Baseball Player Won-Loss Records
Home     List of Articles



The Impact of the Ballot Cap on 2014 Hall-of-Fame Voting

There has been a lot written about who should and shouldn't vote with specific voters called out for specific votes. To be honest, I think the BBWAA generally does a good job and most of the writers who write about their votes strike me as thoughtful and serious about this. They don't all vote the exact same way that I would, but that's why they hold elections.

In general, I have no problem with the Hall-of-Fame electorate. I have one very specific problem with the Hall-of-Fame election process: the damn ballot cap.

Late last summer, I wrote an article asking the BBWAA to allow Hall-of-Fame voters to vote for as many candidates as they wanted, rather than maintaining the historical ballot cap of 10.

To quote myself from that article (my apologies if quoting oneself seems tacky):
On its website, the BBWAA revealed the ballots of 125 voters in the 2013 election (out of 569 ballots total). Almost half of those ballots (61) included at least 8 players who will be on the 2014 ballot as well. This means that these voters couldn't add all of 4-time Cy Young and 355-game winner Greg Maddux, 2-time Cy Young and 305-game winner Tom Glavine, and 2-time MVP and 521-HR hitter Frank Thomas to their ballot without removing somebody they voted for in 2013.

Nearly 70% of the 2013 ballots revealed by the BBWAA (86 of 125) included at least 6 players who will return for the 2014 ballot. These voters could fill their ballots simply by adding Maddux, Glavine, Thomas, and 270-game winner Mike Mussina.

Voters may well be open to persuasion or interested in having conversations about the Hall-of-Fame cases for Mike Piazza or Craig Biggio or Jeff Bagwell or Jeff Kent. But with the 10-man ballot limit, the majority of Hall-of-Fame voters are simply unable to have those conversations. Or, at the very least, the question becomes not simply, "Does Jeff Kent deserve to be in the Hall of Fame?" but "Does Jeff Kent deserve to be in the Hall of Fame more than somebody that I voted for last year?" That becomes a much more difficult and much less "joyful" conversation.

And it's a problem that's only likely to get worse. If there's no room for voters to add Tim Raines to their ballots, then Tim Raines can't gain any upward momentum toward 75%. But meanwhile, Tim Raines is likely to stay on many of the ballots of the 52% of voters who have already decided he's a worthy Hall-of-Famer. Which, in turn, clogs those ballots up from considering other worthy candidates.

If the 10-man limit on the ballot is removed, though, now, the voters are free to engage in those conversations again. Voters who are not already voting for Tim Raines are now free to at least consider him without having to re-consider all of the other players the voter has already decided are deserving Hall-of-Fame candidates. And those voters are, in turn, free to converse with Raines voters and try to persuade them to support other worthy candidates: Alan Trammell, Larry Walker, Curt Schilling, et al. And the new candidates, such as Mike Mussina and Jeff Kent, can be given a full and fair evaluation based purely on their own merits, rather than as part of a complicated strategical evaluation of what subset of deserving players to include on one's size-limited ballot.

More talking about great baseball players. More celebration of baseball history. More joy in Hall-of-Fame debates. What's not to love?
For the 2014 election, the BBWAA kept the 10-man ballot cap in place. And, as I predicted, the result was that a massive number of Hall-of-Fame voters filled their ballots and that there was, perhaps, as much discussion of the proper strategy in filling out a ballot as there was about the worthiness of individual candidates.

So, what impact did the ballot cap have on the 2014 Hall-of-Fame election?

I'll start with my answer. The next table shows my estimate of what the 2014 Hall-of-Fame election results would have looked like had there been no ballot cap. The details of how I came up with these numbers along with some analysis of them follows.

Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Greg Maddux
97.2%
97.2% -
Tom Glavine
92.5%
91.9% -
Frank Thomas
86.0%
83.7% -
Craig Biggio
78.3%
74.8% 68.2%
Mike Piazza
64.4%
62.2% 57.8%
Jack Morris
64.3%
61.5% 67.7%
Jeff Bagwell
59.5%
54.3% 59.6%
Tim Raines
53.1%
46.1% 52.2%
Lee Smith
38.0%
29.9% 47.8%
Mike Mussina
37.7%
20.3% -
Roger Clemens
36.6%
35.4% 37.6%
Barry Bonds
35.9%
34.7% 36.2%
Edgar Martinez
35.6%
25.2% 35.9%
Curt Schilling
35.6%
29.2% 38.8%
Alan Trammell
27.8%
20.8% 33.6%
Jeff Kent
24.5%
15.2% -
Larry Walker
18.2%
10.2% 21.6%
Fred McGriff
18.0%
11.7% 20.7%
Mark McGwire
16.3%
11.0% 16.9%
Don Mattingly
10.0%
8.2% 13.2%
Sammy Sosa
8.4%
7.2% 12.5%
Rafael Palmeiro
5.6%
4.4% 8.8%
Other (Combined)
5.3%
4.2% -
Total Names per Ballots
9.49
8.39 6.60
Players in bold would have been elected under 75% election standard. Players in italics would have failed to meet 5% criterion for staying on the ballot.
The Data
Ryan Thibbs (who goes by the Twitter handle @Weird_Meat) has put together a wonderful spreadsheet that has recorded the Hall-of-Fame votes of over half of the electorate (289 out of 571 as I write this).

