SoxProspects News
|
|
|
|
Legal
Forum Ground Rules
The views expressed by the members of this Forum do not necessarily reflect the views of SoxProspects, LLC.
© 2003-2024 SoxProspects, LLC
|
|
|
|
|
Forum Home | Search | My Profile | Messages | Members | Help |
Welcome Guest. Please Login or Register.
All time WAR leaders/As Criteria for HOF?
|
Post by tonyc on Jan 9, 2014 14:17:52 GMT -5
Having been an "old school" stat-head, growing up with Stato-matic leagues in the late '60's it was not until today I read how FIP and WAR are actually calculated, and I looked at the all time WAR list and found some surprises and wanted some feedback as to whether this criteria may have inherent flaws- or my preconceptions do.
Mariano Rivera is 207th, wheras the long but for the most part not brilliant careers of John Olerud and Frank Tannana are 197th and 199th. Given that Hoyt Wilhelm appears at 330 does relief pitching have less of a component toward victory than we preconceive? I've been under the impression- and perhaps one of you has done a study, that when a team finishes with a better record than their plus minus would predict, such as the Yankees in 2013, that may be in part due to a good closer/bullpen.
Another question involves the age-old debate, when voting for HOF of brief brilliance vs. longer solid numbers. Carl Yaz appears number 35, vs. Joe D. at 67, and Dizzy Dean appears way down at 377- do any of you feel this improperly weights the long-term?
Thanks
|
|
jimed14
Veteran
Posts: 25,831
Member is Online
|
Post by jimed14 on Jan 9, 2014 14:36:52 GMT -5
Good topic. I'm in the same boat as you, having started really getting into the new stats about a year ago.
Good question about relief pitching - I'll leave that to someone else. You could also add to the conversation how much fielding adds/subtracts from WAR and then DHs. It's really hard to believe how low Manny is in career WAR. And hell, Papi is about equal with Chuck Knoblauch at this point. really?
I don't know if we should use WAR as criteria unless it's to compare same positions. I don't have an answer. But I do believe for sure that players who are very good for a very long time as well as players who were unbelievably dominant for a short time should both get in.
At the minimum, they should just throw out the old 400 HR/3000 hits/3000 K criteria.
And please, there is no reason why Dwight Evans is not in the HOF.
|
|
|
Post by mattpicard on Jan 9, 2014 14:46:59 GMT -5
Well, keep in mind that WAR is a counting stat, so the more you play, the higher it's going to be (assuming you produce above replacement level). Thus, a player whose career lasts several years longer than another is obviously going to appear superior in value off of a list of players ranked by WAR. How you use the stat (and others) in this instance really depends on what you personally feel are the best indicators of a player's hall of fame credentials. Is it aggregate career value, as shown (imperfectly) from a stat like WAR? Do you want to focus more on peaks (single out the top seasons) or longevity? Check out Jay Jaffe's JAWS for a happy median. As for relieving, and closing, specifically, keep in mind that since you're only getting ~3 outs of a 27 out game, your value is going to be limited no matter how dominant you are. Hence, you see elite closers such as Rivera, Kimbrel, and Uehara (in 2013) struggle to eclipse the 3 WAR mark in any given season, whereas top tier players at other positions accumulate twice that amount.
Just remember as you get into these more modern stats: they are incredibly useful and telling, but they don't tell the entire story. As for the +/- and record difference, I'm sure the bullpen plays a role in that, but I'm not really well-versed with that topic. I think the manager is another significant factor in teams W-L records outperforming their +/- numbers.
