Election Wonk, but make it baseball

Are home run totals turnout or persuasion?

At the end of every summer sixty writers decide each of baseball's major awards, and every January a few hundred more fill out a Hall of Fame ballot. The results get the full election-night treatment, with every margin dissected and every stray first-place vote being magnified, but the voters themselves mostly stay in the background as a list of names at the bottom of a results page. Voter Profiles, new on Baseball Hopper, is the voter file for that electorate, with a page for every writer. It collects every Hall of Fame ballot a writer has made public since 2009, using Ryan Thibodaux's BBHOF Tracker (the closest thing the Hall of Fame vote has to early returns), and every individual award ballot (MVP, Cy Young, Rookie of the Year, Manager of the Year and the new Reliever of the Year) that BBWAA has publicly published since 2012. That adds up to a page for 1,026 writers, from the ones who have voted once to the ones who have voted every year the records exist.

Each award ballot is also set next to a WAR ballot: the players a voter would have listed if they had simply ranked the eligible players by FanGraphs Wins Above Replacement (fWAR). That means everyone for MVP, pitchers for the Cy Young, rookies for Rookie of the Year and relievers for Reliever of the Year, with Manager of the Year left out because managers have no WAR. A ballot's WAR match is the WAR it captured, weighted by the points each slot is worth, divided by the same total for the WAR ballot. A ballot at 100% carried as much WAR as the leaderboard itself, and most ballots land somewhere between 90% and 98%. The WAR ballot works like a forecast model, and each writer's ballot is the actual vote. WAR is not what the BBWAA asks its voters to measure, and plenty of good ballots depart from it on purpose, so WAR match describes how closely a ballot tracked one widely used measure of value rather than whether the ballot was right.

Breaking Down the Electorate

The major awards each year are not decided by the whole BBWAA. The electorate is drawn by district: each award goes to a panel of two writers from each city with a team in that league, thirty per league, and the chapter chair in each city recommends who those writers will be. That leaves every district to run its own nominating process, and the chapters have used it very differently, some rotating the seat almost every year and others returning the same voters like safe incumbents. From 2012 through 2025, the 840 MVP ballots came from 287 different writers, and 114 of them cast exactly one. Boston is the most competitive district, sending 16 different writers to fill its 28 MVP slots over those fourteen years, and no one held the seat more than five times. Baltimore and Kansas City each used 15 writers, and Miami used 14.

At the other end are the two longest-serving incumbents in the electorate, the only names on every MVP vote in the period. Marc Topkin of the Tampa Bay Times voted for the American League MVP in all fourteen years from 2012 to 2026, and Bob Nightengale of USA Today did the same for the National League. Neither has missed an election, and both chair their BBWAA chapters, the role that recommends who fills the seats. Each holds 14 of his chapter's 28 MVP ballots, and Tampa Bay and Arizona are the two least competitive districts in the electorate, with five different MVP voters apiece. A handful of other districts are nearly as safe: Colorado and San Diego each sent just six writers to fill their 28 MVP slots, and in St. Louis two writers cast 23 of the chapter's 28 ballots between them.

New York, Los Angeles and Chicago vote twice as often because they have two teams, and once that is accounted for they land in the middle, with 22 to 24 different writers across 56 ballots. The districts also lean a little differently. Averaged over each chapter's voters on every award, Detroit sits closest to the WAR ballot (0.7 points above the median voter) and Tampa Bay furthest from it (3.2 points below), with San Diego and Colorado also near the bottom. The districts with more turnover tend to lean slightly closer to WAR, but the whole chapter range is only about four points, much smaller than the gaps between individual writers, and Arizona, with the same five voters as Tampa Bay, sits near the middle of the list.

The MVP swing, all the way down the ballot

The clearest swing in the electorate is in the MVP vote, which has tracked WAR much more closely since 2022. From 2012 through 2021, the median MVP ballot captured between 88.5% and 94.4% of the WAR ballot in a given year, with lows of 89.5% in 2012 and 88.5% in 2016. From 2022 through 2025 the median ran 97.5%, 96.4%, 97.0% and 94.9%. Part of that comes from the top of the ticket: five of the eight MVP races since 2022 were unanimous landslides (Aaron Judge, Shohei Ohtani three times and Ronald Acuña Jr.), against three of twenty in the decade before, and a season like Judge's or Ohtani's makes the top of the ballot easy to agree on. The swing shows up just as strongly down-ballot, though, where no one is deciding who wins. In the fourth through tenth places on an MVP ballot, about 49% of picks from 2012 to 2021 came from that season's top ten in fWAR, and in the four years since it has been about 66%. The typical player named in those spots ranked somewhere between 9th and 15th in WAR in the earlier years and between 7th and 10th since. Voters are filling the back half of the ballot from much closer to the top of the WAR leaderboard than they used to.

