Rethinking What Makes a Hall of Fame Starting Pitcher, Part 3: Can We Make a New Pitching Milestone?

The last two articles in this impromptu mini-series have been building to this question: since the Hall of Fame has been struggling to induct starting pitchers, is there a big, easy-to-understand metric that could serve as a new milestone for Hall of Fame voters and writers? I think that this is a particularly important thing in the context of Hall voting, given that such a large chunk of the electorate does rely on more “traditional” ways of thinking about their votes, and giving a big, round number to easily signal “this is a historic player!” could go a long way in helping those voters see what the rest of us are seeing. In fact, it’s something I’ve tried before in other areas, like when I tried to determine if there was a milestone for Walks to match 3000 Hits and 500 Homers, or when I repeatedly tried to emphasize how Scott Rolen was a top-ten all-time player at his position.

Built with ConvertKit

But as part of answering this, I felt like I had to tackle the Hall’s recent history and their existing standards, and I found that the voters have largely just looked at pitcher wins, with a little bit of focus on strikeouts sometimes. And their recent logjam seems to be in part triggered by the Hall ballots in the 1990s that were heavy on milestones; six pitchers with big milestones under their belt came up for induction in short order, and rather than hurrying to get the obvious picks inducted, Hall voters instead seemed to respond to this by delaying the induction of “weaker” 300 win/3000 strikeout guys, and ignoring everyone else.

However, this was already kind of a big shift in voting behaviors, as before then, the voters would regularly induct players with less than 300 wins and 3000 strikeouts. So part of the answer to our overarching question is just “get voters to accept that not every starting pitcher needs to be a big milestone guy”.

Of course, that’s only one of the two sides of the issue, so having another milestone to point to could be useful, especially given the increasing rarity of 300-Wins pitchers. If some number of old-school voters want milestones, we might as well try to find another milestone for them, to help avoid this type of logjam in the future. And one thing that I noted last time was that, with the attention given to Jack Morris and subsequent candidates like Roy Halladay, it seemed like some of them might have been starting to move towards “most wins over a decade span”.


Of course, Pitching Wins are still riddled with problems as an indicator of skill, but highlighting how a pitcher stacks up against their contemporaneous peers is a useful idea. And we actually do have a basic stat for measuring the best pitchers in a season, a very famous one that even the stodgiest of voters will understand, at that: the Cy Young Award. Of course, there’s the obvious issue of “there are only 2 of those a year”; how do we separate, say, the guys who won for a single fluke year, versus the players who racked up runners-up but never won? A simple “Winner/Not Winner” binary doesn’t really get us the granularity we need here.

And that is where Award Shares come in. For those who aren’t familiar, Award Shares are basically a modification of the existing Award System that gives credit to any player who received votes, not just the winners. Winners of course get the biggest return (especially dominant winners), but now at least your perpetual runners-up aren’t coming away empty-handed. Also nice is that this system works with basically any voting system, so you can use it for MVPs, Cy Young Awards, other sports, really anything as long as the full votes are revealed.

A unanimous first-place win is equal to 1 full award share, and they scale down from there. So for example, in the NL MVP voting last year, Ronald Acuña Jr. got a full 1 Award Share for his unanimous win (30 first place votes, each worth 14 points) while Mookie Betts (who got all 30 second-place votes, each worth 9 points) gets 270/420, or about 0.64 Award Shares for 2023. It’s not necessary to know all of these for this piece, but the full points values for an individual award ballot are as follows (taken from the BBWAA’s own website):

MVP: 10 slots, worth 14-9-8-7-6-5-4-3-2-1 from top to bottom
Cy Young: 5 slots, worth 7-4-3-2-1 from top to bottom

So as you can see, actually getting first place votes and dominating the competition is heavily rewarded, but winning is also no longer the only thing we care about. Close elections are actually reflected here, such as 2017, where Giancarlo Stanton’s 2-point edge on Joey Votto (302 points to 300) shows up as 0.72 to 0.71, rather than 1-0.