Putting together his data, along with the final results released by the BBWAA, the next table summarizes the voting results for Known Ballots*, Unknown Ballots, and Overall.
*I made two changes to Thibbs's data. The BBWAA has confirmed that one voter submitted a blank ballot. The identity of that voter has not been revealed, so (understandably) that ballot was not counted as "known" in Thibbs's spreadsheet when I did this analysis (Thibbs has since added this anonymous vote to his spreadsheet). Since we know exactly how this voter voted, though, I have treated this as a "known" vote. Thibbs does include one ballot by a voter who did not remember his full ballot (he voted for 10 players but could only tell Thibbs 8 of them). I have excluded this ballot from the "known" ballots, since we don't know the exact content of it. The result is that I agree with Thibbs's total number of "known" ballots as of the time I wrote this (289), but disagree with him on the total number of votes associated with those 289 ballots. Thibbs has identified seven additional ballots since that time; I have not updated my analysis to incorporate these additional ballots.

Percentage of Vote
Player Known Ballots Unknown Ballots Difference
Jeff Bagwell 54.7% 53.9% 0.8%
Craig Biggio 75.8% 73.8% 2.0%
Barry Bonds 40.5% 28.7% 11.8%
Roger Clemens 38.8% 31.9% 6.8%
Tom Glavine 94.5% 89.4% 5.1%
Jeff Kent 13.8% 16.7% -2.8%
Greg Maddux 99.0% 95.4% 3.6%
Edgar Martinez 23.2% 27.3% -4.1%
Don Mattingly 4.2% 12.4% -8.3%
Fred McGriff 11.8% 11.7% 0.1%
Mark McGwire 10.0% 12.1% -2.0%
Jack Morris 58.8% 64.2% -5.4%
Mike Mussina 22.8% 17.7% 5.1%
Rafael Palmeiro 5.2% 3.5% 1.6%
Mike Piazza 67.1% 57.1% 10.0%
Tim Raines 54.3% 37.6% 16.7%
Curt Schilling 36.7% 21.6% 15.0%
Lee Smith 23.5% 36.5% -13.0%
Sammy Sosa 6.6% 7.8% -1.2%
Frank Thomas 90.0% 77.3% 12.7%
Alan Trammell 20.4% 21.3% -0.9%
Larry Walker 7.6% 12.8% -5.2%
Other (Combined) 0.3% 8.2% -7.8%
Total Ballots 8.60 8.19 0.41


What We Know
The ballots collected by Thibbs are not a random sample, but are, instead, a self-selected sample: Thibbs knows the votes of voters who volunteered the information. Not surprisingly, therefore, the differences between the Known and Unknown Ballots are somewhat greater than one might expect from a pure random statistical sample. That said, I am actually very impressed at how similar the results are in most cases.

A few observations about the two groups.

Of the 289 Known Ballots, 168 (58.1%) of them included 10 names.

On Ryan Thibbs's wonderful spreadsheet, he not only identifies who each voter voted for, but he also provides a source for each ballot. For those ballots which included 10 names and for which Thibbs provided an online source, I checked the source to see if the voter gave any indication about any other players that he or she would have voted for had there been no limit.

The results were a mixed bag. For some voters, the source simply listed the ballot. Some voters wrote a column that gave no real indication of anybody else the voter might have considered. Some voters referenced other players without being clear about whether they would have voted for them. Some voters made reference to how difficult the vote was but didn't come right out and name names.

But some voters explicitly named the names of some players that they would have voted for but for the ballot cap.