The Cy Young electorate swung the same way, though less smoothly. The median Cy Young ballot captured 87.8% of the WAR ballot in 2012 and 88.7% in 2013, and 97.8% in 2025 and 98.4% in 2026, but the years in between include an 83.7% median in 2016 and 90.2% in 2023. The reason the Cy Young line bounces is the subject of the next section.

Where the electorate breaks from the model: pitchers

The electorate breaks hardest from the model on pitchers, and it breaks in both directions. Four times since 2012 a pitcher led his league's MVP field in fWAR. Clayton Kershaw in 2014 won the award with 18 first-place votes. The other three got almost none: Jacob deGrom in 2018 received one first-place vote and finished fifth, Shane Bieber in 2020 received none and finished fourth, and Corbin Burnes in 2021 received none and finished 15th, in the same season he won the National League Cy Young. MVP voters tend to treat a pitcher as a candidate for a different office, the Cy Young, and those years account for a good share of the dips in the MVP line.

Within the Cy Young, the split is over how to measure a pitcher, not whether to count one, a little like two pollsters weighting the same sample differently. FanGraphs builds pitcher WAR on Fielding Independent Pitching (FIP), which estimates what a pitcher's run prevention should have been from his strikeouts, walks, hit batters and home runs allowed, the outcomes that do not depend on his fielders. Voters tend to weigh the runs a pitcher allowed and the innings he covered, and when those two views part, the winner can come from well down the WAR list, an upset by the model's standards. Robbie Ray won the 2021 AL Cy Young while ranking tenth among AL pitchers in fWAR. Blake Snell won in 2018 ranking eighth and again in 2023 ranking sixth. Sandy Alcantara won the 2022 NL award unanimously while ranking third, and Aaron Nola, the league's fWAR leader, did not receive a first-place vote. The same kind of gap appears in the newest results: Sal Stewart won the 2026 NL Rookie of the Year while ranking 11th among rookies in fWAR.

The outlier voters and persuasion over replacement

Every electorate has its outliers, and this one is no exception. Each writer's profile ranks them by how far their ballots landed from the median voter in the same award and year, and most of the electorate sits close together: of the 229 writers with five or more comparable ballots, 159 land within two points of the median. A handful sit much further out, a few well above the median and a longer tail well below it (the chart names the writers at either end), though their places rest on small samples and could change with a ballot or two. A writer's number reflects which awards and years they happened to vote on as much as how they vote, the way a district's results depend on which races were on its ballot that year, and a ballot that strays from WAR may reflect a view of value the voter would defend on its own terms.

Election analysts sort any shift in a vote into two kinds: a change in who is voting, and persuasion of the voters who were already there. The award vote's swing toward WAR is almost all persuasion. The electorate has turned over steadily: in 2016, about one in five MVP and Cy Young ballots came from writers with no award ballot between 2012 and 2015, and by 2025 it was nearly two in three. In any given year, though, those newer writers and the ones who were voting before them landed within about a point of each other, and first-time voters sat no closer to the WAR ballot than writers on their seventh. The change came from the writers who already held the seats. The 151 writers who cast MVP or Cy Young ballots both before and after 2022 raised their own WAR match by 2.7 points on average, nearly the same as the 3.2-point rise across the whole electorate, and their standing relative to the median voter did not move at all.

The Hall of Fame: persuasion AND replacement

Hall of Fame ballots can be measured the same way using career fWAR: a ballot's WAR match is the career WAR of the players it named, divided by the career WAR of the same number of top candidates on that year's ballot. Here one bloc of candidates decides most of the result. The players the site flags as linked to performance-enhancing drugs, from Barry Bonds and Roger Clemens to Alex Rodriguez and Manny Ramírez, carry some of the highest career WAR totals on any ballot they appear on, so leaving them off pulls a ballot well below the WAR leaders whatever else it does. With them counted, the median public ballot's WAR match ran between 87% and 91% from 2009 to 2012, fell to 73% in 2013 when Bonds, Clemens and Sammy Sosa joined the ballot, and climbed back to 92% by 2020 as more voters added them. With those candidates set aside on both sides of the comparison, the rest of the vote shows no trend: the median landed between 85% and 94% in every year but 2026. Part of the gap from 100% reflects the measure rather than the voters, since relievers like Mariano Rivera, Trevor Hoffman and Billy Wagner have modest career WAR, and some candidates were judged on things WAR does not measure, from voters' concerns about Curt Schilling's and Omar Vizquel's conduct off the field to Carlos Beltrán's role in the Astros' sign-stealing.