And then, like any other stat, we can compile it over the course of a career and directly compare players’ lifetime results. In fact, Baseball-Reference has a basic leaderboard just for that!

So let’s start playing around with those leaderboards on our own, since B-R doesn’t have filters. Let’s begin with a basic question, just to test the waters: Who are the best players by MVP share who aren’t in the Hall of Fame?

Overall Rank	Name	MVPs	MVP Shares
1	Barry Bonds	7 wins	9.30
3	Albert Pujols	3 wins	6.91
4	Mike Trout	3 wins	6.53
11	Álex Rodríguez	3 wins	5.23
15	Miguel Cabrera	2 wins	4.68
19	Mookie Betts	1 win	3.80
22	Pete Rose	1 win	3.68
24	Freddie Freeman	1 win	3.44
30	Paul Goldschmidt	1 win	3.24
33	Dave Parker	1 win	3.19
36	Joey Votto	1 win	3.08
38	Manny Ramírez	0 wins	3.06

Yeah, that looks about what you’d expect. Bonds, A-Rod, Rose, and Manny all have other issues holding them back. Pujols and Cabrera will waltz in a few years from now. The active guys all look like favorites for some 2030s inductions. Parker is probably a little overrated here, but there’s a reason he held on for the full fifteen years of BBWAA ballots, and he even reached 7 votes on his third VC ballot back in 2020 (just over halfway to induction); he’s the exact type of hitter that the Hall has generally overrated, and I don’t think he’d be an egregious choice if/when he gets in. This seems like a solid idea so far, the list generally lines up with what we’d expect, and we have a better idea of the scale we’re working with.

Okay, that’s a good starting point. Let’s try it again, but the snubs on the Cy Young Shares; has our recent drought of inductions left us with a list that we can start picking from?

Overall Rank	Name	CYAs	CYA Shares
1	Roger Clemens	7 wins	7.66
3	Justin Verlander	3 wins	5.21
5	Max Scherzer	3 wins	4.61
6	Clayton Kershaw	3 wins	4.58
14	Gerrit Cole	1 win	2.90
15	Johan Santana	2 wins	2.72
16	Corey Kluber	2 wins	2.60
17	Félix Hernández	1 win	2.46
18	Jacob deGrom	2 wins	2.44
19	Bret Saberhagen	2 wins	2.20
20	David Price	1 win	2.11
22	Zack Greinke	1 win	2.03

Hmm… I can’t quite put my finger on it, but something about this list feels different? These lists are of course always going to skew a little more modern, because the best modern players haven’t had a chance to be elected yet. And it probably doesn’t help that the hitters side is littered with special exceptions that make it seem a little older, while the pitchers only have Clemens. But still, isn’t it a little weird that a full nine of our top twelve non-inductees haven’t even been retired long enough to qualify for Cooperstown? Shouldn’t we have, like, a Tommy John or a Curt Schilling here, to match the hitters a little more closely?

And on a different note, the numbers here are definitely lower: Manny Ramirez brings up the twelfth spot for non-inducted hitters with just over 3 MVP shares, and he’s 38th all-time, but Zack Greinke is our twelfth pitcher, while placing just 22nd all-time and with just over 2 Cy Shares. What’s going on here?

Well, it’s actually due to a variety of factors, but they can largely be simplified to “there are fewer Cy Young Award votes to go around than MVP votes, and this disparity was even more pronounced the further in time you go back”. You can start to break it down by going back and looking at the voting breakdown I posted earlier; there are half as many ballot spots for Cy voters compared to MVP voters, and the point range is smaller and more compressed.* Honestly, that’s probably merited, given the differences in pitching and not-pitching, although it does mean that we’re still not going to quite get these onto a one-to-one basis.

