Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Saturday, January 12, 2013

Nate Silver: In Cooperstown, a Crowded Waiting Room

Bill Pecota had 538 career chances at 3B. Must ask Nate about this.

Between the protest voters on the one hand, and the maxed-out voters on the other, the players are being squeezed at both ends.

Finally, some players may be harmed by the psychology of the ballot. If Clemens were not on the ballot, for example, then you could credibly make a case that Curt Schilling was the best pitcher on the ballot (if you don’t think that Morris is). But Schilling’s accomplishments look poor by comparison to Clemens’s, as do those of almost any pitcher — even if you aren’t willing to vote for Clemens because of his steroids use. The same holds for outfielders whose statistics might be compared with Barry Bonds’s.

There is even something to be said for the so-called “paradox of choice”: that when presented with too many options, we may be overwhelmed with information and have trouble making any decisions at all.

Hall of Fame voting is ultimately designed to be a consensus process. One reason that players tend to gain votes over time is because the writers are looking at what their peers are doing and value the endorsements of their colleagues. Moreover, because they have as many as 15 chances to elect a player, many writers tend toward conservatism initially. There is no way to remove a player from the Hall of Fame once he has been elected, but you can change your mind to include him later. When a writer initially votes “no” on a player, it really means “wait and see” in many cases.

But consensus is harder to achieve when members of a group have divergent values and ideologies. Instead of the typical friendly arguments about how a player’s lifetime accomplishments might be weighed against how dominant he was in his best seasons, or how to compare players at different positions, the writers are now spending most of their time arguing about who used steroids and when, and how this should affect Hall of Fame consideration. Many have passionate beliefs about this, whichever side of the argument they take. An increasing number of writers would like to elect a dozen or more players; an increasing number would like to lose the whole “steroids era” to history. Good-natured debates may be replaced by tactical considerations, as voters make guesses about who everyone else might vote for, or where their ballots might be wasted.

Repoz Posted: January 12, 2013 at 03:41 PM | 22 comment(s) Login to Bookmark
  Tags: hof, sabermetrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Ray (RDP) Posted: January 12, 2013 at 04:27 PM (#4345702)
I just skimmed TFA, but I think one point Nate is making is that the typical player progression where he sees a steady increase in his vote totals over time (with 1.1 being the multiplier), slowing down after his first several years, and then perhaps with a final 15-year bump, might not continue going forward as the logjam continues to affect voting.

Quoting:

Based on an analysis of Hall of Fame voting between 1967 and 2011, I found that the increase in a player’s vote total is typically proportional to his percentage from the previous year. In his second year on the ballot, for example, the typical player’s vote share increases by a multiple of about 1.1.

Thus, a player who received 10 percent of the vote in his first year would be expected to receive about 11 percent on his second try, while a player who got 50 percent of the vote would go up to 55 percent.

The pace of improvement is typically highest in the first several years that a player spends on the ballot, slowing down once he has been eligible five or six times. (The exception is in a player’s 15 and final year of eligibility, when he may receive a fairly large boost.)


Thus, we saw a number of returning candidates see atypical drops (9 of the 13), including Smith, Trammell, McGriff, Mattingly, and Williams.

And I'm not sure if Silver would agree with this, but to me it means that we might not be able to expect that Biggio or Schilling will soon get in, to say nothing of the faux-steroids-tainted candidates such as Piazza or Bagwell or Sosa.

It's a loss for Bernie and perhaps Lofton, who may have fallen off of the ballot because of this and who deserved the chance for voters to consider their cases over a few more years. Perhaps Lofton wasn't going to make the 5% cut, but Bernie went from 9.6% to 3.3% and done.

So the damage, even to non-tainted players, is already being felt.
   2. snapper (history's 42nd greatest monster) Posted: January 12, 2013 at 05:14 PM (#4345713)
I just skimmed TFA, but I think one point Nate is making is that the typical player progression where he sees a steady increase in his vote totals over time (with 1.1 being the multiplier), slowing down after his first several years, and then perhaps with a final 15-year bump, might not continue going forward as the logjam continues to affect voting.

