Wednesday’s Cooperstown results today, 2013 editionby Chris Jaffe
January 07, 2013
Damned if I know.
I’ll be damned if I know what’s going to happen. I probably shouldn’t begin like that, right? After all, you’re reading an article with a title that very clearly indicates that I do know what I’m talking about.
Each January for the last several years here at the mighty THT, I’ve done an annual column where I predict what the BBWAA will do. I don’t merely guess will/won’t, but dare to give specific voting percentages for everyone on the ballot.
The track record is pretty good so far. Through five election cycles, I’ve predicted vote totals 77 times and been within five points of the result 64 times (and within one percentage point of the actual results 25 times) with an average margin of error of 3.3 percent.
I had to mention that because there ain’t no way my results will be that good this year. My system is based on looking at past elections and overall trends in BBWAA voting to help decipher what the future will hold. But there’s never been an election like the one coming up this year.
My system has always been shakiest with first-year candidates, and there’s never been an incoming bunch of new candidates like the group we have here: Barry Bonds, Roger Clemens, Craig Biggio, Mike Piazza, Sammy Sosa, Curt Schilling—poor Kenny Lofton gets completely lost in the shuffle, dosen’t he?
We’ve had bumper crops of newbies before. 1999 gave us Nolan Ryan, George Brett, Robin Yount, Carlton Fisk and Dale Murphy. 1989 presented Johnny Bench, Carl Yastrzemski, Fergie Jenkins, Gaylord Perry and Jim Kaat.
But, of course, the 2013 ballot isn’t just about baseball talent. It’s also about steroids. Yes, we’ve had Mark McGwire and Rafael Palmeiro, but there’s been not one like Bonds and Clemens to really inflame the debate.
Okay, so this is the toughest election to predict that’s ever been. That doesn’t mean we can’t try to estimate what Cooperstown will say Wednesday. As I said, my system is based on studying the past, and here are the 10 guidelines I use in trying to predict what a player’s vote total will be.
All things being equal, assume the guy will finish around where he did last year. This makes sense, but since things aren’t always equal, we have another nine guidelines.
2. Strength of ballot
The biggest determinant in figuring out what will happen to members of the backlog in an election is how strong the overall ballot is. If you have a weak crop of new candidates, the backloggers should rise. If you have a strong crop of newbies, the backloggers should suffer.
Obviously, 2013 looks to have a very strong crop of newcomers, even with the taint of steroids on Bonds and Clemens. For perspective, let’s look at the two strongest crops of newbies in the last half-century: 1989 and 1999.
As noted already, Yaz, Bench, Jenkins, Perry and Kaat all showed up in ’89, and the backlog paid for it. Most notably, Jim Bunning entered the ballot on the verge of election, having received 74 percent of the vote in 1988. In 1989, he fell to 63 percent. Behind Bunning was Tony Oliva, whose vote collapsed from 47 in 1988 to 30 percent in '89. Luis Tiant fell even worse, from 31 to 11 percent.
Technically speaking, there was plenty of room for the new guys without having to cut into the vote totals for the old-timers. The 1988 pre-deluge ballot had 6.60 names per ballot, including those elected (Willie Stargell) or those who ran out of time by 1989 (Roger Maris, Elston Howard).
Sure, there was officially room, but backloggers lose support well before the Baseball Writers Association of America approaches the maximum 10 names per ballot. Voters who have put five or six names on their ballot every election for 20 years suddenly feel it’s cheapening things to fill out a full 10 names. In 1989, the overall names per ballot remained about the same—going up from 6.60 to 6.75—despite newbies averaging 3.35 names per ballot on their own.
In 1999, with Ryan, Brett, Yount and Fisk, it was the same thing. There were 17 backloggers—including future Hall of Famers Tony Perez, Gary Carter, Bert Blyleven, Bruce Sutter and Jim Rice—and all 17 saw their vote totals go down. Many had their worst showing ever that year, including Rice (29 percent), Blyleven (14 percent) and Carter (34 percent).
This is pretty bleak news for Jack Morris and the rest of the backlog, but as we'll see in the third guideline, there is some good news for them.
Before moving on, let’s note that it’s not just backloggers hurt by the strength of ballot; newbies can be, too. Jenkins and Perry weren’t elected right away in 1989, not with Yaz and Bench on the ballot. Similarly, Fisk had to wait.
