Just passing it along since it was new to me - the site HitTracker has nice information on homerun distances calculated through a spreadsheet model that is briefly described. It also has links to video clips of the homeruns from MLB.com which is nice.
I especially like the historic homers page and the estimates on Mickey Mantle's 1963 homer along with Glenallen Hill in 2000, Ted Williams in 1946 and Barry Bonds in 2002.
Wednesday, May 31, 2006
Just passing it along since it was new to me - the site HitTracker has nice information on homerun distances calculated through a spreadsheet model that is briefly described. It also has links to video clips of the homeruns from MLB.com which is nice.
Posted by Dan Agonistes at 4:25 PM
Saturday, May 27, 2006
Just returned from a week-long trip to northern New Mexico and southwestern Colorado with my family. The intellectual highlight of the trip was certainly the detour through Chaco Canyon in northwestern New Mexico.
For those unfamiliar with the canyon it is the largest of the Anasazi (a Navajo term translated "ancient ones" or "ancient enemies") sites that dot the four corners region and include the more famous Mesa Verde area which we also visited along with some 200 other sites. The term Anasazi although still found in the literature has been superseded by "Ancestral Puebloans" as the result of the connection modern Pueblo tribes (the 19 tribes that now occupy the Rio Grande valley in the region of Albuquerque) feel for the Anasazi and the fact that most now believe that the Anasazi migrated to the Rio Grande basin.
Of course, what most folks find interesting about the Anasazi and Chaco Canyon in particular is that they largely abandoned these settlements in the 1100s in what has become somewhat of a mystery for archaeologists.
This topic was even more interesting to me since the ruins at Chaco Canyon were the subject of a chapter of the 2004 book Collapse: How Societies Choose to Fail or Succeedby Jared Diamond that I'm currently reading. Diamond is the author of the Pulitzer Prize winning Guns, Germs, and Steel and while that book tracked the rise of civilization since the last ice age, this one discusses societal collapses throughout history and posits the reasons for their decline using a five-point framework. In addition to the Anasazi, Diamond profiles other ancient societies including the Easter Islanders, the islanders of Henderson and Pitcarin, the Mayans, and the Greenland Norse settlements. Diamond then hopes to tease out some lessens from these previous collapses for modern society.
To get to Chaco Canyon you take a 21-mile trip on a dirt road off of state highway 550 several hours northwest of Albuquerque. My wife was not amused as we rumbled along through the desert to what is now a National Park and a World Heritage Site that sees 75,000 visitors per year. After having a little lunch at a picnic site near the Visitor's Center we registered at the center and were ready to take a look around.
The visitor's center is located in the southeast corner of the canyon and we took the loop road northwest towards Pueblo Bonito (PB), the largest of the Chacoan "Great Houses" in the canyon. PB was occupied from the 800s to the mid 1100s and consists of 600 rooms and 40 kivas (the below ground cylindrical structures the use of which is still debated). During its heydey PB was the largest man made structure in pre-Columbian North America and contained structures four or five stories high.
Following Diamond's five point framework he views the contributing causes of the collapse of the Anasazi civilization like so:
The second environmental impact was related to farming techniques used by the Anasazi. When they first populated the canyon they likely relied on flood plain agriculture by allowing their crops (corn primarily) to be watered by the periodic summer rains that washed through the canyon. The canyon itself was well situated to catch rain from a variety of locations on the mesa above. However, as the population grew the Anasazi began diverting water into channels for irrigation, clearing vegetation, and building dams. The unanticipated result was that by 900AD rushing water cut deep channels (arroyos) in the canyon. Once water ran down these arroyos (a couple of which we passed on the loop road and which were cut more than 10 feet below the level of the bottom of the canyon) it was actually below the level of the fields and without pumps it became useless. The Anasazi dealt with the problem by using other schemes allowing them to stay in the canyon but it certainly decreased the efficiency of their food production in an already somewhat difficult environment.