FYI I may soon move this to the off-topic forum, as it doesn't specifically pertain to anything Red Sox-related.
|
|
|
Post by tonyc on Jan 9, 2014 16:24:28 GMT -5
Good responses Jim and Matt. Jim, in the post I nearly added my delight in seeing both Dwight Evans and Luis Tiant around 125th- a level eclipsed largely by HOFers, and ahead of a number of them. Why Dewey isn't in? He represents the perfect storm of excelling in every underrated criteria, less than the overrated criteria in traditional baseball analysis: He started his career not making the great initial impression and became better late. His greatest strength was in being about the best defensive, not offensive right fielder in our lifetime, save Roberto Clemente. In terms of offense, batting first or second, he excelled in runs, not rbi, and finally, at a time when batting average ruled, not ops, he derived fine stats due to ok ba, but better obp due to walks, and good slugging. Jim, that is a good point also about using lifetime WAR to compare players at the same position, it does seem there are some inherent flaws in there. Matt, great point about it being a counting stat, so that a player continuing to add above marginal stats, who however, compromises great earlier performance with mediocrity will pad his lifetime WAR. This accounts for some of the high WAR numbers of long time players such as Olerud, Tananna, and even my beloved Yaz. For HOF purposes, while I agree that both the brilliant short termers and consistent long-termers have a place, this represents somewhat of a flaw in cross-comparison (an overweighting toward longevity).
Thanks, any other analysis quite welcome here.
|
|
|
Post by rjp313jr on Jan 9, 2014 18:29:08 GMT -5
You have to be careful using numbers in general when identifying a HOF player. I personally care more about peak performance then piling up stats as a slid player. But a sustained period of excellence needs to be factored in. However, if a guy was never considered one of the tru elite players during his time I have trouble calling him a Hall of Famer. Jim Rice gets a lot of flack but at least he was truly one of the greatest hitters for a period of time. Feared by the opposition.
I struggle with a guy like Biggio because he was real good for a long time but was only ever borderline great.
Evans isn't a Hall of Fame caliber guy and there's nothing wrong with that. This is for the elite. Best of the best.
|
|
|
Post by tonyc on Jan 9, 2014 20:21:41 GMT -5
Rip,
On the surface it appears Dwight is not, but please read Bill James clear analysis that he IS entitled "an open letter to the hall of fame about Dwight Evans." He makes an excellent case, just as I was delighted when (granted a stonger case) Bert Blyleven made it due to modern analysis, and Jack Morris didn't.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 9,027
|
Post by ericmvan on Jan 13, 2014 18:04:02 GMT -5
You have to be careful using numbers in general when identifying a HOF player. I personally care more about peak performance then piling up stats as a slid player. But a sustained period of excellence needs to be factored in. However, if a guy was never considered one of the true elite players during his time I have trouble calling him a Hall of Famer. Jim Rice gets a lot of flack but at least he was truly one of the greatest hitters for a period of time. Feared by the opposition. I struggle with a guy like Biggio because he was real good for a long time but was only ever borderline great. Evans isn't a Hall of Fame caliber guy and there's nothing wrong with that. This is for the elite. Best of the best.This is almost entirely bullshit, but all of it is bullshit that every member of the BBWAA promulgates nowadays, so you're not guilty of anything but going along for the ride. Point by point: 1) IOW, if baseball writers were idiotically clueless about how valuable a player actually was back when he played, we have to ignore all the evidence that they were being idiots, and continue to refuse to give the player the credit he deserves. So if a guy retires as the 14th best player in baseball history, based on WAR5 (average of bWAR and fWAR, total of his 5 best seasons), but gets just 3.9% of the BBWAA vote, and stays on the ballot for 15 years while saber-types declare the situation a travesty, but never gets higher than 43% of the vote, by all means, let's let him die from complications of the diabetes that shortened his career before we finally induct him. 2) Biggio's 5 best seasons total 32.2 WAR, which ranks 97th among position players, 13th among 2B. The 13th best 1B (with 34.3) is Frank Thomas. Jim Rice? 30.3. But let's continue to define greatness while ignoring defensive value! It's easier! 