Perhaps the most interesting case was Yount, and it’s especially interesting for what it might tell us about Biggio this year. Sure, Yount became a first-ballot Hall of Famer in 1999, but with just 77 percent of the vote, the lowest total for any 3,000-hit guy not named Rafael Palmeiro in over 30 years.
That’s odd. In the last quarter-century, almost all 3,000-hit players can be put into one of two categories on the ballot. Category one: those who topped 90 percent of the vote. This is guys like Cal Ripken Jr. , Rickey Henderson or Tony Gwynn. These guys were going to be first-balloters even if they retired with 2,900 hits.
Category two: compilers who needed great career records to get into Cooperstown. Think Paul Molitor or Dave Winfield. As it happens, both Molitor and Winfield (and Eddie Murray) won election with 85 percent of the vote, the same percentage for each of them. It’s one of the most impressive bits of consistency in Hall of Fame voting: The three 3,000 guys—all coming up for election in different years—all finished with the same vote total: 85 percent. So you have to figure that under normal circumstances that’s the floor for a 3,000-hit player.
The only 3,000-hit club members in the last 25 years who don’t fit that trend are Yount and Palmeiro.
The problems of Palmeiro are obvious and unrelated to this. Yount? He should’ve been a 90-percent guy with his pair of MVP trophies, but instead he barely skated in with 77 percent. This is notable because there is a 3,000-hit newcomer in 2013: Biggio. He’s the sort of player who in a normal election would get 85 percent. Damn shame for him it isn’t a normal election.
Biggio might get in anyway. If Bonds and Clemens weren’t dinged by steroids, Biggio would be in trouble, but guess what—they are dinged by steroids. For everyone not voting for Bonds and Clemens, Biggio should be an 85-percent guy. Among those who are voting for Bonds and Clemens, Biggio will probably get 60-some percent of the vote.
The less support Bonds and Clemens get, the better the odds are for Biggio.
3. Candidates per ballot
That last bit sounded really bleak, but there is some good news here.
Fun fact No. 1: through 1986, every BBWAA Hall of Fame election averaged at least seven names per ballot.
Fun fact No. 2: from 1987-onward, no BBWAA Hall of Fame election has averaged as much as seven names per ballot.
Last year’s ballot averaged just 5.10 names on it, an all-time low. With Barry Larkin entering Cooperstown, the backloggers averaged 4.16 appearances per ballot. Which is to say, we might have a historic crop of new candidates coming up, but at least there’s a historic chunk of space to give them.
It all depends how much support Bonds and Clemens get. Under normal circumstances, they’d both appear on nearly ever ballot, Biggio and Sosa would be on three-fourths or so, Piazza on most, and Schilling on a bunch. Without the steroids stink, the new crop’s names would average four appearances per ballot.
Well, Bonds and Clemens will perform well below normal expectation, while Sosa looks to do as poorly this year as Palmeiro did last year, somewhere around 10 percent. This new crop is still one of the strongest ever. Along with 1989 and 1999, it will be one of the three best of the last half-century.
But say the newbies average three names per ballot. Combined with the backlog’s 4.10 names per ballot in 2012, that’s just a little over seven names per ballot this year.
Now, seven names per ballot is higher than the BBWAA goes these days. It’s unlikely it’ll go that high this year, and we certainly should expect some overall drop-off from the backlog. But it won’t be the horrific crunch that obliterated Bunning, Oliva and Tiant in 1989. It won’t be the historic hurt that lowered the support for the entire 1999 backlog. Some guys can still rise up this year.
Morris is the highest surviving returning candidate, with 67 percent of the vote last year. My hunch is that he won’t get in. Normally, a guy can rise up from 67 to 75 percent, but Morris is seriously swimming upstream this year. It’s one thing to say his candidacy won’t be obliterated but quite another to say he’ll rise up notably.
4. “Over the top” surge
Well, it should be a bad year for the backlog overall, but that doesn’t mean it’ll be equally bad for all. As a rule of thumb, players on top of the backlog do the best job holding their support. If a player is that high up, a lot of voters consider him the best candidate on the board. Remember how in 1989 Bunning fell from 74 to 63 while Oliva and even Tiant fell even worse.
Logically, that shouldn’t happen. A guy with 74 percent should lose more than a guy with 47 percent because he has more to lose. But that just ain’t the way things play out.
The key point is 50 percent. When a majority of BBWAA voters support a candidate, the logs start rolling in his direction. Voters who haven’t voted for a candidate start going with the wisdom of the majority.