Overall, of the 168 Known Capped Ballots, I counted 64 voters who explicitly mentioned at least one additional player that they would have voted for. These 64 voters identified a total of 189 votes - about 3 per voter - they would have liked to have made but for the ballot cap. I tried to be fairly conservative in making this determination, so this is most likely an under-estimate of how many additional players these 64 voters would have voted for and I am fairly certain that there are other voters (even other Known Voters) who would have voted for additional players who I have not counted here.

The 189 known casualties of the ballot cap break down as follows.

Jeff Bagwell 9
Craig Biggio 6
Barry Bonds 2
Roger Clemens 2
Tom Glavine 1
Jeff Kent 16
Greg Maddux 0
Edgar Martinez 18
Don Mattingly 3
Fred McGriff 11
Mark McGwire 9
Jack Morris 5
Mike Mussina 30
Rafael Palmeiro 2
Mike Piazza 4
Tim Raines 12
Curt Schilling 11
Lee Smith 14
Sammy Sosa 2
Frank Thomas 4
Alan Trammell 12
Larry Walker 14
Other (Alou, Nomo) 2
Total Ballot Casualties 189


Based on these results, what can we say with absolute certainty?

Craig Biggio was named on 427 of 571 ballots. There were also at least six additional voters who believe that Craig Biggio is worthy of making the Hall of Fame but left him off of their ballots because of the 10-man ballot cap. Added together, that is at least 433 of 571 voters who wanted to vote Craig Biggio into the Hall of Fame this year, 75.8% of the electorate.

It seems clear to me that Craig Biggio should have been elected to the Hall of Fame this year. The standard for election is 75% and we know with absolute certainty that more than 75% of the electorate believe that Craig Biggio belongs in the Hall of Fame.

Craig Biggio's failure to be elected to the Hall of Fame this year is a clear failure of the BBWAA's electoral process that has resulted in the will of the electorate being thwarted.

What We Can Guess
Biggio's election is the only result that changes based purely on what I call Known Ballot Cap Casualties. But that doesn't mean that there aren't additional things that we can guess.

Looking at the above numbers, here are some observations.

The Most Frequently Mentioned Cap Casualty
The player named most frequently as a deserving Hall-of-Famer left off of a ballot because of the 10-man limit was Mike Mussina. And there really wasn't a close second.

To be honest, this result really surprised me. I think that voter Mark Zuckerman perhaps explains voters' issue with Mike Mussina best:
"A colleague of mine in the Baltimore-Washington sports media corps has insisted for years that Mussina was a Hall of Famer, a statement I always greeted with mock. (Mock directed at my colleague, mind you, not Mussina.) I just didnít think of him as a Hall of Famer, and I was sure the overwhelming majority of fellow voters would feel the same way. Well, a funny thing happened once I actually started examining Mussinaís career last month: I realized he truly does have a case. He may not have the awards and reputation of Tom Glavine, but the two were quite similar pitchers. No, he never won the AL ERA title, but he finished in the top five an impressive seven times. He has one of the best strikeout-to-walk ratios in history. Heís got a strong postseason record. And, yes, he pitched his whole career in hitter-friendly ballparks against tough AL East lineups. (Though I think that last point is exaggerated a bit. He didnít have to face the Oriolesí explosive lineup while pitching for Baltimore, and he didnít have to face the Yankeesí modern murdererís row while pitching for New York.) But I will admit I seriously considered this guy and might very well have voted for him Ö if not for that pesky, 10-player limit rule. Mussina was definitely the 11th guy on my ballot. If I was allowed to vote for 11 players, I very well might have included him. But Iím only allowed to vote for 10, so he unfortunately got caught in the numbers game."
Basically, for a lot of voters, Mike Mussina failed the "feels like a Hall-of-Famer" test and when your ballot is already too crowded, you have to start looking for reasons to exclude guys and not "feeling" like a Hall-of-Famer is an excuse. Not a very good one - as Zuckerman acknowledges - but you take what you can get.

There might be a similar factor at play with respect to the second-most frequently named Cap Casualty, Edgar Martinez. Martinez also has sort of a ready-made criticism - he was a DH, which some voters don't necessarily view as a "real" position. Again, it's not a particularly good reason to exclude somebody and it's not absolutely decisive (voters did just elect Frank Thomas, after all) but when you have to cut one of your 11 choices, it's perhaps something easy to latch onto. Thomas also appearing on this ballot may have also been a factor: maybe some voters were reluctant to vote for two DHs at the same time, although I didn't read anything that came right out and said that.

I suspect that Mussina was also hurt by this being his first year on the ballot. Another first-year player, Jeff Kent, was third in snubs with 16.

