Peter Keating of ESPNtheMag: Agrees w/ Blake

Announcements, Ballots, Voting Results and Other Assorted Stuff.
User avatar
cheekimonk
BBA GM
Posts: 5327
Joined: Sat Oct 23, 2010 6:46 pm
Location: Birmingham, AL
Has thanked: 157 times
Been thanked: 128 times
Contact:

Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by cheekimonk » Tue Jan 10, 2012 10:12 am

Peter Keating published an article in this month's ESPNtheMag about baseball HoF criteria (an excerpt is here and you can find the entire article if you're an Insider) where he basically tackles guys who were truly outstanding vs. guys who were just above-average but for a very long career. He lands on WAR as a perfect "comparison" stat because, well, that's the whole point: it measures how much better (how many more wins the player is worth) a player is than "a benchwarmer or a Triple-A guy."

But he says it's problematic now that it's being used for MLB HoF candidates. WAR is a cumulative stat, so over a 15-year career a player's WAR is simply the sum of all his 15 WARs. As Keating points out, "Bob Boone...finished with only 10.1 fewer WAR (26.1) than Roy Campanella by virtue of playing nine more years. Campanella exceeded 5.0 WAR in each of the three MVP seasons of his 10-year career, while Boone's best single mark was 3.5 WAR..."

So he has proposed a stat called Wins Above All-Star Level (WAAS) where he basically sets a minimum threshold for a marginal All-Star's WAR in a given season. For MLB he pinpoints it at 2.5 WAR. So a player's WAAS for each season would be the difference between their WAR and 2.5 (subtract 2.5 from their WAR). For a player's career you do that for every season, toss out negative seasons (since WAAS is only measuring the player's peak performance), and then add them together. Using that, Keating writes, "Sandy Koufax has 54.5 WAR, fewer than Frank Tanana (55.1 WAR) who pitched 1,800 more innings. But Koufax has 32.6 WAAS as compared to Tanana's 19.7."

I want to put together some numbers for MBWBA's HoF if I can, but I think it's a pretty simple and smart adjustment for HoF purposes.
Ben Teague, GM Boise Spuds
2682-3175, .457 PCT (5,857 games, 36 seasons)
11 Playoff Appearances, 1 Championship

Former BBA GM: Many (Monty Brewster Memorial Series champion: 1997)
Former GBC GM: Jerusalem, Buenos Aires


Boise Home Page (roster, prospects, etc.)

Al-Hoot

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by Al-Hoot » Tue Jan 10, 2012 10:24 am

wow...my mind glazes over...
j/k

Interesting stuff. I still haven't figured WAR out yet. Aren't there different formuli for calculating it?

Anyway, the new WAAS is an interesting concept.

Except I think the Hall of Fame benefits by having some "flavor" in it: a lot of people argue that such players as Koufax and Dizzy Dean don't belong in it, because of the brevity of their careers. But both those pitchers were absolutely brilliant. Yeah they might have blazed a lot shorter than other pitchers,... but their impact on baseball can never be forgotten. (Not that that in itself makes them HOF candidates). But I don't think there is one right system, nor one right criteria to base HOF selection on. And so the arguments will go on and on...

I spoke to the daughter of a MLB pitcher who some feel should be in he MLB HOF, and she says that whenever her dad was asked about this, he just shrugged it off and said It [who gets into the HOF and who doesn't] is "all political" anyway. Now this is a pitcher from the 30s and 40s, but I doubt things have changed much.

Gosh, did Blyleven finally get in? Bully for him I think he's deserving. Jokes aside.

Al-Hoot

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by Al-Hoot » Tue Jan 10, 2012 11:16 am

Two more thoughts:

1. "For a player's career you do that for every season, toss out negative seasons (since WAAS is only measuring the player's peak performance), and then add them together."

well, that' an arbitrary decision right there that not everyone might agree with. Shouldn't every season be taken into account.