Last year, Morris, Bagwell and Lee Smith all topped 50 percent. Normally, that would foretell good things for their eventual candidacies. In this mess of an election? It means they should fall less. They might maintain or even gain a bit, but that’s about it.
2013 just isn’t a year for the backlog.
5. Comparable candidates
Another key item is if there is a really similar candidate arriving on (or departing from) the ballot.
Look back at Bunning and Tiant. They both fell because three new pitchers—Perry, Jenkins and Kaat—all arrived. All three were more impressive than Tiant, which is why he lost two-thirds of his support. Bunning was arguably better than Kaat and had the over-the-top surge going for him, but he still fell.
Looking at 2013, we have all kinds of new names entering the ballot, though: sluggers, hitters, pitchers—you name it.
The most distinctive backloggers are Smith and Tim Raines, as there are no great relievers or base stealers arriving. Plus, Smith and Raines have some good support going for them already. Alan Trammell also might be helped, as the most comparable person on last year’s ballot was Larkin, and he’s now in Cooperstown.
Steroid backloggers might take a hit. Think for a second: if you’re willing to vote for steroid candidates, you have Bonds and Clemens and McGwire and Palmerio and Sosa. That’s half a maximum ballot. Someone could easily get squeezed off in that scenario, especially if you want to make room for some of the other fine candidates.
6. Steroids (and other muddy matters)
Look, there are essentially two different elections going on at the same time: an election in which voters are willing to vote for PED-associated players, and an election in which voters aren't. However, the results will be mixed together.
Bonds and Clemens will get near 100 percent of the one election and zero percent in the other election. If you’re not willing to vote for them, then you’re not willing to vote for anyone considered dirty because no one—absolutely no one—thinks their numbers or talent are wanting. But if you refuse to support anyone considered unclean, then their stats don’t matter.
How many voters are in one category and how many in the other? That’s a guess.
The steroids dispute has been going on for a while, but it’s never been so dramatic. People could always say that McGwire or Palmeiro wouldn’t be a Hall of Famer if it wasn’t for PEDs. But Bonds and Clemens were already first-rank talents before people think they began juicing.
More than that, if you’re willing to vote for steroid users, the ballot is impossibly crowded with qualified candidates. If you’re willing to vote for Palmerio and Sosa, maybe you think Alan Trammell is the 11th-best candidate. Maybe you think he’s worthy, but the ballot gives you only 10 slots.
Folks, please realize the BBWAA elections typically are exercises in consensus building. That’s why the over-the-top surge happens. There is anything but a consensus when it comes to ‘roids. There’s at least a rough idea of what makes a Hall of Fame pitcher or slugger or hitter or whatever. Not everyone agrees to it, and there is always a gray area, but there’s enough of a consensus to make the elections work.
There’s also some muddiness with relievers. The BBWAA is still trying to figure out how to handle them. The voters are willing to put them in, as witness the 21st-century elections of Sutter, Dennis Eckersley and Rich Gossage. Then again, Sutter and Gossage had to come from way back to get elected.
Smith has had a very strange Hall of Fame arc. In 2003, he debuted with 42 percent of the vote and in 2011 was still at 45 percent. It’s rare someone with so much support to have so little movement. He finally broke 50 percent last year, but that’s because it was a weak ballot.
Base on points made earlier, Smith should do better than many other backloggers, but I’m skeptical. His candidacy used to be based on being the all-time saves leader, but Trevor Hoffman and Mariano Rivera both have blown him away.
7. Last year on the ballot
Players in their 15th and final go-around typically receive a slight bump, around three percent. This year Dale Murphy is the only player in his final try. He may not get the bump, though, given how big the incoming crop of new candidates is. In 1999, when Ryan-Brett-Yount appeared, last-year candidate Minnie Minoso actually had his vote percentage go down, one of the few times that’s happened to a player in his final year.
At the very least, Murphy shouldn’t lose as much support as he otherwise would, and he might still go up a tick from his 15 percent showing last year.
8. Repoz’s Hall of Fame Ballot Collecting Gizmo at Baseball Think Factory
Every year for the last several years, Repoz, the editor-in-chief at BTF collects and tallies every publicly posted BBWAA Hall of Fame ballot. He typically ends up with about a quarter of the overall electorate.
With returning candidates, I find the other criteria here work better, but with new candidates, Repoz and my own gut sense are the only things I have to go on.