Voter Strategy
Looking at the table of Known Ballot Cap Casualties above, voters tended to leave off the names of players who seemed to have less chance of actually being elected (although, when I actually tried to put some math behind that sentence, the correlation between 2013 vote percentage and Known Ballot Cap Casualties is surprisingly weak).

What Might the Voting Results Have Looked Like If There Was No Ballot Cap?
I'm a numbers guy. So, enough speculating and talking, let's put some meat on the bones. Based on what we know, what is the best guess of what the 2014 Hall-of-Fame election results might have looked like if there had been no ballot cap?

The next table shows the distribution of Known Hall-of-Fame ballots by the number of votes. The first column shows the actual numbers; the second column includes the 189 Ballot Casualties identified above.

Names Actual No Cap
0 1 1
1 1 1
2 0 0
3 7 7
4 16 16
5 11 11
6 9 9
7 20 20
8 35 35
9 21 21
10 168 104
11 - 15
12 - 18
13 - 11
14 - 10
15 - 6
16 - 0
17 - 0
18 - 2
19 - 1
20 - 1


Even identifying "known" ballot casualties, the number of ballots with exactly 10 names is greater than the combined number of ballots with 8, 9, 11, or 12 names. If the 10-man ballot cap were removed, that result would seem highly unlikely to me. To smooth out the distribution, then, I doubled the number of ballots with more than 10 names. The resulting distribution is shown in the next table.

Names Known
(estimated)
0 1
1 1
2 0
3 7
4 16
5 11
6 9
7 20
8 35
9 21
10 40
11 30
12 36
13 22
14 20
15 12
16 0
17 0
18 4
19 2
20 2


The most popular ballot size here is still 10, but the number of ballots with 7, 8, or 9 names is similar to the number of ballots with 11, 12, 13, or 14 names. The general distribution looks fairly reasonable to me.

As was shown in the very first table, the number of names per ballot is somewhat greater on Known Ballots than Unknown Ballots, by about 0.4 names per ballot. Hence, the distribution of Unknown Ballots by the number of voters must differ somewhat from the distribution of Known Ballots. It seemed to me that the easiest way to adjust the Known Ballot distribution to approximate a distribution for Unknown Ballots was to take the difference in average ballot size - 0.41 - treat that as a percentage, and reduce the ballot size of 41% of Known Ballots by one name.

That implies a ballot distribution for Unknown Ballots as follows.

Names Unknown
(estimated)
0 0
1 2
2 3
3 10
4 14
5 10
6 13
7 25
8 29
9 79
10 23
11 17
12 21
13 13
14 12
15 7
16 0
17 0
18 2
19 1
20 1


That creates a somewhat quirky result that the Known Ballot spike at 10 shows up as a spike in Unknown Ballots at nine names. There's no real reason why we should expect that, so to smooth things out, I shifted 10 ballots from 9 names to each of 7, 8, 10, 11, and 12 names. That leaves the average ballot size unchanged (given the 10-man ballot cap) but creates a somewhat smoother distribution. The overall assumed distribution, then, for Known, Unknown, and Overall Ballots looks as follows.

Estimated Ballot Distribution
Names Known Unknown Overall
0 1 0 1
1 1 2 3
2 0 3 3
3 7 10 17
4 16 14 30
5 11 10 21
6 9 13 22
7 20 35 55
8 35 39 74
9 21 29 50
10 40 33 73
11 30 27 57
12 36 31 67
13 22 13 35
14 20 12 32
15 12 7 19
16 0 0 0
17 0 0 0
18 4 2 6
19 2 1 3
20 2 1 3


Based on what we know, that looks like a pretty reasonable distribution to me.

This ballot distribution implies a total of 624 Ballot Cap Casualties. These were distributed across players based on the sample distribution of Known Ballot Cap Casualties identified above.

Putting all of that together would have produced the following vote percentages for the 2014 Hall-of-Fame ballot. Actual vote percentages for both 2014 and 2013 are shown in the same table.

Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Jeff Bagwell
59.5%
54.3% 59.6%
Craig Biggio
78.3%
74.8% 68.2%
Barry Bonds
35.9%
34.7% 36.2%
Roger Clemens
36.6%
35.4% 37.6%
Tom Glavine
92.5%
91.9% -
Jeff Kent
24.5%
15.2% -
Greg Maddux
97.2%
97.2% -
Edgar Martinez
35.6%
25.2% 35.9%
Don Mattingly
10.0%
8.2% 13.2%
Fred McGriff
18.0%
11.7% 20.7%
Mark McGwire
16.3%
11.0% 16.9%
Jack Morris
64.3%
61.5% 67.7%
Mike Mussina
37.7%
20.3% -
Rafael Palmeiro
5.6%
4.4% 8.8%
Mike Piazza
64.4%
62.2% 57.8%
Tim Raines
53.1%
46.1% 52.2%
Curt Schilling
35.6%
29.2% 38.8%
Lee Smith
38.0%
29.9% 47.8%
Sammy Sosa
8.4%
7.2% 12.5%
Frank Thomas
86.0%
83.7% -
Alan Trammell
27.8%
20.8% 33.6%
Larry Walker
18.2%
10.2% 21.6%
Other (Combined)
5.3%
4.2% -
Total Ballots
9.49
8.39 6.60


Finally, rearranging the table to rank the players by estimated cap-free 2014 vote percentage, we get the results which I first showed earlier in this article.

Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Greg Maddux
97.2%
97.2% -
Tom Glavine
92.5%
91.9% -
Frank Thomas
86.0%
83.7% -
Craig Biggio
78.3%
74.8% 68.2%
Mike Piazza
64.4%
62.2% 57.8%
Jack Morris
64.3%
61.5% 67.7%
Jeff Bagwell
59.5%
54.3% 59.6%
Tim Raines
53.1%
46.1% 52.2%
Lee Smith
38.0%
29.9% 47.8%
Mike Mussina
37.7%
20.3% -
Roger Clemens
36.6%
35.4% 37.6%
Barry Bonds
35.9%
34.7% 36.2%
Edgar Martinez
35.6%
25.2% 35.9%
Curt Schilling
35.6%
29.2% 38.8%
Alan Trammell
27.8%
20.8% 33.6%
Jeff Kent
24.5%
15.2% -
Larry Walker
18.2%
10.2% 21.6%
Fred McGriff
18.0%
11.7% 20.7%
Mark McGwire
16.3%
11.0% 16.9%
Don Mattingly
10.0%
8.2% 13.2%
Sammy Sosa
8.4%
7.2% 12.5%
Rafael Palmeiro
5.6%
4.4% 8.8%
Other (Combined)
5.3%
4.2% -
Total Names per Ballots
9.49
8.39 6.60
Players in bold would have been elected under 75% election standard. Players in italics would have failed to meet 5% criterion for staying on the ballot.

Are These Results Reasonable?
In 2013, the average voter voted for 6.6 players. Because nobody was elected, the only players who came off of voters' ballots were Dale Murphy (18.6%), Bernie Williams (3.2%), and first-year players who failed to reach 5% and received a combined 52 votes (9.1%).

So, even with no ballot cap, my estimate is that returning players would have lost support as a group (of about 1.4% per player). Does that make sense?

Might I Be Under-Estimating the Number of Ballot Cap Casualties?

On the one hand, that seems to suggest that I might be under-estimating the actual number of ballot cap casualties. And, indeed, I might.

For example, consider Mike Nadel. He wrote a column that walked through his thought process in filling out his ballot. Near the end, he writes the following:

"Finally, it got a little more difficult. Should I stop at nine? Or should I add one more to reach the maximum votes we can cast?"
Nadel answered his final question "Yes", and voted for Mike Mussina to fill in his 10th spot.

But what if somebody answered the same question "No"? I only checked for "ballot cap casualties" from voters who actually hit the ballot cap. But what if a voter would have voted for 12 players but couldn't choose which of his final three choices deserved the 10th spot and ended up dropping all three and only voting for nine players? That seems possible.

For example, two voters had a full ballot and included one of Bonds or Clemens and explicitly named the other one as a ballot cap casualty. What if somebody had nine names other than Bonds and Clemens and decided it was easiest to sacrifice them both? Take away the ballot cap and maybe they both get an extra vote.

I skimmmed through articles from voters who listed nine players. The only one I found that was close to this was Jack McCaffery who wrote that "Mattingly, Schilling, Trammel, Smith, Bagwell, Raines and others I have voted for previously belong in Cooperstown too". McCaffery only voted for 9 but "filled" his ballot by insisting on including Pete Rose. I have not included any of the above players as "Ballot Cap Casualties" of McCaffery's ballot, since, despite what he might think, his vote for Pete Rose doesn't count, so he didn't actually reach the BBWAA's ballot cap.

But the existence of McCaffrey's ballot as well as the possibility that one's ballot could be affected by the ballot cap even if it wasn't filled could suggest that I am under-estimating the true number of ballot cap casualties and, therefore, under-estimating the vote percentages that players might have gotten in the absence of the ballot cap.

