations. Nate and Voros among others looked at this issue in more depth.)
The other interesting chart is the following one, which shows that there’s a marginal decrease in pennants added, per win above 4 WAR. So, for those arguing that you should pay more per extra win, that’s not the case. And for those who intuitively believed that giving Pujols a 300 or 400 million$ salary would have been ludicrous, because at some point, you should pay less for every extra win (studes, among others), go to the head of the class. However, with 8 WAR really being about the maximum level, we see that a linear fit is pretty much warranted. Not only is it much easier to calculate, it holds up pretty well here.
Friday, April 20, 2007
Phil points to an article by Chris Isodore asking about where Black ballplayers went. I recently read that the number of black players was in the 30% range in the 1970s. It is under 9% today.
The number of black people is around 10% in USA. But, that’s not the pool of people to select from. The pool is the 17-23 year old males athletes, mostly in USA, but also from Latin America among other areas. And, in USA, it’s not just 17-23 year old male athletes, but the 17-19 year old male athlete in HS, and 18-23 year old male athlete in college. And, as Chris points out, teams are targetting college players. So, what percent of male athletes in HS are black? I dunno… 30%? How many male athletes in college are black? I dunno… 10%? One-third of one and two-thirds of another gives you 17%, which is still twice as much as what we have in MLB.
So, where are they? Football and basketball are also viable places for athletes to go. Maybe 30% of male athletes between 17-19 in HS are Black, but what’s the split between the 4 majors sports? 60% basketball? 30% football? 20% baseball? 1% hockey? As someone else who commented in Phil’s blog said, Michael Jordan may be responsible for the huge shift, if there was one, of black athletes choosing basketball. (In hockey for example, Patrick Roy is single-handedly responsible for all the Quebec goalies in the league; Bobby Orr may be responsible for all the rushing defensemen in the league.) And as Chris notes in his article, football scholarships give more opportunity than baseball scholarships in college.
I also think that while the Vince Colemans, Otis Nixons, Miguel Dilones, and Gary Pettises were all the rage in the 1970s and 80s, the move away from small ball to long ball means power is emphasized over speed. And, real or perceived, Black = speed. Is it any surprise that the players that MLB.com choose for its all-hustle ballot was the “scrappy white guy”? Also, the “extra player” on MLB roster is now a pitcher, not a position player. And, the talent pool of pitchers that are Black is much smaller than for position players.
What we have here is an entire set of independent circumstances, each plausible, each on its own not having enough of an effect, but when accumulated results in a huge swing in Blacks in MLB.
By the way, I can’t stand the term “African American”. I’m from Canada, so am I supposed to say “African Canadian”? In Canada, we have a huge number of people from Haiti and Jamaica. What am I supposed to say? And my parents are from Europe, so should I be termed as “European Canadian”? I don’t have a problem with the label, but rather the labelling process. Why target only one group for labelling? The correct answer is that the group be called whatever it wants to be called. But, I hear “African American” mostly from White people not Black. Then again, alot of my Black exposure is Dave Chappelle and Chris Rock and they definitely don’t say “African American”.
Monday, April 09, 2007
Patriot asks about “greatness”. I answer as follows:
Read More
Thursday, April 05, 2007
Six or seven years ago, I dared ask the question on Baseball Boards. I laid out the process and came up with the answer: yes. The discussion that followed was illuminating and exasperating. The least appealing of the comments, rather than criticizing the process, criticized the conclusion! The problem, which no one pointed out at the time, was regression toward the mean, the single-most important concept to understand, if you are going to analyze sample data. I didn’t know about the concept back then. Once you handle that, the answer changed dramatically: no. Ruth would still be RUTH, but not so RUTHIAN.
The process was similar to what Dick Cramer did, as explained in The Hidden Game of Baseball, but I handled the age adjustment (he didn’t, it seemed). His results can be dismissed. I intended to finally write the followup for THT two years ago, but ended up shelving it. I’ve always intended to finish it up.
