Wednesday’s Hall of Fame results todayby Chris Jaffe
January 04, 2010
It's that time of year again. This Wednesday the Hall of Fame announces the results of the BBWAA election for new immortals. For me, that means I make my annual predictions for what the numbers will be in the Cooperstown announcement.
I first predicted the vote in 2008 based on a series of guidelines I'd created based on my analysis of voting trends in the history of BBWAA elections. I guessed specific percentages for everyone on the returning backlog and all new candidates likely to get over 5 percent of the vote. I did a good job, with an average margin of error of 4.3 percent. In comparison, Repoz (head linker at the Baseball Think Factory) tracked every BBWAA member who published his ballot online and was off by an average of 4.7 percent. Plus I successfully predicted Gossage would go in but Rice just miss being named on the 75 percent of ballots needed for election.
Last year, I did even better, with an average margin of error of 3.2 percent in my predictions of the top 15 candidates. Also, I again correctly figured that Rickey Henderson and Rice would go in.
So now I'm back for Round Three, and I must say - it's a good thing I've got two successful predictions under my belt because this could very easily be the year I screw up royally. I have far less faith in my predictions than in previous seasons. My guesswork will be wilder this year than in years past.
To understand why the 2010 ballot is so tough to predict, it's necessary to look over my guidelines for predicting how people will do. I have 10, which I previously discussed in detail, and so will only provide a quick summary here.
Guideline 1: Consistency
This guideline's simplicity matches its importance. Want to know how someone from the backlog will do in an upcoming election? Look at how they have done. All other things being equal, expect them to end around the same spot.
Guideline 2: Strength of ballot
That said, all other things aren't always equal. The single biggest factor explaining shifts is the strength of the ballot. When a bunch of strong candidates first arrive on a ballot, the vote totals for the backlog go down. When a weak crop emerges, their vote total goes up.
A typical rookie crop garners about 1.6 votes per BBWAA ballot and this year's rookie crop, featuring Roberto Alomar, Barry Larkin, Fred McGriff, and Edgar Martinez, looks like it should easily beat the 1.6 average.
Guideline 3: Comparable candidates
If a new candidate is directly comparable to someone on the backlog, that can have an unusually pronounced impact on the backlogger's vote. The best example of this came in the late 1980s when the arrival of Jim Palmer and Fergie Jenkins caused vote totals for Luis Tiant, Mickey Lolich and Jim Bunning to plummet.
There is one good example this year of comparable candidates: Barry Larkin versus Alan Trammell. My hunch is that Trammell will lose ground this year.
Alternately, it can only help Tim Raines that Rickey Henderson is off the ballot.
Guideline 4: The "Over the top!" surge
Once a candidate cracks the 50 percent barrier, he becomes a central focus of attention and writers start looking for reasons to include him on their ballot. Simply put, the logs start rolling in his favor.
As a general rule of thumb, guys near the top of the ballot experience the biggest gains in their vote total.
Last year actually had one of the most tepid such surges in history. I'm not fully sure why, but my hunch is that's because there was an unusually loud cry against the guy on top of the ballot, Jim Rice. His surge shrank and that affected Dawson and Blyleven. I'll need a greater sample size than one election before I modify this rule.
Actually, last year's ballot was strange in other ways. The entire backlog scarcely moved, which was especially odd given that it came one year after an election with the lowest total of names/ballot in BBWAA history. My personal guess is that the limited over-the-top surge caused less movement across the ballot as a spillover effect. It ain't the most rational of reasons, but that's how crowd psychology goes sometimes. I assume things go back to normal this year.
Guideline 5: Primordial conversations
A lot of the guidelines work because the Hall has a 70-plus year history built up that helps define what is and isn't a Hall of Famer. Some times the history isn't as important, which causes all the rules to be a bit shakier. In particular, there are two concerns: 1) relief pitchers (in 2008 I called this rule "Relief pitcher wackiness"), and 2) steroid cases.
Relievers Rich Gossage and Bruce moved up from unusually low points to eventual induction. Mark McGwire is the first real steroid case. His 2009 vote total was virtually identical to his 2008 level. We'll see if that continues.