Or Would Support for Returning Players Have Declined Even Without the Ballot Cap?
Having said all of that, though, some voters impose their own personal ballot limits, and would undoubtedly do so regardless of the BBWAA's limit. For example, Mark Purdy dropped Biggio, Morris, and Raines from his 4-person ballot because he "only vote[s] 3 or 4 every year". Marty Noble explained his ballot by saying, "I don't want 28 people entering the Hall at once, so I limited my checks on the ballot to three."

There is no reason to think that Purdy, Noble, or other voters like them would have changed either their approach or their final ballot if the ballot cap was eliminated.

Overall, I think the results here are pretty solid. They're estimates, of course, but I think they are very realistic estimates that are built upon a solid factual foundation.

Analysis of Final Results
So, let's take a closer look at the results.

The 2014 Hall-of-Fame Class That Should Have Been
Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Greg Maddux
97.2%
97.2% -
Tom Glavine
92.5%
91.9% -
Frank Thomas
86.0%
83.7% -
Craig Biggio
78.3%
74.8% 68.2%


As I already mentioned, if the ballot cap had not been in place in 2014, the BBWAA would have elected four players, with Craig Biggio receiving an estimated 78.3% of the vote.

As for the other three players who were elected even in spite of the ballot cap.
No Known Voters identified Greg Maddux as a ballot cap casualty (although comment #1928 here suggests that there were "at least three or four voters who did just this"). Hence, my estimate of Maddux's uncapped vote percentage is identical to his actual vote percentage (97.2%). I suspect, however, that Maddux would have done a bit better on an uncapped ballot and might have come closer to the record for overall vote percentage (98.8%).

Tom Glavine would have received an estimated 92.5% of the vote in the absence of the ballot cap (versus 91.9% actual). This would have almost exactly matched Jim Palmer's vote percentage (92.6% in 1990). I wrote an article comparing Glavine and Palmer as part of my 2014 Hall-of-Fame ballot series last month.

Frank Thomas would have received an estimated 86.0% of the vote in the absence of the ballot cap (versus 83.7% actual). This is similar to the first-ballot percentages of Paul Molitor (85.2% in 2004), who was the first Hall-of-Fame player who played more games at DH than at any other single position, and Eddie Murray (85.3% in 2003), who was the last first-ballot electee with 500 or more career home runs before Thomas.


On the Path to Induction
Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Mike Piazza
64.4%
62.2% 57.8%
Jeff Bagwell
59.5%
54.3% 59.6%
Tim Raines
53.1%
46.1% 52.2%


In 2013, nobody was elected to the Hall of Fame by the BBWAA, but five players received over 50% of the vote, a level of support that, historically, has led to eventual induction.

As noted above, in the absence of the ballot cap, one of these five players, Craig Biggio, would have been elected this year.

Jack Morris's Hall-of-Fame candidacy before the BBWAA has now expired. He is dicussed in the next section, which looks at pitchers on the 2014 Hall-of-Fame ballot.

That leaves three players who were on a clear path toward eventual Hall-of-Fame induction coming into 2014: Jeff Bagwell, Mike Piazza, and Tim Raines.

Jeff Bagwell finished third in his third season in 2013 balloting, behind only Biggio and Morris, with 339 votes (59.6%). In actual 2014 balloting, Bagwell's support declined by 29 votes (to 310) and 5.3% (to 54.3%). Including estimated ballot cap casualties, I estimate that Jeff Bagwell would have received 340 votes (59.5%). In other words, but for the ballot cap, Jeff Bagwell's support would have held constant in 2014. This is not necessarily ideal in terms of Bagwell moving toward eventual induction, but one year not gaining votes in the wake of a strong freshman class is neither fatal nor unusual. For example, Tony Perez saw his support decline in both his 4th and 8th years on the ballot (1995, 1999), but was able to more than make up these losses the next year, gaining 16.4% of the vote from 1999 to 2000 to win election in his 9th year on the ballot.

Mike Piazza was one of only two returning players who saw his actual support increase from 2013 to 2014, with his votes increasing from 329 (57.8%) to 355 (62.2%). Without the ballot cap, I estimate that Piazza's support in 2014 would have been 64.4%. In the absence of the ballot cap, Piazza would seem to be following Carlton Fisk's path to induction with a one-year lag. Fisk debuted in 1999 on a strong ballot that included three first-year inductees with 66.4% of the vote. Fisk was elected the next year with 79.6% of the vote. Hopefully, the BBWAA will do away with the ballot cap this year and Mike Piazza can follow Fisk's lead and be inducted into the Hall of Fame as early as next year.