One can imagine that after several years of crop failures the peasants may have become disenchanted with the ruling elite (like those living in PB) and revolted. The result was that the canyon was almost completely abandoned between 1150 and 1200 (the last dated construction at PB was from just after 1110, again using tree ring analysis, and the last construction anywhere in the canyon is dated at 1170). Interestingly, there are several places within PB where you can see the cores cut out of logs used to build various rooms that have been used to date the room. Evidence suggests that the people simply moved on because of the lack of pottery and other things of value that would have been taken.
It's not hard to imagine that once support from outlying areas ceased, perhaps due to their own problems dealing with drought or a lack of confidence in the rulers at Chaco, the society might collapse in a hurry.
In addition, there is evidence of warfare and cannibalism (a notion completely absent from the standard tourist fare at Chaco and Mesa Verde as you can imagine) from other Anasazi sites suggestive of increased competition for decreasing resources. One of the sites we visited at Mesa Verde was "Balcony House" which because of its accessibility only through ladders and toe hold cuts into the limestone, was almost certainly a defensive settlement and one of the last occupied at Mesa Verde. Our kids, of course, loved the climbing making their parents somewhat nervous.
While at the visitor's center I picked up a copy of The Archaeology of Chaco Canyon: An Eleventh Century Pueblo Regional Center, which is a collection of essays on a variety of topics related to the research at the canyon under the auspices of the "Chaco Project" which began in the mid 1960s and continued through the mid 1980s. The book is a kind of final report on the findings of the project. It is written in a fashion that even non-experts can enjoy and I found it very entertaining.
Many times the authors disagree but what I found most interesting was the ongoing debate regarding the interpretation of Great Houses like PB. Was their significance primarily political or was it ritual? In other words, was the Anasazi society a hierarchical one with centralized authority centered in places like PB or was it a more egalitarian society and PB served as a pilgrimage site?
Diamond assumes the former interpretation, that Stephen Lekson articulates in his introductory chapter "Chaco Matters" published in the book I referenced above. One of the most interesting aspects of that essay is how Lekson ties changing modern intellectual models to the interpretation of the Anasazi in vogue at any particular time. The current trend seems to favor the interpretation of Chaco as a ritual center in the midst of an essentially egalitarian society. It seems to this non-expert, however, that the idea of Chaco as the center of a hierarchical society with a ruling elite makes more sense. Some of the points that Lekson makes in its favor include:
Interestingly, there is evidence of road systems throughout the four corners region and roads that lead out of Chaco Canyon to the north and southwest. From my cursory review of the book mentioned above, it doesn't appear significant work has been done on identifying the uses and paths for these roads. Some may been used for commerce, and other had a ritual or spiritual significance. Either way, the presence of such structures indicate central organization. In addition, it is now known that there was potentially a line of site system in place that could have been used for communication among at least some of the Great Houses. For example, from Farview House at Mesa Verde one can see all the way to Chaco.
In any case, one of the things I took away from both books and our recent trip is that although Chaco Canyon is one of the most intensely studied sites in the world, there is still plenty of research to be done and plenty of interpretation that is controversial.
Posted by Dan Agonistes at 4:38 PM
Thursday, May 18, 2006
The third part of my series of articles on hit batsmen is up on BP. Part one looked at the big picture trends from 1901-2005, part two explored a few alternate theories I hadn't considered and that were related to the big picture. Part three today takes a look at the differences between leagues related to the introduction of the designated hitter, expansion, and the double-warning rule.
In the latest article I considered the effect of players who were plunked much more than average. In doing so I referenced Don Baylor, Chet Lemon, Craig Biggio and others. For those curious here are the top 20 players in terms of HBP per 1,000 plate appearances since 1901.