2. Longevity in itself counts for something. Since baseball is a physical sport requiring players to stay in condition, those great players who can do that for two decades should have that taken into account, even if they weren't the absolute best during some of those years--some might argue. ;)

Like I said, a HOF can have different flavors to it.

jcrmoon42
Ex-GM
Posts: 1559
Joined: Thu Nov 18, 2010 5:51 pm
Has thanked: 1 time
Been thanked: 8 times

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by jcrmoon42 » Tue Jan 10, 2012 11:41 am

Interesting stuff. In my opinion, the truth is somewhere in the middle, as it usually is. Some players with long careers who accumulate stats are deserving of induction. Some players who were truly great for short periods are deserving of induction. It seems, again in my opinion, that too many land on the definition that a Hall of Famer is someone who is truly great for a long time. The problem with this opinion is that the definitions of "truly great" and "a long time" are completely arbitrary. What makes a long time? Is it 20 years? 15? 12? It is totally made up in the mind of the person doing the defining.

One of the guys on MLB Network who votes uses WAR, but he combines career and peak, much like Bill James does with his Win Shares rankings. This seems a very reasonable way of doing it, because it absolutely recognizes the truly great while also giving credit for those who make short-term greatness and long-term consistency arguments.

One of these days, when time allows, I will do some of that stuff for the MBBA like I have done for the ACB and other leagues before.

User avatar
cheekimonk
BBA GM
Posts: 5327
Joined: Sat Oct 23, 2010 6:46 pm
Location: Birmingham, AL
Has thanked: 157 times
Been thanked: 128 times
Contact:

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by cheekimonk » Tue Jan 10, 2012 12:05 pm

On WAR: there is only one formula now that the stat is widely accepted (and I think an official MLB stat, if I'm not mistaken). There are stat geeks that came up with stuff in their basements over beers in the 70s, but are now full-fledged businesses with proprietary formulas for what they provide. They come up with all kinds of stuff and sell it to teams, media, other geeks, etc. and part of what they ALL do is come up with their "improvements" on non-sacred statistics (i.e., Sabers mostly) and slap a different acronym on it.

WAR basically leverages other micro-stats in Sabermetrics to get a measure of a players actual contribution on offense and defense. I think most people get confused because WAR is a comparative stat. That means you can't project it (unlike batting average where I can guess how many hits a guy will have in a certain number of at bats) because: it doesn't exist until it happens and it is only meaningful by comparison. So saying a player has a WAR of +0.8 really doesn't tell you anything at all except that he's somewhere above average for a guy playing his position and as often as he does. But saying that the player has a WAR of +0.8 and a guy your team is looking to trade for has a WAR of +1.5 is very meaningful. That's the value to me because it's easy for us to look at a slugger with 1.000+ OPS and a .250 hitter with speed and compare the guys. But what if the slugger is a marginal first baseman and the .250 hitter is Ozzie reborn at SS? How do you compare that? WAR takes it all into account and gives you one number vs. another number.

I think, too, that people get confused by "wins" in that they try to say that a player with +12 WAR was responsible for 12 of his team's wins. That could be argued, but there's no need...just think of it as a number that is meant to be compared to similar numbers. If a guy with +10 WAR goes down for the year to injury then you need to be out looking to trade or sign an FA or you're hosed. But if a guy with +0.3 WAR goes down, you could probably call up a rookie or vet from the minors and not lose too much.
Ben Teague, GM Boise Spuds
2682-3175, .457 PCT (5,857 games, 36 seasons)
11 Playoff Appearances, 1 Championship

Former BBA GM: Many (Monty Brewster Memorial Series champion: 1997)
Former GBC GM: Jerusalem, Buenos Aires


Boise Home Page (roster, prospects, etc.)

User avatar
cheekimonk
BBA GM
Posts: 5327
Joined: Sat Oct 23, 2010 6:46 pm
Location: Birmingham, AL
Has thanked: 157 times
Been thanked: 128 times
Contact:

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by cheekimonk » Tue Jan 10, 2012 12:31 pm

Al-Hoot wrote:Two more thoughts:

1. "For a player's career you do that for every season, toss out negative seasons (since WAAS is only measuring the player's peak performance), and then add them together."

well, that' an arbitrary decision right there that not everyone might agree with. Shouldn't every season be taken into account.