Tim Raines broke 50% in balloting in 2013 for the first time in his sixth appearance on the BBWAA's Hall-of-Fame ballot. In actual 2014 voting, Raines's vote total declined by 36 (6.2%). Raines was one of the more frequent players named as a Known Ballot Cap Casualty, being named 12 times. Projecting that up, that leads to an estimated vote percentage for Raines of 53.1% if the BBWAA had no ballot cap. This would have been Raines's fifth straight season of increased support in Hall-of-Fame voting. This is somewhat similar to the path taken by Raines's former Expos teammate, Andre Dawson, who gained steadily (outside of a downward blip in 2007 voting). Dawson received 65.9% of the vote in his seventh year on the ballot (2001) before being elected two years later. Raines is a bit behind that pace but is hopefully still on a path to eventually join Dawson in Cooperstown.
Pitchers
Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Jack Morris
64.3%
61.5% 67.7%
Lee Smith
38.0%
29.9% 47.8%
Mike Mussina
37.7%
20.3% -
Curt Schilling
35.6%
29.2% 38.8%


Even without a formal ballot limit, it's only natural to re-evaluate one's standards when new players appear on the ballot. Consider Jack Morris. The primary statistic driving Jack Morris's Hall-of-Fame candidacy is pitcher wins - 254 career wins, most in the 1980s, 7 World Series wins. Setting aside Roger Clemens (as 62.4% of voters did), Jack Morris had the most wins of any pitcher on the 2013 Hall-of-Fame ballot.

But in 2014, in addition to Clemens, Morris also had fewer career wins than three newcomers to the ballot: Greg Maddux (355), Tom Glavine (305), and Mike Mussina (270).

It seems perfectly natural to think that suddenly having three new players show up who beat Jack Morris at his strength might cause some voters to reconsider. And, in fact, by my estimate, even eliminating the ballot cap, Jack Morris's vote total declined by an estimated 3.4% from 2013 (67.7%) to 2014 (64.3%).

Curt Schilling also saw a similar decline - from 38.8% to 35.6% (3.2%) - likely for similar reasons: he went from the strongest steroid-free sabermetric case to perhaps as low as the fourth-best case, behind three first-year pitchers.

As I mentioned earlier, Mike Mussina was named more often than anybody else as a player that voters would have voted for but for the ballot cap. Adjusting for estimated ballot cap casualties produces what I think is the most interesting result of any here. In reality, Mike Mussina received the 7th-most votes of any pitcher on the 2014 Hall of Fame ballot with 116 votes (20.3%). But with no ballot cap, I estimate that Mike Mussina would have received 215 votes which would have placed him fifth among pitchers on the ballot, and only two votes out of fourth place.

The pitcher with two more votes than Mussina on my estimated 2014 Hall-of-Fame ballot is relief pitcher Lee Smith. Even with no ballot cap, I estimate that Lee Smith's support would have declined in 2014 (for the second year in a row) from 47.8% to 38.0%, after declining from 50.6% in 2012. It seems that, upon further review, many BBWAA voters are concluding that Lee Smith is not a Hall-of-Famer.

Steroid Users
Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Roger Clemens
36.6%
35.4% 37.6%
Barry Bonds
35.9%
34.7% 36.2%
Mark McGwire
16.3%
11.0% 16.9%
Sammy Sosa
8.4%
7.2% 12.5%
Rafael Palmeiro
5.6%
4.4% 8.8%


The 2014 ballot is the 8th ballot on which Mark McGwire has appeared. I estimate that, had there been no ballot cap, Mark McGwire would have received 93 votes this year. That would have made Mark McGwire's Hall-of-Fame vote totals by year since his debut in 2007 the following, in order: 128-128-118-128-115-112-96-93.

The evidence suggests that, if anything, anti-steroid sentiment is growing over time among BBWAA Hall-of-Fame voters. And my best guess is that this is not merely an artifact of the ballot cap.

There were 22 Hall-of-Fame candidates on this year's ballot who were either holdovers from earlier years or who received enough support this year to carry over to next year's ballot. On average, these players were identified as a Known ballot cap Casualty 8.5 times apiece.

There are five players among those 22 who are most strongly associated with steroids: Barry Bonds, Roger Clemens, Mark McGwire, Rafael Palmeiro, and Sammy Sosa. These five players were identified as Known Ballot Cap Casualties a total of 17 times, or 3.4 apiece, with over half of those being McGwire. The other four of them - Bonds, Clemens, Palmeiro, and Sosa - were identified as Known Ballot Cap Casualties only twice each.