BP’s Between the Numbers looked at the issue, but, the execution was lacking. The drawing of the adjusted line really didn’t make much sense. It almost seemed like the author realized the problem, and couldn’t put his finger on it. That work too I would dismiss. Bill James’ timeline is also an effort to just put something in. Now we’ve got David Gassko handling the problem. If we look at his chart, we see that a player around 1950 would have his wOBA of .400 drop to .340 today. That’s a drop of 31 runs per 600 PA. This seems almost as preposterous as Cramer’s findings, even though he took care of the regression toward the mean issue. (My guess is that David didn’t handle the age adjustment.) But, there’s a part two coming up, so we’ll see what he did. IIRC, my work suggests about a 10-run change per 600 PA in that time period, and virtually flat in the last 30 years or so. I really ought to dust that off. The process is semi-reasonable, and the results pass the sniff test.
Wednesday, April 04, 2007
Excellent analysis from Eric Walker from an old article. He neatly shows the change in run scoring was a sudden spurt in 1992-1994, while the years before and after had their own plateau with a certain variation. Sudden one-year jumps in plateaus are not caused by longterm trends like change in player personnel, player conditioning, etc. And expansion can hardly explain a shift of such magnitude. Parks could cause such a change, if you have enough of them changing so quickly, and you can demonstrate the changes. So what we’re left by is what everybody thought almost 15 years ago: the ball was juiced.
Tuesday, April 03, 2007
Using a process similar to that recently published by Chone and jinaz, I come up with each player’s plays made above average for each of the three years, 2004-2006. (Everett from 2004-2006 is +10, +40, +39).
I take the top 30 fielders at each position by balls in zone (i.e., opportunities), for each year. Those are my regulars. Everyone else is a bench player.
The bench 1B is a better fielder than the regular 1B, by 0.7 plays per 162 GP. Here are how the regulars fare relative to the bench player:
Read More
Friday, March 30, 2007
Patriot checks in with some cool graphs as to how much offense was generated by each position, historically. He’s got some data for you to download as well. A few months ago, I also published data on Google Docs. Back then, I hadn’t split the three OF positions (excepts since 1996). I should get that fixed at some point.
I don’t know why the offensive production of 1B and CF would be equal in the late 1940s. Clearly, it would be a mistake to think that the average fielding 1B back then would match the average fielding CF (a good reason not to use offense production to apply a positional adjustment). The huge drop in SS offense in the 1970s was probably due to the artifical turf putting a premium on fielders.
Friday, March 23, 2007
Clay will be starting a chat soon. I submitted the following question, which we’ll see how it will be addressed:
Read More
Monday, March 19, 2007
John Beamer checks in with his forecast for team wins, using the player forecasts of THT. The first sanity check is…
Read More
Thursday, March 08, 2007
Studes follows the Voros approach in describing some players. It is in fact Voros’ approach that allowed me to create aging charts. (See Legend at the bottom)
As you can see, each rate describes something specific.
Now, there’s no reason that you must look at things this way. It assumes a certain independence that perhaps is not warranted. You could for example, look at things in other ways. Rather than removing HBP from the denominator first, then the BB, then the K, then the HR, you can remove all four right away.
So…
Read More
Monday, February 05, 2007
Bill James popularized the Defense Spectrum. It’s one of those brilliant insightful ideas that is clear, concise, and obvious. It goes something like:
P-C-SS-2B-CF-3B-RF-LF-1B-DH
It has two specific meanings. The first is that the fielding premium is higher on the left-side than the right side. The second is as a player’s fielding abilities diminish, he gets shifted over to the right side.
How true is the second part?
Read More
Thursday, January 18, 2007
Yes. Here’s how to tell.