I don't think it will be that bad. The writers know they have to elect someone, or they'll eventually lose their vote (like previous VCs have).

The consensus will form around clean or "cleanish" players, and some will start getting in. Biggio had a great debut, he's a lock within the next 2-3 years (probably next year), Piazza had a very strong debut, and Bagwell made a little progress despite the deluge.
   3. Ray (RDP) Posted: January 12, 2013 at 05:20 PM (#4345717)
Piazza had a very strong debut, and Bagwell made a little progress despite the deluge.


But Bagwell is instructive for Piazza, isn't he? They're similar players in that they're viewed as suspicious for steroids even though there's been no actual evidence. Bagwell debuted at right around where Piazza did -- 56% for Bagwell, 57.8% for Piazza -- and yet Bagwell increased only to 59.6%.

I wouldn't be surprised to see Biggio slip down the grease pole a little.
   4. snapper (history's 42nd greatest monster) Posted: January 12, 2013 at 05:26 PM (#4345722)
But Bagwell is instructive for Piazza, isn't he? They're similar players in that they're viewed as suspicious for steroids even though there's been no actual evidence. Bagwell debuted at right around where Piazza did -- 56% for Bagwell, 57.8% for Piazza -- and yet Bagwell increased only to 59.6%.

I think increasing at all was a very good sign for Bagwell, given that you added 5 guys who are clear HoFers just on the numbers.

I would hope writers (especially those who don't care about PEDs) will start redirecting their votes to the qualified guys who can get in.
   5. Greg K Posted: January 12, 2013 at 05:43 PM (#4345732)
How about some kind of cumulative vote?

Each submitted ballot has to rank the top X% of retired players who have reached a certain threshold in the history of MLB (4000-5000 PA, 1500-2000 IP something like that) which matches the % (+ a little bit to give a bit of breathing room) of such players in the Hall before the new crop of retirees are included in the population. So I guess you don't necessarily have to rank all of them just however big that top X% group is. This can include current Hall of Famers. So theoretically you could vote for no one if you think all of the top X% players are already in the Hall of Fame.

Voting begins immediately after retirement, but election is only possible after five years at the earliest. A player's vote total is determined by where he falls in an averaged ranking of the ballots. In the top percentile gets you 20 points. Just scraping the bottom of the barrel gets you a small amount...2 or 3 say. So a top guy gets in after five years while a guy on the bubble has to wait a long time (longer if the arithmetic works against him and some new guy pushes him out of the X%). There is no limit to how long you stay on the ballot.

I suppose the relevant comments are
A) This would never happen
and
B) It's a lot of work for a ballot...but hey, this is important stuff. It SHOULD be a lot of work. And if it's not your idea of fun to rank the hundred or so best baseball players of all time then I'd say you shouldn't be deciding who's in the Hall of Fame.

and finally C)
This isn't so much a real suggestion as floating an idea...what are the major problems you guys see with a system like this? (Technical rather than practical...as in, do you think it will spit out terrible results? Not, would the Hall of Fame ever use it in reality.)

I do like the idea of guys gradually marching towards Cooperstown. A cumulative vote achieves that without the odd tendency to have votes gradually build over time (which I've never fully grasped as a phenomenon).
   6. snapper (history's 42nd greatest monster) Posted: January 12, 2013 at 05:49 PM (#4345734)
the odd tendency to have votes gradually build over time (which I've never fully grasped as a phenomenon).

I don't think it's at all odd when you look at the history of voting.

When they started in 1936, they had 50+ years of players to choose from, so only slam dunks got in on the first ballot. Perfectly deserving HoFers had to build gradually as the creme-de-la-creme got in. That tradition has simply persisted.
   7. ajnrules Posted: January 12, 2013 at 05:54 PM (#4345735)
Bagwell debuted at right around where Piazza did -- 56% for Bagwell, 57.8% for Piazza -- and yet Bagwell increased only to 59.6%.

