Dissecting the 2009 Hall of Fame voteby Chris Jaffe
January 19, 2009
Another year, another Hall of Fame vote.
Two weeks ago in this space I tried to predict how the Hall of Fame vote would go. On the one hand, I did a heckuva job. I successfully predicted Rickey Henderson and Jim Rice would go in, named the top seven vote recipients in order, and correctly figured that Henderson would be the only new candidate to receive more than 5 percent of the vote. I tried to guess the exact vote percentage for 15 different candidates and had an average error of 3.1 percent, besting the forecasting by anyone else (that I know of, anyway).
My point isn't to blow my own horn. I have to remind myself of all the above because, quite paradoxically, I was absolutely stunned by the results of the vote.
Let me clarify. The entire ballot didn't stun me. The two inductions were widely predicted, for example. Yet the results near the top surprised me.
Heading into last week, I was banking on a sizable (and frankly, typical) rise in support for those on top of the backlog. It didn't happen. Sure, Rice received enough support to enter Cooperstown, but he experienced one of the weakest rises of anyone who ever topped a backlog.
He had everything going for him—top of the backlog, it was his final year on the ballot (which generally causes a small upward nudge), strong advocacy from his supporters—yet he only rose from 72.2 to 76.4 percent. In contrast, the only other person to make Cooperstown in his final year on the ballot, Ralph Kiner, skyrocketed up by 16.5 percent when he won entry.
I'm not a fan of Rice's candidacy, so soft support for him doesn't bother me. I was much more troubled by what happened below him. Andre Dawson and Bert Blyleven were the top of the non-Rice backlog. Both stayed at a standstill. Dawson went up by 1.1 percent and Blyleven by only 0.8 percent. As noted in a previous column, guys between 60-70 percent usually have another 6-7 percent added onto their total for the next year.
In fact, it was an exceptionally flat year all around. Compare the vote for all this year's non-new candidates with their support from the year before:
Name 2009 2008 Dif Jim Rice 76.4% 72.2% 4.2% Andre Dawson 67.0% 65.9% 1.1% Bert Blyleven 62.7% 61.9% 0.8% Lee Smith 44.5% 43.3% 1.2% Jack Morris 44.0% 42.9% 1.1% Tommy John 31.7% 29.1% 2.6% Tim Raines 22.6% 24.3% -1.7% Mark McGwire 21.9% 23.6% -1.7% Alan Trammell 17.4% 18.2% -0.8% Dave Parker 15.0% 15.1% -0.1% Don Mattingley 11.9% 15.8% -3.9% Dale Murphy 11.5% 13.8% -2.3% Harold Baines 5.9% 5.2% 0.7%
As tepid as Rice's bounce was, it outdid anyone else's change. In comparison, the year before saw nine separate backloggers have their vote totals change by at least 4.6 percent. I've checked every election in the last half-century, and they all had at least one (and usually numerous) players rise or fall by more than 4.2 percent. It's like they just re-ran this year's vote. In all, 13 backloggers had their votes change by a paltry 1.7 percent average.
There is a better way to check and see if 2009 experienced the smallest vote change from 2008 of any in Hall of Fame election in BBWAA history. Take everyone from the backlog, determine how much their votes changed (a drop of -5 percent and a rise of +5 percent would both be considered 5 percent in this case), find the average change, and compare to other years.
I looked at the last 38 years of elections using this. In all that time, here are the elections with the smallest changes from the previous year:
Years Avg Chng 2008 to 2009 1.7% 1992 to 1993 2.0% 1994 to 1995 2.1% 1987 to 1988 2.6% 2002 to 2003 2.8%
Yup, it's official: 2009 is the winner for most look-alike election with its predecessor.
In and of itself, the above is merely interesting, but no cause of concern. Yet I found it disturbing.
To explain my qualms, look at the year with the second-deadest vote total, 1993. A profound difference separates 2009 and 1993. As I noted in this and last year's columns predicting the Hall of Fame vote, the single most important determiner for figuring how the backlog will do is overall strength of the ballot. Stronger new crops cause them to decline; weak ones allow backloggers to rise.
1993 was a very strong crop of newbies. Reggie Jackson won election in his debut, Phil Niekro broke 60 percent while Steve Garvey topped 40 percent. In all, an average ballot contained 2.09 newbies.