Name PA HBP Rate Period
Craig Wilson 1836 81 44.1 2001-2005
F.P. Santangelo 2014 83 41.2 1995-2001
Ron Hunt 6033 243 40.3 1963-1974
Jason LaRue 2290 85 37.1 1999-2005
Fernando Vina 4685 157 33.5 1993-2004
Jason Kendall 5908 197 33.3 1996-2005
Charlie Babb 1350 45 33.3 1903-1905
Bert Daniels 2178 72 33.1 1910-1914
Reed Johnson 1465 48 32.8 2003-2005
Red Killefer 1719 54 31.4 1907-1916
Dan McGann 4000 125 31.3 1901-1908
Olmedo Saenz 1539 47 30.5 1994-2005
Kid Elberfeld 4939 147 29.8 1901-1914
Steve Evans 3829 111 29.0 1908-1915
Don Baylor 9270 267 28.8 1970-1988
David Eckstein 3151 89 28.2 2001-2005
Doc Gessler 3394 92 27.1 1903-1911
Jack O'Neill 1024 27 26.4 1902-1906
Dick Padden 1663 43 25.9 1901-1905
Joe Yeager 1806 46 25.5 1901-1908
You'll note that there are lots of current and modern players in the list. This is due to the fact that historically we're at almost an all time high for HBP rates and a rising tide lifts all boats. Interestingly, I heard Bert Blyleven interviewed on the local ESPN radio affiliate yesterday and he, like many former pitchers, was bemoaning the lack of aggressive pitching on the part of modern pitchers. As is typical he implied that hitters aren't hit as much which of course is patently false.
In thinking about this common perception it occurs to me there are two explanations. First, there is a human tendancy to view our past as a "golden age" when things were as they should be and things were better. In this case pitchers were better since they stood up to hitters etc. On the other hand, Blyleven could very well be correct. Pitchers today hit more batters, true, but they do so out of accident rather than intent. This is consistent with the fluctuating strike zone, aluminum bat, and even the body armor theories I discussed in the articles.
Posted by Dan Agonistes at 11:27 AM
And now this from The Onion:
MINNEAPOLIS—Sequestered in his parents' basement for the past 18 months, mad sabermetrician Gary Neeman emerged Wednesday after achieving the perfect combination of objective knowledge and functional predictors of future performance, successfully assembling the ideal baseball player's 2006 season statistics. "The VORP of Ty Cobb, the win shares of Barry Bonds, the equivalent average of Ted Williams—I have created the mathematical representation of the ultimate run-producing machine!" said Neeman, holding the sacred Bill James Historical Baseball Abstract above his head and invoking the isolated power of Babe Ruth. "I will soon share this creation with the entire baseball world, as long as I can find an overhead projector in time for the annual SABR Convention in Seattle this June!" Neeman's earlier attempts to produce the perfect player's statistics failed, as each of his first eight results was identical to Albert Pujols' 2005 batting line.
You gotta love it.
Posted by Dan Agonistes at 8:38 AM
Interesting article in yesterday's New York Times by Alan Schwarz on the controversy surrounding the licensing of fantasy leagues (a business that touches 15 million people with a $1.5 billion annual impact) by MLB Advanced Media. As an employee of MLBAM (as a stringer for MLB.com's GameDay system) I'm probably not in the best position to comment but what folks have asked me is if the lawsuit has any bearing on the analysis of statistics. The following quote from Schwarz appears relevant in this regard.
"Major League Baseball Advanced Media is not making a copyright claim to the statistics themselves; a 1997 decision in the United States Court of Appeals involving the National Basketball Association ruled sports statistics to be public-domain facts that do not belong to the leagues.
Rather, the central issue concerns celebrities' ability to control use of their names in commercial ventures, and how this "right of publicity," which has developed under state common law and statute over the last half-century, may commingle with Constitutional press protections under the First Amendment."
So it would appear the issue revolves around how the players are portrayed and can be interacted with in the venture in question and not really their statistics, although some are worried that it's a slippery slope. Schwarz writes that the case could go either way and should be decided in September.
Maury Brown over at The Hardball Times had a nice piece on this as well back in January that has additional background info.
Posted by Dan Agonistes at 8:19 AM
Wednesday, May 17, 2006
I promised myself that I wouldn't write anything on this topic and was actually feeling pretty proud for not having done so thus far. Given what I've written on the topic previously and the polarization that this issue engenders, I've spent the last three weeks or so concentrating on the historical trends in hit by pitch data as well as watching the Cubs struggle to score runs.