Bagwell actually debuted in 2011 with 41%, so he's actually got quite a bit of momentum. I think Piazza is going to get in first, but Bagwell could get in as early as 2016 alongside Griffey, although 2017 may be more realistic.

I can also see Biggio getting in next year, quite possibly at the expense of a supposed "lock" like Thomas or Glavine.
   8. Walt Davis Posted: January 12, 2013 at 06:17 PM (#4345746)
Thus, we saw a number of returning candidates see atypical drops (9 of the 13), including Smith, Trammell, McGriff, Mattingly, and Williams.

But these were not atypical drops. If what you quoted is accurate, then Silver's model is simply wrong. Drops for the backlog, especially the deep backlog, are exactly what was expected.

It's not a complicated model. You have votes going off the ballot due to election or 15th year or <5%. You have votes coming onto the ballot -- these aren't known beforehand but you can usually have a good guess whether more are coming on than are going off. Then you have the number of votes per ballot -- also not known beforehand but you can have a not-so-good guess based on general trends and the quality of the ballot.

2012 had seen a record low for names per ballot, around 5.2 if memory serves, continuing a long trend. The 2011 ballot was only losing Larkin (86.4) and the <5% (7.3) so about .94 names per ballot freed up. Then you had the rush of new candidates coming on. The most optimistic estimates for the 6 big names looked something like: Biggio (75), Piazza (65), Bonds (50), Clemens (50), Schilling (40), Sosa (20) or 3 names per ballot. The most pessimistic estimates were pretty much what we got which was 2.51 names per ballot. Add in .09 for the new <5% crowd and you've got a range from about 2.6 to 3.1 names per ballot being added.

That looked like a blood bath for the backlog -- 2.6 minus .94 leaves 1.7 fewer names per ballot. Fortunately the total names per ballot grew by nearly as much, about 1.4. That's pretty typical -- you almost never see a jump in average names per ballot to fully absord the newcomers. If anything, this was better than usual.

That still leaves you short though and that means, on average, we expect everybody in the backlog to drop. Any gains among backloggers have to be balanced by even bigger drops for other backloggers. Guys high in the backlog are probably more stable than guys lower down -- the more guys who debut above you, the greater danger you are in.

That was all perfectly predictable based on historical trends and it's pretty much exactly what Dag predicted -- if anything he was more pessimistic as I think names per ballot took a bigger jump than he expected. I think I'm on record for small jumps for Bagwell, Morris and Raines and drops for Smith, Edgar, Walker, Trammell and McGriff although Edgar didn't. However all the newbies debuted at least a little lower than I expected, leaving more votes for others.

It's simple demographics -- a whole bunch of rich people moved into the neighborhood while only a handful moved out and they didn't build enough new houses to handle them all. The poorer folks in that neighborhood either have to get out (<5%) or share their votes with others.

FYI: Bagwell debuted at 42; 56 was his 2nd year total.
   9. Ray (RDP) Posted: January 12, 2013 at 06:19 PM (#4345747)
Bagwell actually debuted in 2011 with 41%,


Thanks to you (and Walt) for the correction.
   10. Arbitol Dijaler Posted: January 12, 2013 at 06:21 PM (#4345750)


When they started in 1936, they had 50+ years of players to choose from, so only slam dunks got in on the first ballot. Perfectly deserving HoFers had to build gradually as the creme-de-la-creme got in. That tradition has simply persisted.


Plus, it's not the Hall of Merit - it's legitimate for a writer on the fence to see a level of support for a player and reconsider whether the player should be celebrated in this way. Sort of a proxy for the player's overall importance in the game.

   11. Ray (RDP) Posted: January 12, 2013 at 06:24 PM (#4345751)
the odd tendency to have votes gradually build over time (which I've never fully grasped as a phenomenon).


Nate explains it this way:

Hall of Fame voting is ultimately designed to be a consensus process. One reason that players tend to gain votes over time is because the writers are looking at what their peers are doing and value the endorsements of their colleagues. Moreover, because they have as many as 15 chances to elect a player, many writers tend toward conservatism initially.