Guideline 6: Last year on the ballot
Players in their final year of consideration usually receive a small bump in support. This year, no players are in their final year so it doesn't apply.
Guideline 7: Candidates per ballot
Here's where things start to get fun for predicting 2010's ballot.
In recent decades, the BBWAA has placed fewer names per ballot than at any previous point in their 70-plus years of electing. The 18 lowest average names-per-ballot have all come in the last 19 elections. (The exception, if you're curious, came in 1999 when Nolan Ryan, George Brett, Robin Yount, and Carlton Fisk all debuted). The BBWAA never fell below six candidates per ballot until the 1990s, but three of the last four elections have been below 5.7 names/ballot.
The last two elections had the two smallest averages in BBWAA history: 5.35 in 2008 and 5.38 in 2009. Those elections were low even by modern standards, but the BBWAA hasn't topped 6.60 this millennium and last broke seven names per ballot in 1986.
Yet this makes predicting 2010 a bit odd. As noted above, last year the BBWAA averaged only 5.38 names per ballot. Well, that included Rice and Henderson, who were elected and thus are no longer on the ballot. It also included Tommy John, who ran out of time, and guys like Mark Grace who received less than 5 percent of the vote and thus fell out of the race.
This year's returning backlog received 3.24 votes/ballot - which has got to be the smallest total in the history of the backlog. I suppose that makes sense, since there are only 11 men in the backlog, also a record low.
Given that, it's actually possible that the rookie crop of hopefuls could be better than average and the entire backlog moves up. That rarely happens, but there's room for it this year. (In fact, both could happen and there could still be a record low names per ballot, but I'm skeptical that will actually occur.)
I expect the backloggers to rise up overall, but how much is where the guessing game begins.
Guideline 8: Repoz's BBWAA tallying
For at least the third straight year, Repoz, the chief linker at the Baseball Think Factory, has collected every BBWAA ballot posted and adds them up, which provides some nice advantages.
First, this is the only source I have for guesstimating new candidates, as the rest of this list deals with looking at how people did in previous votes. The results aren't perfect (this isn't a scientific sample, just a self-selecting one of BBWAA members who publicize their vote), but it gives us something to work with.
This year, I'm leaning on Repoz's tally a bit more than in years past partially because there is a deeper crop of newbies (with Alomar, Larkin, McGriff, and Martinez all having reasonable cases that can be made on their behalf), but also because it's so difficult to figure what will be the average names per ballot.
As I write this, he's collected 73 ballots, with an average 6.3 names/ballot. I find that shockingly high. In part, but only in part, because all the newbies are doing better than I would've guessed. I don't think the average will be as high as 6.3, but I'll make my estimate higher than I otherwise would.
Actually, since Repoz has been doing this a while and it's mostly the same ballots being counted, his tally does help with the returnees. For example, in past years his tallies have always been low on Lee Smith's results. Alternately, they've exceeded the final results for Jack Morris, Trammell, Raines, and especially Bert Blyleven.
That said, there is a downside to using the tally in my predictions. (And this really hurts me when I try to do my predictions), all the above estimates are based on the full tally, which isn't completed until the eve of the BBWAA announcement, whereas I obviously have to get this column done several days earlier. The results of a three-quarters tally can be quite different than the full tally at times. Still, the results are so much higher than I would've guessed it would be wise for me to lean on this.
Guideline 9: Beware 5 percent
Players with no chance to be inducted are likely to be dropped by their supporters once the ballot gets a bit too crowded for them. People that spend a while near the cut-off marker are good bets to fall off the ballot eventually.
Guideline 10: Guidelines ain't laws
Tendencies ain't written in stone, so use your judgment. This is nice because, as noted above, some of the guidelines are clashing this year.
In the three years I've been doing this prediction, this is by far the trickiest. Three guys could get elected this year in Blyleven, Dawson, and Alomar. That's what Repoz's tally says. Then again, he's been high on Blyleven before and who knows if he's high on Alomar.