But, with 714 approaching I can't resist any longer. My plan had simply been to ignore 714 and 715 (as I have Barry's ESPN show) because in my mind the entire Bonds persona has become a freak show. But then I read this article from Dayn Perry who reminds us not to get too sentimental about Ruth. After all Ruth was a drunk, a philanderer, a bad teammate, a glutton, ad infinitum - facts which journalists of the day ignored. Interestingly, I heard the author of the new Ruth biography Leigh Montville interviewed on an MLB radio podcast this morning and he noted that Ruth's grand daughter believes that Ruth also had ADHD.
While I have no problem with Dayn's article (and he even brings Yankee Stadium park factors into the discussion which you gotta love) and agree that we shouldn't get all upset about Bonds passing a guy like Ruth, I think some have used Ruth's moral failings to try and excuse Bonds' actions. The two men and their situations are totally separate issues and should remain so. My distaste for what's going on is rooted in what Bonds has done, not what Ruth did. The overwhelming evidence - the kind that lets reasonable people make judgments every day - clearly indicates that Bonds cheated. Attempting to excuse him because Ruth may have been a louse much of the time makes no sense. As to what the unenhanced Bonds might have done, my opinion on the subject is similar to this quote by Cory Lidle.
"What he could have done without performance-enhancing drugs--which he hasn't been proven guilty of [using], which I'm not buying--you can maybe take what he had done in his prime, before his head started growing at an enormous rate, and just make those projections. Say that, 'This is what he could have done.' Maybe it's 550 home runs. I don't know. It definitely wouldn't have been anything close to 700."
What's more dicey in my opinion is the view espoused by some that Ruth's feats should be downgraded because he played in an era before integration, night games, the slider and all the other changes in the game since the mid 1930s. That train runs both directions, however, as Bonds had access to (legal) medical care, better equipment, and simply the accumulation of knowledge that goes with any activity as it is refined over time. Even without steroids Bonds' career has been lengthened by the operations on his knee and treatment he's received on his elbow.
I have no doubt that if Bonds and Ruth were to match their hitting skills Bonds would be by far better. An interesting analysis of this question was performed by Nate Silver in a chapter of Baseball Between the Numbers. There Silver creates a league difficulty factor based on Davenport Translations by examining the performance of players in successive seasons. He then uses these factors to translate statistics across eras. His analysis of Ruth concludes as follows:
Ruth's career EqA would be .274. He probably would have made the All-Star team a couple of times, with an EqA in his best seasons approaching .300. But he'd be remembered as merely a good player and certainly wouldn't be a credible candidate for the Hall of Fame. In modern terms, Ruth might be a Tino Martinez (career .274 EqA) or Raul Mondesi (.278).
And this idea holds for all players of the past. It's simply a fact (akin to athletes in other sports such as basketball or track and field although not as severe) that most major league players today are better than most players of the 1920s and 1930s - which is why in the end the argument seems kind of silly. Players should be judged in their own context.
But on a different note I ran into this article that tells the stories of two kids who witnessed Ruth's final three homeruns in his visit to Forbes Field on May 25th, 1935.
Posted by Dan Agonistes at 11:40 AM
Tuesday, May 16, 2006
In the latest Cubs debacle on Sunday afternoon Brian Giles of the Padres set a team record by walking five times. However, Dave Smith of retrosheet reports that pitcher Clay Hensley also struck out 5 times which led him to do some research. His conclusions (with some commentary) as posted on the SABR-L list this morning include:
What's really disturbing of course for Cub fans is that while Giles walked five times on Sunday Jacque Jones (he of the .303 OBP) has walked five times all season in 102 AB, Ronnie Cedeno (.309 OBP) has walked 4 times in 132 AB, Juan Pierre just 8 times in 151 AB (.269 OBP), Neifi Perez just 4 times in 64 AB (.235 OBP), Jerry Hairston four times in 60 AB (.313 OBP), John Mabry all of once in 46 AB (.250 OBP), and Fred Bynum once in 32 AB (.212 OBP). I sincerely hope the Cubs front office isn't scratching their heads trying to figure out why it is that their team is dead last in runs scored behind even the Pirates.