On those candidates, the Hall of Fame electorate shifted through both channels at once, persuasion and replacement. In 2013, 46% of public ballots named Bonds or Clemens, and by 2022, their last year on the ballot, it was 72%, while Bonds's official share rose from 35% in 2014 to 66%. The public ballots work something like an exit poll with a known lean: Bonds ran between one and six and a half points better on public ballots than in the official count every year from 2014 to 2022, so the writers who publish are not a perfect sample of the whole electorate. Part of that came from writers changing their minds: among 242 writers with public ballots in both 2013 to 2015 and 2020 to 2022, 45 began voting for Bonds or Clemens and three stopped. The rest came from new voters entering the electorate. Writers whose first public ballot came in 2017 or later (by then most voters published their ballots, so a first public ballot that late usually marks a newly eligible writer) named at least one PED-linked candidate on 72% to 85% of their ballots from 2017 to 2024, against 64% to 75% for writers who had been public since 2014 or earlier, and the gap held after Bonds and Clemens left the ballot. Newer writers also fill out longer ballots, about one more name per ballot than the longer-tenured group every year, and with the PED-linked candidates set aside their picks have sat as much as four points closer to the WAR leaders in recent years. Unlike the award vote, which moved almost entirely by persuasion, the Hall of Fame vote has been reshaped both by the writers who reconsidered and by the ones who replaced them.

Reading the Voter File

The profile pages are built for that kind of precinct-level reading. Topkin's page is a good place to start because it has the longest award record on the site. Every profile opens with a summary of the writer's tendencies, including how many names they put on a Hall of Fame ballot compared with other public voters, how often their first-place pick won and how their WAR match ranks. Below that is a grid of every Hall of Fame candidate they could have voted for, year by year, and each award ballot shows where every pick finished and where he ranked in fWAR, with the year's WAR ballot underneath for comparison. The last section lists the five writers whose Hall of Fame choices most resemble theirs.

Each profile also shows how full a writer's Hall of Fame ballots ran, which matters because the ballot itself changed: from 2014 to 2019, public ballots averaged between 8.2 and 8.8 names and roughly half used all ten spots, before falling back to about six or seven names once the backlog of candidates cleared. The list of voters is searchable by name and outlet, and anyone curious about who represents their city can pull up their district's voters and check the record.

Method and sources

Award ballots come from the BBWAA's published voting pages (2012 to 2026, plus the 2009 NL Cy Young). Hall of Fame ballots come from the BBHOF Tracker (2009 to 2026) and include only ballots writers made public, so they are a subset of the full electorate. WAR is FanGraphs WAR, player ids come from the MLB Stats API, and Baseball-Reference award pages were used to resolve MVP vote-getters listed by surname only.

  • The WAR ballot is the top players by fWAR among those eligible for each award in each league, with as many places as the real ballot (ten for MVP, five for Cy Young and Reliever, three or five for Rookie of the Year).

  • WAR match weights each spot by the official points (MVP 14-9-8-7-6-5-4-3-2-1, Cy Young 7-4-3-2-1, three-place awards 5-3-1). Five-place Rookie of the Year ballots (2025 on) are assumed to use the Cy Young scale.

  • Rookie eligibility follows FanGraphs' rookie flag, and a player traded between leagues counts only the WAR he produced in that league.

  • Writer and chapter rankings count only writers with three or more comparable ballots, and chapter averages place each writer in their most recent chapter.

  • Hall of Fame WAR match uses FanGraphs career WAR (batting plus pitching) and compares each public ballot's picks with the same number of top-WAR candidates on that year's ballot. Blank ballots are left out.

  • PED-linked candidates are the ten the site flags: Mark McGwire, Rafael Palmeiro, Barry Bonds, Roger Clemens, Sammy Sosa, Gary Sheffield, Manny Ramírez, Andy Pettitte, Alex Rodriguez and Ryan Braun. David Ortiz, whose link rests on a reported 2003 survey test result that MLB later called inconclusive, is not included.

Get Balls & Sticks in your inbox

Subscribe for free to get new posts by email as soon as they publish.