And it sounds odd, but as I consider and reconsider how to evaluate players, including being presented with things like WARP and WAR and the various defensive/baserunning metrics, I find that I've changed my mind about some players. Or are in the process of changing my mind. Players like Larry Walker and Edgar Martinez and even Dwight Evans I'm getting closer to supporting.

And some of my re-evaluation has been the result of a heightened appreciation I've gotten for peak/prime value with respect to HOF arguments. This explains my shifting views on Walker/Edgar. And could well lead to me supporting someone like Giambi in the future.

If I were a voter and players like Bernie/Lofton remained on the ballot you can be damned sure I'd continue to consider them seriously -- and might even support them some years down the line. (Though I would, under no circumstances, vote for them if I still found them wanting just to try to ensure that they'd stay on the ballot.)
   12. Walt Davis Posted: January 12, 2013 at 06:49 PM (#4345758)
I'll add to the above that the hardest variable to project in all of this is names per ballot. You know where it was last year, you have a pretty good idea of where it will have to be to maintain the backlog exactly where they are, you're pretty sure it will be somewhere in between those two numbers. But when those two numbers are far apart, it's hard to be very precise.

As to the future ...

it will continue to be hard for the backlog to make progress. This would have been true without steroids. It will be hard because you have so many great players coming onto the ballot. This year's ballot must set a record low for votes leaving: Murphy and the <5% crowd add up to only .3 names per ballot. The newcomers must project to something like 2.6 names per ballot. It seems impossible that we would see a jump in names per ballot to keep up (that would be an average of nearly 9). Any writer who throws a feel good vote to Shannon Stewart or even Luis Gonzalez or maybe even Jeff Kent should be tried for treason.

This also likely means that next year's newbies will have some "disappointing" results despite their "slam dunk" status. I can see Thomas anywhere from Biggio's 40% to induction. I doubt Mussina can do better than Schilling did this year. The lower backlog is gonna get wiped out -- Palmeiro's gone, I suspect Sosa is too.

Maddux sails in. But if they can push a second non-Morris player across, they might be able to keep their head above water. That and the <5% crowd would mean something around 2.5 names opening up. That should be enough to absorb Johnson/Pedro/Smoltz (assuming Sheff is ignorable). Assume Johnson and Pedro across (hard to see a 3rd person at this point if they get 2 across in 2014), "freeing up" about 1.7 votes which is more than enough to absorb Griffey with room left over for some backlog growth in 2016 (although we will likely see a small drop in names per ballot in 2016).

If they can push a second person across in 2016 there's lots of room for major backlog progress in 2017 although Pudge and Vlad (and not Manny) will eat up a good chunk of that. If they can only get Griffey across though I've got to admit that 2017 looks pretty ugly and could end up a lot like this year -- which is not what I've argued earlier. I'm guessing somebody manages to get across but it will be a jumble of guys in the 50s and 60s like this year sort of election.

So they need to push at least 6 guys across in the next 3 years. And they really do need to extend the ballot to more than 10 names -- that will make it more likely that names per ballot can grow to 9+ to absorb the onslaught. I would also put greater pressure on the selection committee to stop passing through the truly non-viable. Alomar, Franco, Finley, Green and Sele ate up .05 names per ballot. I suppose you can argue those morons would have voted for nobody in that slot instead, so freeing them up wouldn't matter but it's counterproductive to even give them an option for wasting votes like that. Guys like Alou and Durham were very good players but you know they're going nowhere.
   13. Ray (RDP) Posted: January 12, 2013 at 07:39 PM (#4345781)
Any writer who throws a feel good vote to Shannon Stewart or even Luis Gonzalez or maybe even Jeff Kent should be tried for treason.


Walt, as it so happens I just took a close look at Kent for the HOM thread. Your comment seems quite extreme. I think Kent is a HOFer. Certainly he's very arguable.

What are your particular problems with his case? I imagine defense and durability.