3) The number that predicts HOF election for players before 1960 is WAR5 + 1/2 WAR, which we'll call HW (Hall Worthiness, and, yes, if I get off my ass this will be an article for someone, probably THT). Almost everyone with HW > 51 is in the HOF; the exceptions are Sherry Magee, Minnie Minoso, Jimmy Sheckard, Heinie Groh, a quartet of 19th century guys, and a quartet of guys who padded their numbers with huge years from 1943-5 when half the talent in baseball was at war (Bob Johnson, Charlie Keller, Stan Hack, and Bob Elliott). (In the HW range that covers those 12 exceptions, there are 48 HOFers from before 1960). So, are Paul Waner, Mickey Cochrane, Bill Dahlen (19th century SS), Joe Cronin, Luke Appling, Home Run Baker, Gabby Hartnett, Sam Crawford, Goose Goslin, Joe Gordon, George Sisler, Bobby Wallace, Fred Clarke, Richie Ashburn, Jesse Burkett, Jack Glasscock (19th century SS) , Sherry Magee, Bill Terry, Ralph Kiner, Elmer Flick, Zack Wheat, Hughie Jennings, Joe Medwick, Joe Kelly, Billy Herman, Jimmy Collins, and Larry Doby the "elite, best of the best"? That's their rank by HW (which also includes a boost for catchers of up to 30%). All but Dahlen, Glassock and Magee are in the Hall. The post-1960 players spanning the same range are Larry Walker, Bobby Grich, Joe Torre, Craig Biggio, Ted Simmons, Ryne Sandberg, Paul Molitor, Willie McCovey, Ozzie Smith, Alan Trammell, Tim Raines, Dick Allen, Barry Larkin, Graig Nettles, Rafael Palmeiro, Eddie Murray, Edgar Martinez, Mark McGwire, Roberto Alomar, Thurman Munson, Tony Gwynn, Kenny Lofton, Bill Freehan, Dwight Evans*, Lou Whitaker, Sammy Sosa, Andre Dawson, Billy Williams, Buddy Bell, Ken Boyer, Harmon Killebrew, Sal Bando, Reggie Smith, Keith Hernandez, Willie Stargell, John Olerud, Bobby Bonds, Darrell Evans, Jason Giambi, Jim Wynn, Willie Randolph, Dave Winfield, Gene Tenace, Robin Ventura, Tony Perez, Cesar Cedeno, Jason Kendall, Willie Davis, Lance Parrish, Ron Cey, and Fred McGriff. (That's a few more names than there ought to be, even accounting for expansion, and I bet you can spot the guys who didn't get there just by talent.) *I've given Dewey some credit here for the peak WAR that the 1981 strike cost him. Otherwise, he's between Stargell and Olerud. BTW, I could keep going: Minnie Minoso, Willie Keeler, John McGraw, Ernie Lombardi, Buck Ewing, Bobby Doerr, Dave Bancroft, Frank Chance, Max Carey, Jimmy Sheckard, Charlie Bennett (19th century C), Roger Bresnahan, Earl Averill, Kiki Cuyler, Joe Sewell, Heinie Groh, Tony Lazzeri, Enos Slaughter, Bid McPhee, Joe Tinker, Jake Beckley, Chuck Klein, Cupid Childs (19th century), Nellie Fox, Wally Schang, Hack Wilson. 19 of 24 in the Hall, and Schang's only in there to make the number of catchers on the list to that point match the other positions. The post-1960 equivalents are Luis Gonzalez, Vada Pinson, Chet Lemon, Jeff Kent, Jim Rice, Dale Murphy, Will Clark, Fred Lynn, Norm Cash*, Jim Fregosi, George Foster, Rocky Colavito, Jose Cruz, Dave Parker, Jim Sundberg, Kirby Puckett, Brian Downing, and Orlando Cepeda. *Cash makes it solely on the basis of his corked-bat 1961. (For post-1960 players, you'll notice the ridiculous bias in favor of slugging without regard to defensive value. Except for McGriff, every power-hitting non-PED-tainted 1B has made the Hall -- McCovey, Murray, Killebrew, Stargell, Perez, Cepeda.) You should realize that, to the best of our knowledge, the difference between these rankings and what seems to make sense to you is the long-term ineptness of player evaluation by baseball writers and broadcasters, which comprise the nearly sole architect of your sense of greatness. Defense has been ignored entirely unless a guy was famous as the very best; if you saved 120 runs in your most valuable seasons, that gets you into the Hall, but if you only saved 108, that'll get you 1.7% of the vote even though you had 2500+ career hits (the only stat many of these guys look at, seriously*) and your combination of peak and career value puts you right between HOFers Billy Williams and Harmon Killebrew (the "Brooks / Buddy rule"). The defensive spectrum for these guys doesn't exist; there's SS, and then there's everyone else, so that Eddie Matthews can show up on the ballot as the greatest 3B of all time and get 32.3% of the vote, or Gary Carter as the 2nd greatest catcher and get 42.3%. Oh, and if you're a SS, only defense matters, so Arky Vaughan can retire as the 2nd greatest SS of all time and not be one of the 100 players screened to get on the ballot. And, of course, walks don't count for hitters. In fact, in so far as they reduce your ability to get high hit totals, they count against you. And, as I said, all of these absurd and idiotic distortions of evaluation have underlined the degree of attention that players have gotten from the media. And hence shaped what most fans think of them. Another low point came in 1992, when Bobby Grich appeared on the ballot. With his 17-year career featuring a very good bat, tremendous batting eye and superb defense at a difficult position, he was one of the 50 Hall-worthiest position players ever. And he got 2.6% of the vote -- and no one complained. Yet despite the idiocy of the voters, an actual HOF standard was indeed set by the BBWAA's own members (much of it by the Veteran's Committee, fixing ballot oversights) . To create it, mind you, they had to induct a trove of lesser players: Johnny Evers, Travis Jackson, Harry Hooper, Edd Roush, Hugh Duffy, Sam Rice, Heinie Manush, Sam Thompson, King Kelly, Earle Combs, Deacon White, Jim O'Rourke, Red Schoendienst, Rabbit Maranville, Ross Youngs, Jim Bottomley, George Kell, Pie Traynor, Freddie Lindstrom, Chick Hafey, Rick Ferrell, and Ray Schalk. The "best of the best" has never been true. It's just "best." The original standard was roughly two guys per position per decade, which is to say, best guy in the league at your position in your decade. Which seems to be fair. The Hall as originally stocked was for guys good enough that they should not be forgotten. Not merely for the legends. Ducky Medwick is not a legend. Just a guy with a legacy worth remembering. The subsequent generation of BBWAA writers has, however, bought into this "best of the best" nostalgic nonsense and is in the process of keeping out of the Hall 50 or so guys who are incontrovertibly Hall-worthy by the standards for the first 60 years of modern baseball. *In 2010 the 11 members of the Historical Overview Committee nominated 5 position players for consideration by the Expansion Era Committee. According to HW, the top 5 candidates were Grich, Ted Simmons, Graig Nettles, Thurman Munson, and Buddy Bell. Here, however, is a ranking pf eligible players by career hits, with their HW rank among eligibles: 1. Al Oliver (23rd in HW) 2. Rusty Staub (17th) 3. Steve Garvey (28th) 4. Buddy Bell (5th) 5. Ted Simmons (2nd) 6. Dave Concepcion (24th). Their five nominees were the above, less Bell.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 9,027
|
Post by ericmvan on Jan 13, 2014 18:27:32 GMT -5
Check out Jay Jaffe's JAWS for a happy median. Of course, Jay just pulled that formula out of whatever was handy. I decided to see what the actual criteria for Hall inclusion was. Should peak be 3, 4, 5, 6, or 7 years? How much weight versus career WAR? I took the 39 players in the chunk of the JAWS ranking of pre-1960 guys where there are some HOFers and some non-HOFers, and tested a bunch of regressions, and what actually predicts the results is WAR5 + 1/2 career WAR. Which also has the advantage of scaling almost exactly the same as career WAR. When you look at these players, both JAWS and my HW agree that Heinie Groh was overlooked, and that Heinie Manush, Earle Combs, Red Scheondienst, and Rabbit Maranville were mistakes. However, JAWS also has Bobby Veach, Art Fletcher, and Tommy Leach as oversights and considers Hack Wilson as a mistake. HW has all of those as having been judged correctly, and adds Johnny Evers, Travis Jackson, and Edd Roush as mistakes. Evers is widely viewed as being in the Hall only because Tinker and Chance are, and I remember Jackson's selection being criticized. It's pretty clear that HW creates a much neater dividing line. The stronger emphasis on a shorter peak reflects the way voters have actually judged players. The two big outliers in HW are defensive specialists Maranville and Schoendienst. When you remove them, the regression becomes tremendously significant.