Voter Jim Henneman offers some insight, I think, into why this might be the case.
"Although I'm not ready to commit to never voting for someone who has been tied to performance-enhancing drugs, for the second straight year I did leave off my ballot the five names most prominently mentioned in that regard -- Barry Bonds, Roger Clemens, Mark McGwire, Rafael Palmeiro and Sammy Sosa -- and had little trouble doing so, given the qualifications of at least another dozen or so candidates." (emphasis added)
It's a difficult and uncomfortable issue for a lot of people. And if you can avoid difficult, uncomfortable issues, most people will do so (that's not a criticism; it's just a fact). If a voter started filling out his ballot, saving the steroid guys for the end, if he had already checked 10 names by the time he got to the steroid guys, why bother to think about whether you'd vote for them if you had more space? And, then, why bother talking about the steroid guys in your column if your willingness to vote for or against them didn't matter to your final ballot?

Removing the ballot cap could, therefore, prompt some writers to re-evaluate their stance on PEDs and the players associated with them. And the extent to which these players may see their votes increase if the ballot cap was removed could be greater than the numbers here might suggest. But that's pure speculation on my part.

All of that said, even with the small number of voters to identify steroid users as Known Ballot Cap Casualties, there is one result here of at least some significance. Rafael Palmeiro's estimated vote percentage increases by only 1.2% with the elimination of the ballot cap, but this is just enough to push him across the 5% threshold for staying on the ballot.

Everybody Else
Percentage of Vote
Player Estimated
Uncapped 2014
Actual 2014 Actual 2013
Edgar Martinez
35.6%
25.2% 35.9%
Alan Trammell
27.8%
20.8% 33.6%
Jeff Kent
24.5%
15.2% -
Larry Walker
18.2%
10.2% 21.6%
Fred McGriff
18.0%
11.7% 20.7%
Don Mattingly
10.0%
8.2% 13.2%


That leaves six players who received a meaningful number of Hall-of-Fame votes who I have not discussed.
Edgar Martinez is estimated by me to have received 35.6% of the vote in 2014 in the absence of the ballot cap. That would be virtually identical to his 2013 vote total of 35.9%. So far, Edgar Martinez has appeared on five Hall-of-Fame ballots. His vote totals on the first four of those ballots ranged from 32.9% (in his second year, 2011) to 36.5% (in 2012). Ultimately, even removing the ballot cap, Edgar Martinez's Hall-of-Fame candidacy appears to be stuck.

Alan Trammell appeared on his 13th Hall-of-Fame ballot in 2014. He only has two more chances left with the BBWAA. After spending eight years getting 13% - 18% support in Hall-of-Fame balloting, Trammell saw his support rise from 17.4% in 2009 to 36.8% in 2012, a fairly remarkable surge. In 2013, his support slipped from 36.8% to 33.6%. Even without the ballot cap, I estimate that his support would have slipped some more in 2014 to 27.8%. Unfortunately, Trammell is not going to be elected to the Hall of Fame by the BBWAA despite being highly deserving of the honor.

Jeff Kent debuted on the 2014 ballot with 15.2% support. That is very similar to Alan Trammell's debut support (15.7% in 2002) and it is possible that Jeff Kent will suffer the same Hall-of-Fame fate as Trammell. But if there had been no ballot cap, I estimate that Jeff Kent would have received 24.5% of the vote in his debut year. This is still a total that far from guarantees eventual election. On the other hand, it is not that much lower, for example, than Jim Rice's debut in 1995 of 29.8%. As such, it is probably too early to say for sure how Jeff Kent might fare in future Hall-of-Fame voting.

Larry Walker, Fred McGriff, and Don Mattingly all saw their support plummet by 5-10% from 2013 to 2014. Much of that was the ballot cap in all three cases, but not all of it. Without the ballot cap, I estimate that Walker, McGriff, and Mattingly would have still seen declines in their vote percentages of 3.4%, 2.7%, and 3.2%, respectively. This is likely indicative of the fact that none of the three of them have shown any positive trends in ballot support over time which could be leading some of their supporters to abandon them.


All articles are written so that they pull data directly from the most recent version of the Player won-lost database. Hence, any numbers cited within these articles should automatically incorporate the most recent update to Player won-lost records. In some cases, however, the accompanying text may have been written based on previous versions of Player won-lost records. I apologize if this results in non-sensical text in any cases.

Home     List of Articles