No walks + No power (they're also last in SLUG) = No Runs
And of course with this collection of hitters it is simply not surprising that this is the result. They've intentionally loaded their team with players with poor plate discipline and a heavy reliance on batting average and while you can accept a shortstop like Cedeno to hit this way, you can't accept it from your two corner outfielders and every one of your reserves. That's a recipe for disaster cooked up by Jim Hendry and Dusty Baker.
Posted by Dan Agonistes at 8:29 AM
Friday, May 12, 2006
A fellow BP'er alerted me to this interesting use of performance analysis in the NBA from the Fort-Worth Star Telegram:
Del Harris said no one on the staff raised an eyebrow when Johnson said Devin Harris would start.
"We had some pretty good statistical evidence from Mark [Cuban's] people to back up the decision," Del Harris said.
"We felt like we at least needed a 95-point game to win. They're difficult to beat in an 80-point game. We can push the ball better with Devin because he gives us two guards who can really get out and run."
Very interesting and logical approach.
Posted by Dan Agonistes at 6:27 AM
Friday, May 05, 2006
Well, an exciting finish at Coors Field tonight where as I was scoring the game as Jamey Carroll drew the walk off walk in the bottom of the ninth to seal the 5-4 Rockies victory over the Astros. The inning was setup by a throwing error from Morgan Ensberg on a Clint Barmes sacrifice attempt after Brad Hawpe had doubled off Brad Lidge to lead off the inning with the Rox down 4-3. Normally I'm not a big fan of the sacrifice in that situation but with the way Barmes has been swinging the bat and the fact that he's not left-handed that's probably a good move.
To me, the most controversial managerial move of the night occurred in the top of the 3rd inning with the Astros leading 1-0. They had runners on second and third and Preston Wilson coming to the plate to face Byung-Hyun Kim. Again, usually I'm not one to make alot of matchups nor am I a big fan of the intentional walk but...in this case Wilson was 10 for 14 against Kim coming into the game and had plated the first Astros run with a single to right in the first inning with two outs. So when Wilson came up again in the third he was 11 for 15 against Kim in his career.
If you assume at bats are Bernoulli trials in a binomial distribution then the odds of Wilson going 11 for 15 off of Kim are approximately .0015%. That assumes Wilson is normally a .254 hitter against Kim (a value I calculated using Bill James' Log5 method for my article "Tony LaRussa and the Search for Significance". In other words with those assumptions Wilson would be expected to go 11 for 15 approximately 1 out of every 68,000 trials of 15 at bats.
With first base open and a struggling Jason Lane on deck (he's hitting .200 but he has walked 23 times and hit 6 homeruns) I think I might have intentionally walked Wilson and taken my chances.
Kim pitched pretty well with the exception of the Wilson at bats and struck out nine and only walked one in 7 innings of work. Todd Helton, fresh off the DL, singled in four trips and looked a little rusty.
Posted by Dan Agonistes at 9:59 PM
Hate to say I told you so but the Cubs are struggling mightily to score runs. It didn't take a rocket scientist to see that everything would have to go right for this team to compete offensively. Everything has not gone right with Derrek Lee being out for two months (although Will Carroll is optimistic that he'll be back sooner because technically he did not break his wrist) and Aramis Ramirez fighting a slump.
Over the last six games the Cubs are 1-5 and have scored a grand total of 5 runs. They've been shut out three times and been outscored 45-5 over that span. They've now fallen back to 14-13
Juan Pierre has an OBP of .313, Aramis Ramirez is sitting at .207/.310/.391, Jacque Jones is at .239/.286/.451 and the combination of Neifi Perez and Jerry Hairston has an OBP of around .240 with all of three walks. As a team the Cubs are last in the league in runs scored (112 or 4.15 per game) and second to last in homeruns (21) ahead of only the Padres (19).