|
|
|
Post by rjp313jr on Jan 16, 2014 20:26:16 GMT -5
Eric since you're so smart I know you can answer these questions for me. Despite me being snarky since I thought your response was ignorant, I am interested in the answers from an educational point of view.
How does WAR calculate a players defensive value from before advanced defensive statistics? How can the data be reliable? When I look at Biggios outlier 9+ WAR season, I notice his defensive value that year is off the charts better then any other season. Why should I take that number seriously?
Is there a difference between UZR calculations for players in 2013 vs 1980? If so, why should defensive numbers be taken seriously when everyone says they are extremely flawed today let alone what you're trying to extrapolate from archaic information?
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 9,027
|
Post by ericmvan on Jan 25, 2014 17:52:20 GMT -5
Eric since you're so smart I know you can answer these questions for me. Despite me being snarky since I thought your response was ignorant, I am interested in the answers from an educational point of view. How does WAR calculate a players defensive value from before advanced defensive statistics? How can the data be reliable? When I look at Biggios outlier 9+ WAR season, I notice his defensive value that year is off the charts better then any other season. Why should I take that number seriously? Is there a difference between UZR calculations for players in 2013 vs 1980? If so, why should defensive numbers be taken seriously when everyone says they are extremely flawed today let alone what you're trying to extrapolate from archaic information? We have Putout, Assist, and Error totals for everyone who ever played. For many seasons, we have play-by-play data. TotalZone is a system that use PBP data to derive defensive metrics. It has a nice agreement with UZR and DRS, for the years when we have that data. There are at least four systems that use Putouts, Assists, and Errors to derive defensive numbers. There's a separate TotalZone method for those years. Bill James has a system (defensive win shares), Baseball Prospectus has their FRAA, and Michael A. Humphreys (who I had the pleasure of hanging out with at the NESSIS conference) has a great system in his book Wizardry. All of these systems, of course, do a lot of massaging of the raw numbers based on things like hits given up (obviously), the handedness of the pitching staff, and so on. These systems do a darn good job of matching our conceptions. Here are Humphrey's top 10 at various positions, in terms of career runs saved. CF: Andruw Jones, Speaker, Mays, Gary Pettis, Blair, Chet Lemon, Ashburn, Garry Maddox, Devon White, Willie Davis. 3B: Schmidt, Robinson, Buddy Bell, Jimmy Collins, Clete Boyer, Tim Wallach, Tommy Leach, Nettles, Terry Pendleton, Scott Rolen. SS: Art Fletcher, Tinker, Belanger, Bill Dahlen (19th century), Rey Sanchez, Ozzie Smith, Larkin, Jose Valentin, Garry Templeton, Neifi Perez. (He acknowledges that the system is too generous to deadball-era SS, and in his subjective ranking demotes Tinker and Fletcher to 4th and 5th. Ozzie, he points out, was not the same after he tore his rotator cuff in 1985. He had about as high a peak as anyone.) You should be able to go to B-Ref and tool at the TZ leaders for various positions and years, and I think you'll see that they make good sense. Some day, perhaps, we'll have a system that takes the best data-massaging ideas from all of these systems, and is even more trustworthy. Until then, I'm satisfied with using the defensive ratings at b-Ref and FanGraphs.
|
|
|