2. Longevity in itself counts for something. Since baseball is a physical sport requiring players to stay in condition, those great players who can do that for two decades should have that taken into account, even if they weren't the absolute best during some of those years--some might argue. ;)

Like I said, a HOF can have different flavors to it.
1. Well, remember this is only intended by Keating to be used in separating HoF-caliber players from just great players. His central point (explained more in the article) is that the greatest players of all time are defined by their superior seasons. But a really good player can be just below all-star level for 16 years and have cumulatively similar, or maybe even better, stats than the player with 6 legendary seasons. The point is to cull out decent or even poor seasons because he's specifically talking about HoF worthiness and not which player to trade for or sign.

2. This goes to the point I made to Blake in the long HoF thread. What does "Fame" mean? Keating talks about that 2.5 WAAS threshold and says it could be tweaked up or down depending on how big you think a "Hall of Fame" should be. The Alabama Sports Hall of Fame, for instance, is HUGE because it has people like Vonetta Flowers who was a track & field athlete at UAB (in Birmingham) and was the first black woman to win a gold medal at the winter olympics, as well as people like Charles Barkley, Hank Aaron, and "Bear" Bryant. If you follow the criteria you're mentioning then the HoF is going to be quite large and being a member won't mean the same as a Hall based on someone else's stricter formula (being a member of the Alabama Sports Hall of Fame means something, but not THAT much compared to being in Cooperstown, Canton, or even the Rock & Roll HoF). As comparison, what does "MVP" mean? Is that supposed to be "best player" or is it supposed to be "most critical player"? And if it's "most critical player" then is it "most critical" for a playoff team or "most critical" for any team?
Ben Teague, GM Boise Spuds
2682-3175, .457 PCT (5,857 games, 36 seasons)
11 Playoff Appearances, 1 Championship

Former BBA GM: Many (Monty Brewster Memorial Series champion: 1997)
Former GBC GM: Jerusalem, Buenos Aires


Boise Home Page (roster, prospects, etc.)

User avatar
LambeauLeap
Ex-GM
Posts: 1677
Joined: Mon Nov 22, 2010 10:54 pm
Been thanked: 2 times

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by LambeauLeap » Tue Jan 10, 2012 12:44 pm

I prefer the new B.O.O.B formula.
Brad Browne
Editor, Guam Today
---
1986: Chicago Black Sox (73-89)
1987-1991: Valencia Stars/Suns (341-469)
1998-2005: Austin Riverbats/Marquette Suns (697-600)

User avatar
salas
Ex-GM
Posts: 823
Joined: Tue Feb 22, 2011 10:25 am
Been thanked: 3 times

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by salas » Tue Jan 10, 2012 2:31 pm

bateague wrote:Peter Keating published an article in this month's ESPNtheMag about baseball HoF criteria (an excerpt is here and you can find the entire article if you're an Insider) where he basically tackles guys who were truly outstanding vs. guys who were just above-average but for a very long career. He lands on WAR as a perfect "comparison" stat because, well, that's the whole point: it measures how much better (how many more wins the player is worth) a player is than "a benchwarmer or a Triple-A guy."

But he says it's problematic now that it's being used for MLB HoF candidates. WAR is a cumulative stat, so over a 15-year career a player's WAR is simply the sum of all his 15 WARs. As Keating points out, "Bob Boone...finished with only 10.1 fewer WAR (26.1) than Roy Campanella by virtue of playing nine more years. Campanella exceeded 5.0 WAR in each of the three MVP seasons of his 10-year career, while Boone's best single mark was 3.5 WAR..."

So he has proposed a stat called Wins Above All-Star Level (WAAS) where he basically sets a minimum threshold for a marginal All-Star's WAR in a given season. For MLB he pinpoints it at 2.5 WAR. So a player's WAAS for each season would be the difference between their WAR and 2.5 (subtract 2.5 from their WAR). For a player's career you do that for every season, toss out negative seasons (since WAAS is only measuring the player's peak performance), and then add them together. Using that, Keating writes, "Sandy Koufax has 54.5 WAR, fewer than Frank Tanana (55.1 WAR) who pitched 1,800 more innings. But Koufax has 32.6 WAAS as compared to Tanana's 19.7."