As I write this, Repoz has 94 ballots, though he’ll probably have another 50-60 by the time collecting is all done. (MLB.com and ESPN writers don’t release their votes until the day of the election, and that’s more than 30 voters on those two sites alone). As of now, zero candidates are at the magic 75 percent marker.
Biggio is just under, around 70 percent. Piazza is over 60 percent, Clemens and Bonds have fallen below 50 percent, Schilling around 40 percent, Sosa at 14 percent, and everyone else under five percent.
These numbers aren’t perfect, but they serve as a starting point.
It’s worth noting that with some players, Repoz's Gizmo has a consistent tendency to be off. Over the last three elections, it has overestimated Raines’ support by 8.0 percent. It underestimates Jack Morris (off by an average of 5.6 percent), Lee Smith (5.8 percent), Larry Walker (5.6 percent), Don Mattingly (8.4 percent), and Bernie Williams (6.2 percent). To date, the Gizmo is closer with everyone else.
9. Beware five percent
Five percent is the cut-off point for the BBWAA. Everyone finishing below it gets booted from the ballot. Williams is the only returnee who received less than 10 percent of the vote, with 9.6 percent. Even with the crowded ballot, I figure he’ll live to fight another day. Losing half of his support in one year is unlikely.
10. Guidelines ain’t rules
The above are just things to keep in mind. Nothing is set in stone. That’s especially important to keep in mind with such a bizarre election as this year. Sometimes these things contradict each other, and I have to figure out on my own how to best balance them.
Okay, enough of the dilly-dallying. What does my crystal ball foresee? This:
Name Prediction Craig Biggio 76 Jack Morris 69 Mike Piazza 61 Jeff Bagwell 52 Tim Raines 48 Lee Smith 47 Barry Bonds 45 Roger Clemens 45 Curt Schilling 39 Alan Trammell 38 Edgar Martinez 33 Larry Walker 17 Fred McGriff 16 Mark McGwire 16 Don Mattingly 14 Dale Murphy 14 Sammy Sosa 13 Rafael Palmerio 10 Bernie Williams 6 Other guys 7That works out to 6.66 names per ballot, which would be the highest average of the 21st century. But, okay, it is a rather crowded ballot, steroids or not. Besides, it’s only a little higher than 2003, 2004, and 2007, all of which were 6.55 or higher.
I keep going back-and-forth on Biggio. Sometimes I think he’ll just barely nudge over the needed 75-percent marker. Other times, I see him falling short. All that I’m sure of is that he’s on the bubble.
Let’s go back to the Robin Yount comparison again. Yount fell below the Molitor-Winfield-Murray floor of 85 percent for a 3,000-hit guy not just because he was on a crowded ballot, but because there were two clearly superior players to him on the ballot. Brett and Ryan each appeared on virtually every ballot.
For those willing to vote for Bonds and Clemens, Biggio is the third-best candidate. For those not, he’s the most obviously qualified Hall of Famer, with his 3,000-hit credential. By that logic, you’d expect Biggio to get a Yount-like 77 percent from the Bonds-Clemens supporters, and Molitor-Winfield-esque 85 percent from the others. If about half the voters go for Bonds and Clemens, that means Biggio should get 81 percent or so.
But it’s not so simple. First, Biggio’s reputation isn’t as strong as Yount’s was; 77 percent is the highest support Biggio can expect from the Bonds-Clemens backers, not the most likely result.
Second, there is Repoz’s Gizmo at BTF. Not only has Biggio always been under 75 percent, he’s consistently been under 70 percent. That gives me cause for concern. Still, the Gizmo has been off before with guys at the top of the list. When Roberto Alomar debuted on the ballot, the Repoz’s tally pegged him at 87 percent of the known vote. He actually received 74 percent.
Third, it’s such a much more confusing ballot at top. Among Biggio, Bagwell, Raines, Smith, Piazza and Morris, you have a half-dozen guys angling for 50 percent of the vote or more, with Clemens and Bonds right behind them. All those guys could lower each other's vote total.
My hunch is that Biggio just skates in. Many voters see 3,000 hits and just check his name. For half of the voters denying Clemens and Bonds, Biggio is the easiest pick because he comes with that bright shiny number: 3,000.
Would I be surprised if Biggio is on the outside looking in when the votes are announced Wednesday? Heck, no. But I still think it’s more likely than not that he gets the call, even if it is only slightly more likely than not.
Check back on Wednesday to see how wrong I am.
References and Resources
Years ago, I took the results from every previous BBWAA election and put them into an Excel spreadsheet. That's the basis for this analysis.
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.