On the plus side both rookies, Ronny Cedeno and Matt Murton are hitting well along with Todd Walker. An historic start from Greg Maddux has also compensated for the struggles of Carlos Zambrano. We can only hope that averaging 109 pitches per start from ages 22 to 24 and the WBC aren't the culprits.
The underlying problem was masked early on because the Cubs were hitting so well with runners in scoring position and leading the league in that department for more than half the first month. Over the last week, however, their luck has turned with a vengence and despite getting some runners on (they left 11 on base in last night's shutout loss) they're not getting the hits to fall when they need them. They now sit 5th in the league in OPS with runners in scoring position at 812 but have driven in only 93 runs in those scenarios (11th in the league) largely because they have a 700 OPS overall which is 14th in the NL.
Although they're not as bad as they've been over the last week, I'm afraid the overall result is pretty much what we're going to see. Before the season several folks predicted around 84 wins for this team and right now that's looking pretty good.
So are the Cubs struggling? "Yes", in the last week but "No" overall.
Posted by Dan Agonistes at 1:24 PM
Wednesday, May 03, 2006
In case you missed Will Carroll's Under the Knife column today here's a story about a switch pitcher in Division I who plays for Creighton.
It makes me wonder why we didn't see more switch pitchers a couple generations ago. Tony Mullane. who played from 1881 to 1984 did so and of course Greg Harris threw an innning for the Expos with both hands back in 1995. Back then natural lefties (like my Dad) were steered towards right-handedness. You would think there would have been some individuals who continued to be able to use both hands and at least a few would have excelled. Anyway, today you would need a concerted effort to make it work like this boy's father put forth from the time he was 3.
Posted by Dan Agonistes at 1:17 PM
I was lucky to find this excellent post on the Orioles Think Tank related to Corey Patterson. The author runs some comps for Patterson's profile and the results are not very encouraging. Corey has cut his strikeout rate this year (9 in 58 AB) but continues to treat a walk as a personal affront (he has exactly 1). The author makes the point that almost the only way someone with Patterson's profile has been able to log as many plate appearances as he has is that he has other skills (defense and base running and a little pop in his bat) that add to his value. Even so, he's still not league average and so the long and short of it is that unless he finds a way to boost his offensive performance, he'll be driven out of the league sooner rather than later.
Posted by Dan Agonistes at 12:29 AM
Monday, May 01, 2006
Ok, I noted in a previous post that I was kind of on the fence on the whole Allard Baird firing thing and that it seemed to me that the buzzards were circling even though it was clear to anyone familiar with the team that they would be lucky to not lose 95 games.
Then I received this excerpt from Ron taken from a recent column by Rob Neyer.
Last September, we spent a few innings together at a game in Seattle. At one point, Baird gestured toward Terrence Long in left field and said, "I can't believe my phone's not ringing about that guy."
Seriously. At that point, Long was nearly finished with a season in which he did bat .279, but also posted a .321 on-base percentage and a .378 slugging percentage. He probably wasn't the worst everyday player in the major leagues, but then again, maybe he was. Now, it certainly didn't make much sense to continue playing Long down the stretch -- he got more action in September than in any other month -- when younger, more talented players were on hand. Rather, it didn't make sense unless Baird believed that playing Long was akin to showcasing Long, which might have worked except everybody else in baseball was aware that Long couldn't, you know, hit or field.
Just as [David] Glass really didn't know the Royals were going to be terrible this season, Baird really didn't know that Long wasn't good enough to play regularly in the major leagues. Of course, Baird wasn't the only one. Manager Buddy Bell kept writing Long's name into the lineup. In the middle of September, in response to a question, Bell said, "But at the same time, you don't just want to give up the ship. And that, quite frankly, is why Terrence is playing. He's been our best player for the last month. He gives us a better chance to win."