It's worth noting that it's tough to get elected in the first try without some special hook. The most obvious hook is someone with 3,000 hits or 500 homers or 300 wins. Of the 18 players elected on their first try since 1991, all but three belonged to one of those magical categories. Even the remaining trio - Kirby Puckett, Ozzie Smith, and Dennis Eckersley - all had their own little hooks. People considered Smith the greatest defensive player ever. Eck was the ultimate closer. Puckett's was a bit different - he had the sad story of a guy who clearly would've gotten to 3,000 hits had it now been for a tragic eye ailment.
Alomar doesn't quite have that hook. But he could make it anyway. After all, the backlog is record-breakingly small.
This prediction is tough for me because the newbies are much tougher. In 2008-09, I had one slam dunk inductee in Henderson who was obviously going to get well over 90 percent of the vote. Aside from him, I've only had one newbie get over 5 percent, Raines in 2008. Let the record show my prediction for him that year was my worst ever: I guessed 45 percent and the BBWAA gave him 24 percent.
This year, I have four guys falling in that Raines-land beneath Henderson and above 5 percent: Alomar, Larkin, Martinez, and McGriff.
Well, that's why it's tough for me, but this column is about predicting Wednesday's vote, not just dithering. Below are my guesses:
Name 2010 2009 Dawson 84 67 Alomar 82 XX Blyleven 73 63 Larkin 60 XX Smith 52 45 Morris 51 44 Raines 36 23 Martinez 35 XX McGwire 30 22 McGriff 24 XX Parker 18 15 Trammell 17 17 Mattingley 15 12 Murphy 15 12 Baines 6 6 Others 4 XX
Dawson's going in. I might have the percentage wrong, but he'll go in. I expect an unusually strong surge in part because he's so similar (and superior) to Rice, who just went in last year. As it happens, this is the only prediction I feel confident about.
Blyleven could go in this year, but I think next year is a bit more likely. The BBWAA has simultaneously elected three men only once in their last 18 elections.
Aside from them, I gave the biggest bounce in the backlog to Raines. Mostly that's due to the lack of Henderson on the ballot. It also reflects Repoz's tally and that he's got unusually vocal support for someone so far down the backlog, and that never hurts.
Before seeing Repoz's tally, I figured Alomar would debut at 60-65 percent, but he actually has Alomar as the top vote-getter overall. Maybe he will beat Dawson, but it's tough for a first-timer to do that well unless he has 3,000 hits or 500 homers or some similar hook.
By and large I'm hazily following Repoz for the newbies, but I'm stunned with the support he's shown Martinez receiving, which is around 45 percent. I would've guessed one-10th of that because of his seemingly short (for a Hall of Famer) career, that he played DH, and he was outshone by three other teammates for much of his career (and on a team that didn't even win the pennant). Thus for Martinez I'll predict an unusually pronounced gap between the tally and the BBWAA - but this is obviously just pure guesswork on my part.
If Larkin tops 60 percent this year, he's got a strong shot to go in next year, otherwise, he can wait until 2012. He's got a good shot to make it before 2013. That's good because a huge glut of newbies (Craig Biggio, Barry Bonds, Roger Clemens, Curt Schilling) arrives that year which will make it tough for the backlog.
Smith and Morris are the key ones to watch. I think they'll break 50 percent this year, which is when candidates really start to build momentum. They'll have two more years to rise before the 2013 crop emerges. Neither can be counted out, but neither are good bets.
Most everyone else I expect to go up this year, because there is so much room to grow with the small backlog. I kept Trammell down largely because Larkin is his comparable player. I think Baines will survive, but when the ballot becomes much more crowded in 2013, he'll fall off.
Am I right or am I full of it? Check back Wednesday to find out.
References and Resources
Personal annoyance: After last year's ballot, I asked Repoz to email me his full tally. (He'd posted online the results for the main guys, but not for everyone over 5 percent.) He complied, but I mistakenly deleted it. From memory, my margin of error was smaller than his tally last year, but I can't say if that was the case for sure.
History instructor by day, statnerd by night, Chris Jaffe leads one of the most exciting double lives imaginable; with the exception of every other double life possible to imagine. Despite his lack of comic-book-hero-worthiness, Chris enjoys farting around with this stuff. His new book, Evaluating Baseball's Managers is available for order. Chris welcomes responses to his articles via e-mail. Oh, and now he's on twitter.
<< Return to Article