In 2009, you had Henderson and nothing else. No other newbies even topped 4 percent. The 2009 BBWAA averaged 1.06 names per ballot, barely half of 1993's newbies. That should not be enough to stall the entire backlog dead in its tracks. They stopped dead anyway. Last year set a new record for fewest names per ballot at 5.35. If any backlog had room to rise up, it was this year's bunch. They didn't.
That's not good. I'd like to see the current backlog-toppers (Dawson and Blyleven) enter Cooperstown, and next year's ballot has a much stronger crop of newbies with Roberto Alomar, Barry Larkin, Fred McGriff, and Edgar Martinez. I don't think any will make it on the first ballot (Martinez may not top 5 percent, but that's a whole other story), and that means a chance emerges no one goes in from the BBWAA next year.
Perhaps more alarmingly, Bert Blyleven is running out of opportunities. He only has three years left, and he can't afford another year of zero momentum. A few weeks ago I assumed he was a shoo-in for eventually BBWAA election. Now a definite chance arises that this won't happen. That makes the Great Dead Cat Bounce of 2009 so vexing.
But . . . . WHY?
That's what happened and those are the unwanted potential ramifications. Next question: why did it happen? In particular, why did the guys near the top of the backlog fail to rise at all? After all, they're the ones most likely to someday earn induction.
The almost complete disappearance in 2009 of a surge in voter support for those topping the backlog serves as an abrupt change from recent history. In 2008, Rich Gossage jumped 14.6 percent from 71.2 to 85.8 percent, effectively convincing half those not supporting to join the bandwagon. Heck Rice, Dawson and Blyleven—this year's momentum-less trio—moved up by 8.7, 9.2 and 14.2 percents respectively.
Well, admittedly, 2008 had a very small crop of first-balloters (only 0.26 names per ballot) allowing for the backlog to rise up. But that doesn't explain what happened in 2007. With Cal Ripken and Tony Gwynn arriving on the ballot, backlogger Gossage still managed to rise up by 6.6 percent, far more than anyone this year. Looking at other recent elections, the same trend plays out: the failure to surge not only contradicts long-standing BBWAA voting habits, but recent trends as well.
So very little precedent exists for what happened this year at all. That still doesn't answer the 64-cent question: why did everyone run in place this year?
I have no answer I can mathematically prove, but I do lean towards one reason: Jim Rice. I have math that can show the surge effect exists for those on top of the backlog. To me, having the numbers is immaterial unless you can explain the phenomenon using words and ideas.
A simple explanation exists for the backlog surge: as players rise up the ballot, the rest of the electorate becomes increasingly aware of their candidacy. With that awareness, they start looking for reasons to vote for the individual. The higher up a person gets on the backlog, the more the overall electorate is willing to vote for him. In fact, guys who received 70-75 percent of the vote surge higher than those at 65-70 percent, who top 60-65 percenters and so on.
Normally, a person doesn't make the top of the backlog unless he has something really strong in his favor. Rice had two distinctive features in his candidacy.
First, he was a weaker backlog-topper than most. Forget sabermetrics for a minute. By Hall of Fame standards (and especially by BBWAA standards), Rice had a short career with clear holes in his game. I'm old enough to remember when Rice played, and while Rice looked like a possible Hall of Famer in his prime, his sudden decline in the 1980s really dropped him. He was just another guy with a great stretch, not necessarily a Hall of Famer.
Second, Rice had a chorus arguing against his induction. That normally doesn't happen. Like I said, people usually look for reasons to vote someone in when they top the backlog. In this case some argued he not only didn't belong, but was a terrible candidate.
This by no means makes Rice unique. Much of the above could be said about Bruce Sutter as well. However, some of the argument against Sutter was generalized anti-relief pitcher induction, a position that likely lost force with the BBWAA electorate with the induction of Dennis Eckersley a few years early. (Sutter also had the advantage of entering during an exceptionally weak year for first-time candidates on the ballot.)
This doesn't mean that the sabermetric community should pat itself on the back for Rice's relatively muted vote rise. (My hunch is that Rice's career length had the most to do with it.) Still, either 70+ years of voting history abruptly ended without warning, or something special happened last year to explain the minimal surge.
If people aren't going to surge for Rice, it becomes less likely that they'll surge for anyone else near the backlog. Dawson is fairly similar to Rice (though distinctly similar in my mind) so a declined surge for one would hurt the other. Blyleven was the third highest, so if the first two aren't going to surge, he's less likely to as well.