I want to put together some numbers for MBWBA's HoF if I can, but I think it's a pretty simple and smart adjustment for HoF purposes.
Frank Tanana's not the one he should be using as an example....should be Jack Morris, since this whole issue applies most currently and most directly to him!!

jcrmoon42
Ex-GM
Posts: 1559
Joined: Thu Nov 18, 2010 5:51 pm
Has thanked: 1 time
Been thanked: 8 times

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by jcrmoon42 » Wed Jan 11, 2012 12:07 am

WAR being a comparative stat is exactly what makes it useful in a HOF discussion.

I'm certainly not a WAR expert as it is a relatively unknown stat to me, but if it uses the same idea as most replacement level stats, it shouldn't move from year to year by much at all. It may move over time, but not by much. The "replacement level" remains fairly consistent. Because of that, you should be able to look at a player with a 0.8 WAR and get a pretty good sense of his value, and it should have a connection to actual wins. At least, that is part of what I love about Win Shares, the connection to wins.

By the way, I've said it before, but I want to make sure that everyone is aware that the system we use in the MBBA puts almost exactly the same number of players in the Hall on a year by year basis as the MLB Hall of Fame does, including the Veteran's committee. It isn't built in such a way as to define what a Hall of Famer is, but to attempt to put a similar number of players in per year, allowing the voters to decide who those people are.

User avatar
cheekimonk
BBA GM
Posts: 5327
Joined: Sat Oct 23, 2010 6:46 pm
Location: Birmingham, AL
Has thanked: 157 times
Been thanked: 128 times
Contact:

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by cheekimonk » Wed Jan 11, 2012 1:43 am

jcrmoon42 wrote:WAR being a comparative stat is exactly what makes it useful in a HOF discussion.

I'm certainly not a WAR expert as it is a relatively unknown stat to me, but if it uses the same idea as most replacement level stats, it shouldn't move from year to year by much at all. It may move over time, but not by much. The "replacement level" remains fairly consistent. Because of that, you should be able to look at a player with a 0.8 WAR and get a pretty good sense of his value, and it should have a connection to actual wins. At least, that is part of what I love about Win Shares, the connection to wins.

By the way, I've said it before, but I want to make sure that everyone is aware that the system we use in the MBBA puts almost exactly the same number of players in the Hall on a year by year basis as the MLB Hall of Fame does, including the Veteran's committee. It isn't built in such a way as to define what a Hall of Famer is, but to attempt to put a similar number of players in per year, allowing the voters to decide who those people are.
Do you mean a player's WAR shouldn't move from year to year? It's actually just the opposite. Not that it should swing from positive to negative typically, but it would be statistically improbable for a player to generate the same WAR every season...mostly because WAR really does only apply to one season (hence the reason a "career WAR" is cumulative and not derivative...unlike batting average that I can recalculate based on hits and ABs for a career, month, week, 3 years, etc., but that I can't just add or average over a given time period). Now a league's WAR - total and baseline (baseline == the production that is considered to be 0.0) - should be pretty consistent over numerous seasons even though it still only applies to one season. That's true mainly due to the suppression of deviation by the huge sample size.
Ben Teague, GM Boise Spuds
2682-3175, .457 PCT (5,857 games, 36 seasons)
11 Playoff Appearances, 1 Championship

Former BBA GM: Many (Monty Brewster Memorial Series champion: 1997)
Former GBC GM: Jerusalem, Buenos Aires


Boise Home Page (roster, prospects, etc.)

jcrmoon42
Ex-GM
Posts: 1559
Joined: Thu Nov 18, 2010 5:51 pm
Has thanked: 1 time
Been thanked: 8 times

Re: Peter Keating of ESPNtheMag: Agrees w/ Blake

Post by jcrmoon42 » Wed Jan 11, 2012 7:05 pm

No, I meant the baseline shouldn't move that much. In other words, 1.0 WAR one year is roughly equivalent to 1.0 WAR the next year. Replacement level doesn't change much.

Post Reply Previous topicNext topic

Return to “Voting Chatter”

Who is online

Users browsing this forum: No registered users and 5 guests