The Envelope Please: Our 2021 Hall of Fame Crowdsource Ballot Results

The following article is part of Jay Jaffe’s ongoing look at the candidates on the BBWAA 2021 Hall of Fame ballot. For a detailed introduction to this year’s ballot, and other candidates in the series, use the tool above; an introduction to JAWS can be found here. All WAR figures refer to the Baseball-Reference version unless otherwise indicated.

It would be only somewhat hyperbolic to say that the 2021 Hall of Fame election cycle was as contentious and polarizing as the presidential election that preceded it nearly three months ago, but let’s face it, this time around has not been a whole lot of fun. When Hall president Tim Mead opens the envelope to announce the results shortly after 6 pm ET on MLB Network on Tuesday evening, there’s a very good chance that the BBWAA voters will produce a shutout, the writers’ first since 2013 — a ballot that not-so-coincidentally is headlined by some of the same candidates who have split the electorate.

There’s no shutout from FanGraphs readers, however. In our third annual Hall of Fame crowdsource ballot, three candidates cleared the 75% bar, down from four last year and seven in 2019. Not surprisingly, Barry Bonds and Roger Clemens did so, just as they’ve done in each of the past two years. However, both members of the gruesome twosome took a back seat to the top close-but-no-cigar candidate from our 2020 crowdsource ballot, and no, I don’t mean Curt Schilling.

Before I get to the results, a refresher on the process. As with the past two years, registered readers of our site (and participating staff, this scribe included) were allowed to choose up to 10 candidates while adhering to the same December 31, 2020 deadline as the actual voters, but unlike the writers, our voting was conducted electronically instead of on paper. This year, 1,152 users participated, a drop of exactly 20% from last year’s 1,440 voters, but one that’s understandable in light of our pandemic-related traffic dip as well as an apparent lack of enthusiasm towards a ballot that, quite frankly, is headed by heels, in that the top four returning candidates in terms of voting percentage have significant issues that would give any character-minded voter pause.

Schilling’s lengthy laundry list of transgressions — including Islamophobic and transphobic social media posts that long ago cost him his job as an analyst at ESPN, as well as his sharing a tweet in support of lynching journalists — interrupted his trajectory towards election, though last year, he pulled to the doorstep of Cooperstown with 70% of the vote, a position from which 20 out of 21 candidates who still had eligibility remaining were elected the following year. Apparently determined to test the percentages, he could not resist adding fuel to the fire during the presidential election cycle by spewing election-related conspiracy theories; sharing a call for martial law; and comparing Dr. Anthony Fauci, the nation’s top infectious disease expert, to a Nazi. After the December 31 voting deadline, he also tweeted his support of the insurrectionists who stormed the U.S. Capitol building on January 6. Reportedly, at least one voter (and perhaps multiple ones) have contacted the Hall about the possibility of rescinding their votes for him. No dice.

Bonds and Clemens, of course, were superstars, albeit ones credibly connected to the use of performance-enhancing drugs, though at a time before MLB could test and penalize for them. Both have additional allegations as well, namely accusations of domestic abuse against the former, and an inappropriate relationship with a minor against the latter.

Omar Vizquel, already a polarizing candidate given the gap between the impressions of his fielding prowess and raw statistics versus what the advanced numbers tell us, was accused of domestic abuse by his wife on Instagram Live in October, setting off an investigation by The Athletic’s Katie Strang that revealed multiple incidents of violence.

Fun stuff, right? Per the Hall of Fame Ballot Tracker, all four have shed support from returning voters who published their ballots prior to the election, with Vizquel losing the most (a net of 12 votes) as of 12:01 AM ET on January 26.

Anyhow, here are our crowdsource results:

Where Schilling has received the highest share of the vote according to the ballots published in the Tracker (75.3% at this writing, through 172 public ballots and 10 anonymous ones), his share was just the seventh-highest via our crowdsource, and what’s more, he was the only one of the 14 returning candidates to lose ground, doing so by a double-digit percentage. Meanwhile, our voters were overwhelmingly in favor of Rolen, who’s above the JAWS standard at third base and whose arguments, for and against, are simply about his on-field performance rather than his conduct. After Bonds and Clemens, whose support barely budged, our voters also favored Jones, Helton, and Sheffield, all of whose candidacies do have their warts, performance and otherwise, more than Schilling. Two other candidates, Wagner and Manny Ramirez, weren’t far behind. Sosa made the biggest jump of all relative to last year’s crowdsource, and while even the candidates who don’t fare particularly well in JAWS and other advanced metrics gained ground, it’s notable that Vizquel, Kent, and Pettitte posted the fourth- through sixth-lowest gains.