Read More
Tuesday, January 16, 2007
Phil Birnbaum points to an article here with a cool chart that assigns a value to each draft slot. If you work it out, it basically becomes a logarithmic function that looks like this:
Read More
Thursday, January 11, 2007
John Walsh wrote a great piece about OF arms in the Hardball Times Annual. It’s one of those great “well-presented, intensely-researched, easy-to-grasp” articles that I like. You don’t have to get the math-mumbo-jumbo to understand what’s going on. John is one of my favorite sabermetricians, and for that reason, I hold him to a higher standard. In his article:
Read More
Wednesday, January 03, 2007
When we’d play softball with the girls, we’d always put them at 2B and RF for the simple reason that most batters in rec leagues are righties. In the majors, there’s not such a wide gap at all, which is why 2B has had alot of great fielders. What happened in 2006?
Read More
Monday, December 18, 2006
A team of .380 nonpitchers with a team of .410 pitchers will win .300 times per game. We know that using the Odds Ratio Method. This becomes our most reasonable guess as to the replacement level.
A team of non-pitchers being .380, means that an average team of nonpitchers would be .500, or +.120 wins per game. In 162 games, that’s +19.4 wins for 8 or 9 full-time players, or +2.3 wins per 162 GP per player. The pitchers are +.090 wins per 9 IP, or +14.6 wins total.
A team of pitchers being .410 would be .380 for starting pitchers, and .470 for relief pitchers. That gap, .090 wins, is what we’d expect for the same pitcher occupying two different roles. The average starter is around a .490 pitcher, and the average reliever is around a .520 pitcher. So, the average starting pitcher is +.110 runs per 9 IP. Assuming 198 IP per starter, the average starter is +2.4 wins per 198 IP. The average reliever is +.050 runs per 9 IP. Assuming 81 IP per reliever, that’s +0.4 wins per 81 IP.
The total number of wins above replacement (.500 minus .300 times 162) is 32.4 wins. The total payroll above the minimum is about 70 million$. So, the average marginal $ per marginal win is about 2.2 million$ per win.
Of that 70 million$, the nonpitchers get 57% and pitchers get 43% (which is the proportionate numbers of 19.4 and 14.6 wins). Or, 40 million$ for nonpitchers and 30 million$ for pitchers. The other 10 million$ in minimum salary would be split 5.6 million$ for nonpitchers and 4.4 million$ for pitchers. So, the total payroll is 46 of 80 for nonpitchers, or 57% for nonpitchers.
Anyone who wants to say otherwise, do so here. I remain ready to change my mind. I also am ready to do battle.
Wednesday, December 06, 2006
The replacement level winning percentages for nonpitchers, and pitchers is .390 and .420. This makes the average nonpitcher +.110 above replacement, and the average pitcher is +.080 above replacement. This means that 58% of the value is in nonpitchers, and 42% in pitchers. In 2005, teams spent 59% of their payroll on nonpitchers. So, teams get it. They know how to split their money between nonpitchers and pitchers.
But, what about between starters and relievers?
Read More
Thursday, November 02, 2006
I looked at all pitchers who pitched in 1943 and 1944 for the same team. I added up those IP in 1943 and 1944 and compared that to the total league IP. These pitchers accounted for 64% of all IP in 1943 and 71% in 1944. That is, 64% of all IP in 1943 were pitched by pitchers who played on the same team in 1943 and 1944. 71% were pitched in 1944 by pitchers who pitched in both those seasons. Got it?
I did this for all years from 1871 to 2005:
Read More
Monday, October 30, 2006
Here’s a report out of the IIHL, which provides me with a jumping off point.
I wouldn’t be surprised if we can draw parallels to MLB and non-USA players. About 25 years ago, the NHL expanded to 21 teams, and almost all players were North American-born. About 400 of them. Since then, the European influx has mounted, and at the same time, NHL has expanded to 30 teams. Result? Still 400 North American-born players in the league. The 9 new teams have been, essentially, stocked with European players. But, how many Europeans should there be? If you look…
Read More
Thursday, October 12, 2006
From the beginning, and the gap has been widening. Here are some numbers to consider:
Read More
from Amazon
Recent comments
Older comments
Page 17 of 344 pages « First < 15 16 17 18 19 > Last »Complete Archive – By Category
Complete Archive – By Date