Bell said that on the 11th. Over the previous calendar month, Long's average dropped by a point (from .278 to .277) while he hit zero home runs, two triples, and four doubles, and drew eight walks. All while playing in nearly every game. I'm sure that Bell had access to all the numbers, and it's possible that he even looked at them. But he believed that Terrence Long was playing well, and that was all that mattered to him.
In the Royals' organization, everybody seems to live in a sort of Bizarro Baseball World, where left is right, up is down, and Terrence Long is better than Matt Diaz. Everybody's delusional, and it's not coincidental. The owner's delusional, and so it should not be surprising that he hired (and continues to employ) a delusional general manager, who hired (and continues to employ) a delusional manager.
Ummm. Maybe I should take that back. I had kind of assumed that Bell was insisting on playing Long (aka "Magellen") and that Baird was going along since he was paying him $2M. Turns out it would seem Baird clearly doesn't quite understand the baseline level of performance expected from a corner outfielder. That's kind of scary.
Posted by Dan Agonistes at 11:17 PM
As I mentioned in a previous post Retrosheet now has data from the 1922 National League. I downloaded it this evening along with the 1993-1998 data to push my play by play database to over 5.8 million rows.
One of the things I wanted to look at given my recent columns was whether the platoon splits differed during that time. Here are the results.
Throws Left Throws Right Platoon Split
Bats PA AVG OBP SLUG PA AVG OBP SLUG AVG OBP SLUG
L 1602 0.263 0.309 0.324 9632 0.306 0.356 0.436 0.043 0.047 0.112
R 5518 0.303 0.341 0.428 15021 0.281 0.318 0.380 0.022 0.023 0.049
I took out switch hitters. By comparison the 1970-1992 data I used in a previous column showed split differences that weren't quite as large. For example for left-handed hitters it was 24, 26, and 56 points in AVG, OBP and SLUG and for right-handers it was 17, 19, and 33. For the 1922 National League therefore the splits were almost 100% higher for left-handed hitters and about 50% higher for right-handed hitters. Could it be the case that split differences have actually shrunk over time or is this simply a case of small sample size?
I took a quick look at my 1970-1992 data set by year and didn't see any noticeable trend in either direction. There is a lot of variability in the data, especially for lefties where the difference in slugging percentage can go from 44 points one season to 75 the next as it did from 1985 to 1986.
Of course the data since 1970 may not show the trend and so I'll probably need to load all of the data back to 1957 and then take another look.
Steve Goldman of Baseball Prospectus alerted me to the fact that the 1922 data bore out the oft-told story that Casey Stengel learned about platooning from his manager John McGraw who sat the lefty hitting Stengel against southpaws. I ran a query to confirm this and it shows that Stengel batted 246 times against right-handers and just six times against lefties.
However, I'm wondering if the 1922 is missing some games or at bats since the data I loaded contains 252 plate appearances for Stengel that season and baseball reference has him for 283. Hmmmmm.
Casey Stengel 1922
PA AB H 2B 3B HR SO BB IBB HBP SH SF AVG OBP SLUG
R 246 216 82 7 9 6 15 19 1 7 1 2 0.380 0.439 0.579
L 6 6 3 0 0 0 0 0 0 0 0 0 0.500 0.500 0.500
Posted by Dan Agonistes at 10:38 PM
This won't be as good as Aaron's Link-O-Rama posts but I have run across a few interesting things related to performance analysis and other topics recently.
Albert shows that when using these four outcomes as rate statistics, he can get an R-squared of .896, which is right inline with Runs Created and OPS. However, when he adds a differntiation between all hits and doubles and triples the correlation doesn't improve dramatically and only rises to .909. Although he doesn't show it, I would assume that strikeout rate doesn't add much to the picture and that perhaps using doubles and triples instead of strikeouts may in fact yield a correlation greater than .896. He then goes on to use a model to estimate the distribution of talent across his four measures and calculates the regression to the mean that would be used when estimating a player's true skill levels. Of course he shows that hit probablity is far more influenced by luck than any other factor and that strikeout probabilty is the most dispersed of the measures. In any case, some very interesting work here as usual.
Posted by Dan Agonistes at 12:05 PM