*Just to apply actual numbers, for those who want them: MVP ballots get 59 points to distribute over 10 players, while Cy Young ballots get 17 points for 5 players. And since Award shares are based on ratios of first place points, and Cy Young voting is working on a 7-to-1 scale rather than 14-to-1, a second-place Cy vote is actually worth the same as a third-place MVP vote, a third-place Cy vote is equal to a fifth-place MVP vote, a fourth-place translates to seventh-place, and fifth-place is equal to ninth-place.

However, this set-up is also a fairly recent development, dating back to just 2010. From 1970 to 2009, it used a 3-place ballot just like the Rookie of the Year voting did and still does (with a 5-3-1 point distribution); that just feels wrong to me, since it seems like the discussion for “best pitchers” in a given year should be a little wider than the one for “best rookie player”.

But before then, things were even more dire. From 1967 to 1969, it was just a one-person ballot for each league; pick the best pitcher, and the most-named person wins. So, for example, Juan Marichal finished 5th in MVP voting in 1968, but got 0 Cy Young Shares for that season because no one was going to vote for him in 1968 over Bob Gibson. From 1956 to 1966, there was actually just one MLB-wide Cy Young Award with the same process, making the same problem even more extreme. And of course, prior to 1956, there was no Cy Young Award at all.

Compare that to the MVP Award, which has been using two league-divided ten-person ballots since it’s modern inception back in 1931 (although Baseball-Reference’s Vote shares does include earlier MVP votes as well, for completeness’ sake), and has been using the modern point distribution since 1938. Really, prior to 1970, you’re probably better off just using MVP shares for pitchers.

Okay, so our hopes for one big unified Award Shares Hall milestone is dead for the moment, and even trying for one big “Cy Young Shares” milestone is going to give us problems given that anyone pre-2010 is going to get overshadowed by modern guys. To be honest, realizing all of this wasn’t a promising start to this experiment; I even got a little disheartened about whether it was worth continuing with this series a few times, until I noticed something.

Do you remember the big table of starting pitchers on the Hall of Fame ballot? It’s featured in both of the last two articles; the first time it was just “Every recent starting pitcher who reached multiple BBWAA ballots without getting inducted by the BBWAA”, and the second time, I expanded it to include a whole bunch of interesting one-and-done guys, as well as their career strikeout and win totals.

The goal was of course to try and see what the voters were seeing, to try and figure out their process. If reaching a pitching milestone like 300 wins or 3000 strikeouts was a big part of a Hall case, wouldn’t the guys who fell a bit short of those marks be the first runners up? You’d see that some times; Tommy John (288/2245) and Jim Kaat (283/2461) struck around for 15 years on the ballot. But you’d also see guys like Ron Guidry (170/1778) or or Vida Blue (209/2175) hang around for far longer than guys who actually got closer to their marks, like David Cone (194/2668) or Jamie Moyer (269/2441) or Frank Tanana (240/2773).

But I started that project almost independently of my Cy Young Shares idea, so it didn’t occur to me to just line them up. And once I did that, I found… well, era is still the defining issue. Ron Guidry (1.91 Cy Shares) got to hang around for 9 ballots, but Bret Saberhagen (2.2) and Johan Santana (2.72) only got one chance, mostly because Guidry debuted on the ballot in 1994, while the other two made it in 2007 and 2018.

But once you account for the difference in eras to some extent, Cy Young Shares were often the missing factor! Why did Vida Blue do so much better than Jerry Koosman on the 1991 Hall ballot (“so much better” being “well enough to make 4 ballots, to Koosman’s one-and-done showing”, but that’s not nothing!), despite Koosman getting much closer to 300 wins and 3000 strikeouts? Well, Blue had 1.06 Award Shares to Koosman’s 0.62 (even more pronounced, given they played in lower-vote eras), and even if voters didn’t know that, it probably lined up with their idea that Blue had the better peak.