Elsewhere, every candidate received at least two votes from our readers, though only two, Buehrle and Hudson, cleared the 5% bar that would be needed to remain eligible beyond 2021.

As for some general trends, our voters were less generous than in previous years when it came to the average number of candidates per ballot and the percentage of voters using all 10 spots. While we don’t have 2021 BBWAA numbers yet, our voters have been more generous in both departments than the actual voters:

Five of our voters turned in blank ballots, up from two last year, but where that latter pair presumably wanted to ensure that Derek Jeter wasn’t unanimous in our poll (he got “only” 89.9%), this time around the protest was presumably more general.

Per developer Sean Dolinar, who deserves an annual tip of the cap not only for building our crowdsource ballot but also providing me with the voting data in several shapes and forms that I didn’t even know I wanted, there were 752 different ballot combinations including the blank, down from 861 last year but up from 660 the year before. A honking 584 of the 752 combos appeared just once, while five combos appeared at least 10 times (down from 13 such combos last year). Here are the most frequent combinations:

All of the most common ballots included a “Big Four” of Bonds, Clemens, Rolen, and Jones — and for seven voters, that was the entirety of their ballots. Helton, Sheffield, and Wagner were each in nine of the top 11 combos, while Schilling was in eight, Sosa in six, Abreu in four, and Kent in two.

My own ballot, which I cast for real for the first time and replicated via our crowdsource, omitted Schilling while including Abreu, Bonds, Clemens, Helton, Jones, Rolen, Sheffield, Sosa, and Wagner. That exact combination was matched exactly once within our set. I’m not sure whether to congratulate that unidentified voter or recommend they seek help for having their brain untangled.

Anyway, as the election results approach, it’s worth comparing our crowdsource results to the 182 published ballots in the Tracker as of 12:01 AM on Tuesday morning, an estimated 46.0% of the electorate:

Jones, Rolen, Manny Ramirez, and Sosa are all tracking at least 18 points lower than our crowdsource; the first two players have cases built firmly on defensive metrics and the last two have links to PEDs, and while there’s a world of difference with regards to being caught and suspended twice (Manny) than being reported as having failed the survey test (both Manny and Sammy), they’re kind of lumped together in the not-Bonds, not-Clemens PED morass. Abreu, Sheffield, Helton, Pettitte, Bonds and Clemens — a mix of PED guys and stathead favorites — are in the 10-to-18 points lower range. At the other end of the spectrum, Schilling, Vizquel, and to a lesser extent Kent are tracking much better than they fared here.

So, will anybody be elected? I really don’t think so. While Schilling, Bonds, and Clemens are all tracking above 70% at this writing, and while there might be another 30 or so ballots added to the Tracker before the results are announced, one need only look at the public/private splits of that trio’s 2020 support to know that there probably aren’t enough voters hiding in the woodwork to put them over the top:

Yow. While the number of ballots in the post-election and private groups are both smaller than the pre-election one, those are some precipitous drops. Schilling’s support on post-election published ballots last year was 11.6% lower than his pre-election share, and his support on private ballots was 26.5% lower. For Bonds and Clemens, the drops from pre- to post-election published ballots were in the neighborhood of 20%, and from pre- to private 28.0% for Bonds, 24.0% for Clemens.

That’s my back-of-the-envelope analysis. For another view, here’s the projection through 163 published ballots from Jason Sardell, who for the last two years has been the most accurate of the handful of Hall of Fame vote projectionists.

Baseball Hall of Fame results will be announced two days from now, but here are the current "unskewed" trends based on 163 ballots in @NotMrTibbs's Tracker. Still looks like we're going to see a shutout. pic.twitter.com/H9TvpzSY0U

— Jason Sardell (@sarsdell) January 24, 2021

Schilling’s 0.3% chance translates to a 1-in-333 shot. Looking back at our 2020 preseason Playoff Odds page, 0.3% is the same chance we gave the Orioles of winning a Wild Card Seres, and the Marlins of winning the National League Championship Series. Neither of those things happened, of course, but the Marlins came closer than anyone expected, winning a Wild Card Series (the odds of which we estimated at 3.1% at the start of the season) before bowing out in the Division Series.

In other words, the odds of Schilling being elected this year are very long, and in all likelihood, we’ll see a shutout. While that will feel anticlimactic to many observers, true Hallheads will be able to look to the middle of the ballot for more encouraging news, because the tea leaves suggest that Rolen, Helton, Jones, Sheffield, and Wagner are poised to make substantial gains. I’ll be back on Tuesday evening with a closer look at the results.