This is what I posted in the HOM thread:

I just took another look at his career per b-r. I support him. Wasn't particularly healthy, but managed a 2200-game career, including 2,000 games at 2B. Very nice peak, and excellent offense at the position: a 123 career OPS+, with highs of 162, 147, 142, 133, 131. He takes a hit according to b-WAR on defense.

What struck me though in reviewing his career line was that he was hitting well for a 2B even outside of his peak. Never really had a bad year with the bat. Bulk/average value helps build a HOF case for me.

His b-WAR is relatively low because of the defense - just 51 b-WAR - but I use WAR as one part of my inquiry, not as the be-all, end-all. So I support him. I suppose my mind could be changed; one could make some hay arguing against him with defense and durability.

Had absolutely no recollection of him finishing his career with two years for the Astros and four years for the Dodgers.

   14. DL from MN Posted: January 12, 2013 at 09:43 PM (#4345862)
I think Jeff Kent has the most HR for a second baseman. That's not a "feel-good" vote. Aaron Sele is a "feel-good" vote.

I don't think Biggio is one of the top 10 guys for next year. Hypothetically, should I vote for him anyway?
   15. calhounite Posted: January 12, 2013 at 09:53 PM (#4345873)
Few observations ...

The known roider's got about as much chance of being exhibited in honorary section of HOF as a pedestrian in an auto show. Just ain't happening. Would have a better chance of getting in the auto show.

Hanging on for too long especially changing teams every year sending off the signature aura of a journeyman is a sure way to put the immediate kabosh on a perfectly good HOF candidacy.

If a good HOF candidacy is not linked directly to roids, even if suspected, will get in but will take a few years.

Even those not suspected ... may take a few years.
   16. Walt Davis Posted: January 13, 2013 at 12:52 AM (#4345989)
I think Kent is a HOFer

I think he's at least bloody close. I also think he's probably something like 17th on next year's ballot and I'm not sure the voters will treat him even that well. He was my "maybe even". Under normal circumstances he might have a respectable 20% or something, maybe even better. As it is, he might be lucky to clear 5% although the most HR by a 2B and the fight with Bonds might help.

I mean, on a relatively weak ballot, Bagwell started at 41%. Trammell started out pretty low. Walker and McGriff are in the low 20s. The voters aren't usually that kind to 2B. And they've got a ton of other choices. We know what they did to Lofton this year, Bernie this year, Whitaker years ago.

He's a perfectly reasonable candidate ... and a vote for Kent will almost certainly be a wasted vote. It's in that sense that maybe even a vote for Kent should be frowned upon. It's a comment on how packed next year's ballot is.

There are alternatives. The best is to expand the ballot -- then a vote for Kent probably won't be wasted. The other is to dump other pretty hopeless cases like Palmeiro, Sosa, Mac, Walker and McGriff. I haven't looked at it too closely but I think I put him behind all those guys in terms of worthiness.

OK, looked it up, he's 18th in career WAR on next year's ballot, ahead of McGriff. Of the guys ahead of him, I guess you could make a non-crazy case to put him ahead of Sosa or Biggio or Edgar but that's about it. OK, then toss Mac, Palmeiro, Bonds and Clemens and you can almost get him to the top 10. And of course Morris and Mattingly are eating up votes for one more year.
   17. Dr. Chaleeko Posted: January 13, 2013 at 04:20 PM (#4346255)
Agreed, Raffy and Sosa are done. And the two Macs are going to see big drops next year.

But I'd hate to be Lee Smith. His support isn't likely firm enough to withstand Maddux, Glavine, Thomas, Moose (and maybe Kent). With three to five newbies potentially taking his spot, so he's in peril with any voter who has listed 7 or more names this year. In addition, anyone who doesn't vote for first-ballot players may add Schilling or Piazza or Biggio to their ballot. (And Maddux is likely not subject to that penalty, not sure on Glavine and Thomas).

I personally wouldn't be at all shocked to see Smith lose half or more of his current support and go from a 50%er two years ago to a 15%er.
   18. John DiFool2 Posted: January 13, 2013 at 06:17 PM (#4346366)
The other is to dump other pretty hopeless cases like Palmeiro, Sosa, Mac, Walker and McGriff.