This kind of makes a lot of sense, too; writers vote on both the Awards and the Hall of Fame candidates. Of course there’s overlap! There’s some difference too, because memory isn’t perfect, and not every voter weighs in on every award, or maybe they change their mind or just forget how good a guy was over the years. Or sometimes, the split is that our understanding of pitching just takes a step forward, like when Cy Young voters began to de-emphasize win totals around 2010. This isn’t going to be a one-step fix, but then again, neither are the other milestones. I think it’s worth emphasizing this if only to give hard numbers to more amorphous feelings about “dominance”, and maybe refresh voters’ memories on pitchers they’ve forgotten over the years.

Anyway, returning to the overlaps in Cy Young Shares and Hall voting, you can see other interesting points. For example, Don Drysdale had a strong 0.7 Cy Young Shares (despite playing largely in the “one Cy Young for all of MLB” era), which probably lined up with voter’s recollection of a strong peak and helped him overcome lower career totals. Catfish Hunter cruised to induction in the ‘80s not just on the strength of one Cy Young, but 2.02 Cy Shares. And other ‘80s ballot mainstays like Mickey Lolich (0.94) and Wilbur Wood (0.70) don’t fare poorly here either. From 1998 to 2000, four zero-Cy Young Award pitchers hit the ballot in Bert Blyleven (0.45), Frank Tanana (0.23), Nolan Ryan (1.48), and Jack Morris (0.73) all hit the ballot, and their voting results largely lined up closer to their Cy Shares than their career totals (at least until Rich Lederer began publicizing Blyleven’s strengths).

But for whatever reason, Morris kind of was the last of his kind as a candidate. After him, from 2000-2011, we saw a string of failed candidates, with the limited “successes” confined to the earlier half of that span. Only Orel Hershiser (1.32), Fernando Valenzuela (1.55), and Dave Stewart (1.22) stuck around for two ballots; all did well in Cy Young Shares, plus Hershiser and Valenzuela passed the 2000-strikeout line (Stewart appears to have lucked out in this regard). Rick Sutcliffe (1.56) hit the ballot in 2000 with Morris and was slightly better than all of them in Award Shares, but didn’t reach either of the lower career numbers and was one-and-done. Dwight Gooden (1.56) tied with Sutcliffe and did pass 2000 strikeouts, but I imagine his various legal issues didn’t endear him to voters, so he fell off right behind them. Frank Viola (1.2), Jimmy Key (1.25), and Jack McDowell (1.27) were also close to those four, but missed the Win and Strikeout marks and didn’t get a second look. But then again, neither did Dennis Martinez (0.05), who hit the ballot in this span with the highest career totals of this set, but little award success to go with it.

Post-2006, things went even worse. Saberhagen (2.2) and Cone (1.38) would also go nowhere when they reached in 2007 and 2009, respectively. Saberhagen’s superlative awards were maybe countered by his low career totals (167 wins and 1715 Ks), although that previously hadn’t been an overwhelming issue for pitchers as decorated as that. Cone, meanwhile, was well over 2500 strikeouts and just shy of 200 wins, but still went nowhere. Chuck Finley filled Dennis Martinez’s role of “high career numbers, little award success” (200 Wins, 2610 Ks, 57.9 bWAR, 0.01 Award Shares), and it got him just a single vote, fifteen less than Dennis had received. Kevin Brown (1.21) also debuted in 2011 and went one-and-done despite crossing both 200 wins and 2000 strikeouts, but he was also probably hit by external factors due to his PED ties. And of course, once the ballot crush of 2013 moved in, there was little hope for high-peak/short-career stars like Johan Santana (2.72), let alone borderline candidates like David Wells (0.55).

There’s a lot more I could go into here, and not enough space to put it all here without this article getting abnormally long. So instead, I’ll save that all for the big finale piece, which should be going up in the next week or so. If you’d like to be notified with an email when that happens, consider subscribing to the Hot Corner Harbor newsletter below!

New Email List, since Blogger broke the last one!

The old subscription service doesn't seem to be working anymore, so if you'd like to receive emails when a new Hot Corner Harbor post goes up, sign up here!

We won't send you spam. Unsubscribe at any time.

Built with ConvertKit