Upon seeing the ballot last year, I was distinctly worried that Dawson and Blyleven had whacked into a ceiling of support. That is possible, though I likely just overreacted to one year's vote. Dawson remains an excellent candidate for eventual induction. Blyleven—well, I think he should go in, but between his few remaining opportunities and the strength of 2010's newbies, his supporters should consider readying themselves for a push in the next year or two. Better safe than sorry.
One other thing
The 2009 ballot contained 13 backloggers, the lowest in BBWAA history. With the election of Rice, Tommy John hitting the time limit, and no newbies receiving between 5-75 percent of the vote, next year's ballot will contain only 11 backloggers.
Please note, just two decades ago it was common for the backlog to consist of 20 or more candidates. Now it's only half of that. Impressive, eh?
As a result, the overall ballots per name over the last 20 years are at historic lows. Until 1987, there were always at least seven names per ballot; now it's frequently under six.
I think I figured out why over the last week. Look at the 1987 ballot, for example. That year the BBWAA elected Billy Williams and Catfish Hunter. Not only was Catfish Hunter inferior to Bert Blyleven, but was he better than Jack Morris? Hunter sailed into Cooperstown after just a few years on the ballot. Morris has struggled to generate momentum for years.
Williams was superior to outfielders like Dave Parker and Dale Murphy, but is his superiority enough to explain the tremendous gap in their voting? Williams broke 75 percent while those guys are generally milling around 15 percent.
Actually, the cases of Parker and Murphy are very instructive. In the 2000s, Parker received the support of 10 to 19 percent of the BBWAA nine times; Murphy landed there seven times. In the 1980s, no player finished in the 10s that many times. Elroy Face had the most appearances with six. In the 1970s, only Alvin Dark finished in that stretch more than four times.
I couldn't help but notice that Face and especially Dark are far inferior to Murphy and Parker. Murphy won two MVP Awards, and Parker was an MVP who lasted long enough to make the All-Star team in three different decades. Yet they receive the same level of support previously reserved for the likes of Lew Burdette and Phil Cavaretta.
The key revolves around what happened to the vote beginning in 1989. That year, Jim Kaat reached the ballot. And Fergie Jenkins. And Jim Perry. An impressive trio, no? They didn't even headline the newcomers, as Johnny Bench and Carl Yastrmzski arrived as well.
From 1989 to 1999, the BBWAA witnessed an unprecedented surge of newbies. The following guys arrived on the ballot in those years: six 300-game winners, a 500-home run batter widely hailed as the greatest slugger (and best postseason performer) of his generation, a former Triple Crown winner with over 3,000 hits, another 3,000 hitter who once batted .390 in a season, yet another 3,000 hitter who once batted .388, still another 3,000 hit man who won two MVPs, a pitcher with six consecutive 20-win seasons, another hurler with eight total 20-win seasons, arguably the greatest catcher of all time, possibly the game's greatest second baseman, and a man who was by acclamation its best third baseman.
The Hall of Fame has always had those who think its in/out line is far higher than it actually is. It's actually Stan Coveleski (and possibly Waite Hoyt), yet some grouse that its standards have lowered anytime someone worse than Bob Gibson enters it.
But the 1989-99 candidates helped justify the falsely high bar some set for the institution. If it was just a one- or two-year blip in quality of candidates, maybe it wouldn't have been a big deal, but they kept arriving every year.
Upshot: the guys not as good as Robin Yount or Gaylord Perry are seen as wholly unfit and thus receive less support. Alan Trammell scuffles with less support than Harvey Kuenn, Jack Morris falls well under Catfish Hunter, Murphy and Parker can only dream of the support Orlando Cepeda and Tony Oliva enjoyed, and Tommy John had roughly similar support to that of Lew Burdette. Most sadly, with this drop in support, players like Lou Whitaker, who in days past would've been part of the 20+ man backlog, fall under 5 percent.
It's likely no coincidence that shortly after the 1989-99 deluge the VC was recalibrated to largely block future inductions from that body. It's players like Mike Schmidt and Reggie Jackson who seem most eager to uphold the falsely high standards they think Cooperstown represents.
References and Resources
Last year I went to the Hall of Fame's website, and dumped the results of every election into an Excel file. Since then, I've played with the data (and added the 2009 election results), which serves as the basis for this article.
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.