Final thoughts on the 2015 Hall of Fame election

The Hall of Fame results were announced Tuesday. Randy Johnson, Pedro Martinez, John Smoltz, and Craig Biggio will be heading to Cooperstown at the end of July for induction. This is my last post discussing some of the trends behind the voting, some of which amused me, others caused me alarm.

Voters are casting more votes

With a maximum of ten candidates allowed per ballot, most voters should have been able to select all the candidates they wanted. It's difficult now with a backlog of qualified candidates building up due to small classes between 2008 and 2013, but this is the second consecutive year in which writers placed an average of 8+ players on their ballots:

There are no guarantees going forward, of course, but voters are beginning to recognize there are more players worthy of selection. The voters missed by a couple of votes selecting four players in 2014, and by doing so in 2015 demonstrated that they can vote in larger classes.

The PED issue isn't going away

It's becoming increasingly clear that Barry Bonds and Roger Clemens are facing an uphill battle for enshrinement. This was the third year on the ballot for both, and they received almost the same number of votes, with Clemens getting 206 and Bonds 202 votes. Of the 220 ballots that had been made public via Ryan Thibs' Hall of Fame Tracker, this is how voters treated Bonds and Clemens:

Player	Votes	Other?
Bonds	97	95
Clemens	96	95

That's lockstep voting and demonstrates a very clear line of demarcation for the voters in that players are either in or they aren't. I don't see anything changing this, and I frankly don't have an answer -- it's simply impossible to close a chasm this stark and gaping. I've written it numerous times that I have little doubt they will be enshrined by some future Veterans' Committee (and quite likely along with Rafael Palmeiro), and I'm willing to leave it at that.

For them. I'm not willing to drop it regarding Mike Piazza and Jeff Bagwell. The line is just as clear with them, since voters who voted for Piazza and Bagwell also voted for Clemens and Bonds in similar patterns. Neither Bagwell nor Piazza were listed in the Mitchell Report, and the only charges of PED use are innuendo and guilt by association. I'm not stating categorically they didn't use PEDs -- Piazza has admitted using androstenedione prior to it being banned. Two articles, one by Bill Madden on Piazza specifically and the other by Bob Nightengale more generally, show how big the divide is.

I understand the issues with Bonds and Clemens but am puzzled by Bagwell and Piazza. It appears the standards are extremely malleable, suggesting there is no standard at all. I suspect Bagwell and Piazza can make it in next year, since there's only one certain candidate (Ken Griffey Jr.) on the ballot, but there will be more players coming (Manny Ramirez and Ivan Rodriguez) who will undergo the same scrutiny. Every voter has the right to make up his or her mind -- I just hope they begin to separate fact from innuendo.

The rise of advanced stats

I wasn't particularly worried about Pedro Martinez being selected, but at the same time I was a little bit concerned. Living in the alternate universe of different statistics like Fielding Independent Pitching (FIP), Weighted On-Base Average (wOBA) and the like, I sometimes lose track of how far these types of analysis have seeped into the general baseball discussion. Martinez' record is good even on its superficial surface -- his wins (219) are 76th all-time and his .687 win percent is fifth all-time among pitchers with at least 200 starts. But those numbers don't even begin to define his dominance.

Martinez' stretch between 1999 and 2003 was beyond ridiculous when viewed in terms of newer measures. I like using FIP- since it breaks down what a pitcher can control, the number of strikeouts, home runs and walks he gives up, as well as normalizing those numbers by year, league and park. It's not perfect since it rewards pitchers of recent vintage who strike out more batters, but no single number is. These are the best five-year stretches by FIP- in baseball history:

Pitcher	From	To	W	L	GS	CG	SHO	IP	K	BB	HR	FIP-	ERA-	K/9	BB/9
Pedro Martinez	1999	2003	82	21	135	18	5	931.6	1205	181	51	43	45	11.6	1.7
Pedro Martinez	1998	2002	87	24	139	18	7	978.6	1250	201	70	49	48	11.5	1.8
Pedro Martinez	1997	2001	84	28	140	29	11	1020.6	1316	228	73	50	47	11.6	2.0
Pedro Martinez	2000	2004	75	26	139	14	5	935.5	1119	205	68	54	54	10.8	2.0
Randy Johnson	2000	2004	86	40	157	24	12	1116.6	1468	289	102	54	59	11.8	2.3
Randy Johnson	1998	2002	100	38	173	41	17	1272.7	1746	374	121	56	58	12.3	2.6
Randy Johnson	1997	2001	96	37	167	38	15	1225.7	1703	380	115	56	57	12.5	2.8
Pedro Martinez	1996	2000	90	35	155	32	12	1120.6	1375	273	87	57	54	11.0	2.2
Randy Johnson	1999	2003	87	35	157	32	12	1142.6	1542	315	114	57	58	12.1	2.5
Randy Johnson	1995	1999	79	26	136	33	13	1003.5	1363	323	93	59	58	12.2	2.9
Greg Maddux	1994	1998	87	32	155	39	14	1139.4	890	147	45	59	50	7.0	1.2
Greg Maddux	1993	1997	89	33	157	38	10	1155.4	883	154	46	60	51	6.9	1.2
Pedro Martinez	2001	2005	72	28	141	11	2	935.5	1043	220	70	60	62	10.0	2.1
Randy Johnson	1996	2000	80	31	141	35	13	1037.6	1416	334	104	60	59	12.3	2.9
Randy Johnson	1994	1998	75	23	124	30	15	904.3	1203	325	77	60	61	12.0	3.2

To explain, in his 1999-2003 stretch Martinez had a FIP that was fifty-seven percent better than the American League. At the height of the best offensive era baseball has seen since at least the 1930s (and possibly ever), Martinez put up the best five-year span of pitching dominance, and the top four spans of all-time also belong to him. I first noticed this last year in a different regard when I saw that he had several of the best ERA+ seasons ever and was curious how well this would translate into HOF voting. It appears I had nothing to worry about.

This table shows how many five-year spans pitchers have had with FIP- of 79 or lower, suggesting they were twenty percent or better than their league over a sustained period of time (min 750 IP):

Pitcher	Spans
Roger Clemens	19
Randy Johnson	14
Lefty Grove	12
Greg Maddux	11
Curt Schilling	11
Roy Halladay	11
Pedro Martinez	11
Walter Johnson	11
Cy Young	10
Christy Mathewson	9
Rube Waddell	9
Tom Seaver	9
Kevin Brown	9

I show this also to place Curt Schilling and Roy Halladay in context. I'll return some time in the future to new ways to measure pitcher performance other than wins, because that time is upon us. I'll show this table until I'm blue in the face, the career wins leaderboard among active pitchers, led by Tim Hudson at 214 (and probably not many more) and CC Sabathia at 208. And before I leave the subject, Sandy Koufax between 1962-1966 had a 66 FIP-. He's "hurt" by having "only" 9.4 strikeouts per nine innings and playing in very pitcher-friendly Dodger Stadium. I'm not diminishing it, just placing it in context (see more of seasons like this in this Google Docs sheet or in this Tableau data viz).

Final Thoughts

I really appreciated Ryan Thibs' tracker and will definitely return to it again next year, and I also give kudos to the 200+ voters who were willing to make their ballots public. I don't pretend to understand them all -- one voter returned a blank ballot, another voted for only Alan Trammell and Lee Smith, and I'm lost as to what point was being made.

This graph shows the number of ballots for each possible number of votes for the publicly released ballots:

Voters with nothing to hide will release their ballots, voters who vote for zero or one candidate probably won't, but it's still heartening to see the trend toward more names on a ballot.

Magic numbers also seem to be going away, particularly on offense. Carlos Delgado dropping off the ballot didn't surprise me much, but the complete lack of traction for Gary Sheffield caught me off-guard. Noted baseball analyst Tom Tango had this to say:

Brian Giles: latest 50 WAR player without a single vote on a HOF ballot. Mark Langston will pour you a drink.

— Tangotiger (@tangotiger) January 7, 2015

Voters aren't just looking at numerical thresholds and selecting those who surpass them, and I consider that a good thing.

The Hall of Fame is a museum that's taken on outsize influence, but it provides a conversation piece for baseball when the temperature won't get above zero in Iowa today. It also facilitates my favorite type of discussion, one that compares and contrasts players, since that's what I've done ever since I held my first baseball card (1966 Topps). I looked at the stats on the back and said to myself "I wonder how that compares to someone else?" It's fun and generates discussion, and there was no shortage of it yesterday. Congratulations to the inductees, and thanks to the voters for recognizing their excellence.

Data from Baseball-Reference. 2015 voter data from Ryan Thibs' HOF Tracker.

Scott Lindholm lives in Davenport, IA. Follow him on Twitter @ScottLindholm.