A 3,000 hit/500 homer guy and a 600 homer guy are both "hopeless cases".

Re-read that sentence again, very slowly. It's a mad mad mad mad world.
   19. snapper (history's 42nd greatest monster) Posted: January 13, 2013 at 07:00 PM (#4346393)

A 3,000 hit/500 homer guy and a 600 homer guy are both "hopeless cases".

Re-read that sentence again, very slowly. It's a mad mad mad mad world.


Except, in the era they put up that production, those numbers don't make them inner circle guys (like a Bonds or Clemens).

Sosa particularly. He only has 54.8 WAR. The guys near him on the list include a lot of non-HoFers. Even Palmeiro at 66 WAR isn't a "can't imagine a HoF w/o him" guy.
   20. Mess with the Meat, you get the Wad! Posted: January 13, 2013 at 07:04 PM (#4346399)
I put Sosa in as a peak kind of guy
   21. Dr. Chaleeko Posted: January 13, 2013 at 08:50 PM (#4346460)
Palmeiro : 1990s :: Murray : 1980s

Beckley : 1890s :: Start : 1870s

Palmeiro & Murray : 20th Century :: Beckley & Start : 19th Century
   22. Ray (RDP) Posted: January 14, 2013 at 12:03 AM (#4346540)
Yes, Palmeiro is Murray. And nobody questions Murray's HOF worthiness, to my knowledge. That's what's frustrating about the comments that Palmeiro (not considering the steroids issue) is not worthy.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Forsch 10 From Navarone (Dayn)
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogHall of Fame Announces Changes to Voting Process for Recently Retired Players, Effective Immediately
(84 - 6:38am, Jul 28)
Last: Sunday silence

NewsblogSchoenfield: Why didn't the Braves win more titles?
(50 - 6:23am, Jul 28)
Last: Jolly Old St. Nick Still Gags in October

NewsblogFull Count » Mike Carp, Felix Doubront and the challenges of player discontent on a struggling team
(7 - 6:19am, Jul 28)
Last: Jim Furtado

NewsblogOTP - July 2014: Republicans Lose To Democrats For Sixth Straight Year In Congressional Baseball Game
(3292 - 6:14am, Jul 28)
Last: Jolly Old St. Nick Still Gags in October

NewsblogDJ Short: Maximum stay on Hall of Fame ballot changed from 15 to 10 years
(57 - 6:05am, Jul 28)
Last: Sunday silence

NewsblogDodgers and Diamondbacks Triple-A teams involved in wild brawl
(12 - 5:43am, Jul 28)
Last: Robert in Manhattan Beach

NewsblogFull Count » Mike Carp explains why he requested a trade from Red Sox
(18 - 3:18am, Jul 28)
Last: ellsbury my heart at wounded knee

NewsblogGossage on Bonds, McGwire Hall hopes: ‘Are you f–king kidding?’
(106 - 3:02am, Jul 28)
Last: Walt Davis

NewsblogRoger Angell goes into the Hall of Fame
(28 - 12:49am, Jul 28)
Last: bobm

NewsblogOT: Monthly NBA Thread- July 2014
(967 - 12:31am, Jul 28)
Last: clowns to the left of me; STEAGLES to the right

NewsblogGiants purchase contract of 2B Uggla
(8 - 12:09am, Jul 28)
Last: Rickey! On a blog from 1998. With the candlestick.

NewsblogOMNICHATTER 7-27-2014
(145 - 11:46pm, Jul 27)
Last: Los Angeles El Hombre of Anaheim

NewsblogHoVG: John Rocker Shows Up in Cooperstown… “Survivor” Up Next
(31 - 10:52pm, Jul 27)
Last: depletion

Hall of MeritMost Meritorious Player: 1956 Ballot
(7 - 9:28pm, Jul 27)
Last: neilsen

SABR - BBTF ChapterWho's going to SABR??
(100 - 6:40pm, Jul 27)
Last: Scott Fischthal

Page rendered in 0.4232 seconds
52 querie(s) executed