Tag Archives: Cy Young

2015 Baseball Hall of Fame ballot, Part 2

IMG_0125

When I left you hanging in the manner of the Hardy Boys at the end of Part 1, you had suffered through a 4,000-word post just to eliminate the flotsam from this year’s Baseball Hall of Fame ballot. I had crossed out 19 of the 34 names, which still left 15 to share the 10 votes I’m allowed. So I’ve provided the shortcut above for those who just want the bottom line and aren’t up for another 5,000 words. If that’s you, there you go, and see you around. For the rest of you, masochists that you are, here we go.

The 15 names that remained after Part 1:

Starting pitcher: Randy Johnson, Pedro Martinez, Mike Mussina, Curt Schilling.

Hybrid pitcher: John Smoltz.

Relief pitcher: Tom Gordon, Lee Smith.

Catcher: Mike Piazza.

First base: Jeff Bagwell.

Second base: Craig Biggio, Jeff Kent.

Shortstop: Alan Trammell.

Third base: Edgar Martinez. (Right. I know. But that’s how they do it.)

Left fielder: Tim Raines.

Center fielder: None.

Right fielder: Larry Walker.

I began, as I did in Part 1, with Jay Jaffe’s JAWS system to get a feel for what the quants would say. The 15 candidates are listed in order, from best to worst, based on their premium or deficit to the average JAWS score of players at their position already enshrined in the Hall of Fame. As a reminder, JAWS is the average of a player’s career wins above replacement and that player’s “peak” WAR — the total of his best seven seasons.  The theory, and I stress the word theory, is the perfect blend of longevity and awesomeness. Presumably, if you believed the JAWS method delivered on this theory, you would vote for the first 10 names on this list and be done with it, especially because the first 10, conveniently, are above the average of the players at their position already in the Hall, and the next five are below. The idea behind JAWS is to prevent dilution of the quality of players in the Hall by inducting only players equal or superior to the existing residents.

  1. Randy Johnson (+20.2)
  2. Jeff Bagwell (+9.7)
  3. Pedro Martinez (+9.3)
  4. Mike Piazza (+8.1)
  5. Alan Trammell (+2.8)
  6. Curt Schilling (+2.7)
  7. Tim Raines (+2.3)
  8. Mike Mussina (+2.0)
  9. Edgar Martinez (+1.0)
  10. Larry Walker (+0.5)
  11. Craig Biggio (-3.6)
  12. Tom Gordon (-5.1)
  13. John Smoltz (-7.6)
  14. Lee Smith (-9.0)
  15. Jeff Kent (-11.6)

In Part 1, I wrote a bit about the difference between quantitative and qualitative analysis. In this case, I am using JAWS to represent quantitative analysis and my own subjective “eye test” — did the dude look like a Hall of Famer to me — to represent the qualitative. (If you want to know why Barry Bonds and Roger Clemens, whose JAWS scores are off the charts, are not on this list, you need to go back and read Part 1, where they are listed under the category “Obvious cheaters,” and this will refer you back to my post on last year’s ballot, where I described at length my thought process on the PED issue as it relates to the Hall of Fame.)

Appraising the quantitatively-derived list with my subjective, qualitative eye, several anomalies jump out.

Anomaly No. 1: Craig Biggio.

Considered by most observers a lock for the Hall since the day he retired, Biggio, according to JAWS, was an inferior player to the average second baseman now in the Hall. He was the top vote-getter in his first year on the ballot, 2013, the year nobody was inducted, receiving 68.2 percent of the vote.  Last year, leapfrogged by first-timers Greg Maddux, Tom Glavine and Frank Thomas, he received 427 votes from the 571 voters, or 74.8 percent. Had he received two more votes, he would have joined Maddux, Glavine and Thomas in the class of 2014.

Significantly, Biggio’s JAWS score is inferior to that of Lou Whitaker, the under-appreciated longtime Tigers second baseman who failed to receive the minimum 5 percent required to stay on the ballot in his sole appearance in 2001. That, of course, is a crime against nature and baseball, as is the continuing under-appreciation of his teammate and double-play partner, Alan Trammell, on the ballot this year for the 14th time. But we’re getting ahead of ourselves.

How is it possible that Whitaker, the second-best second baseman in history not to be inducted (Bobby Grich is the first) according to JAWS, would get 2.9 percent of the vote on his first try and Biggio, his inferior by quantitative analysis, would get 68.2? Well, as the old lady in the church says at the end of The Birdcage when asked to identify the mother of the groom, whose parents are Robin Williams and Nathan Lane, “I just don’t know.”

Prone as I am to stirring up trouble, and knowing the howls of outrage this would trigger among veteran members of the Baseball Writers Association of America, I might hypothesize that race had something to do with it — Whitaker is black, Biggio is white — and order up a quantitative analysis of voting on comparable players by race, and a parallel analysis of the racial makeup of the voting population. I don’t know what such a study might find on the first question, although I would note that Tony Gwynn and Frank Thomas sailed in a lot faster than some white comparables, and Jim Rice got in with a JAWS score lower than Minnie Minoso, Lance Berkman and Jose Cruz. On the second, my guess is the racial makeup of the voting population looks something like the racial makeup of Maine, New Hampshire and Vermont, which is why the hypothesis must be considered.

Another hypothesis would be that the absence of a significant sabermetric influence in 2001 revealed an embarrassing blind spot in qualitative analysis. Whitaker’s offensive numbers in the traditional, non-sabermetric categories — .276/.363/.426, 244 homers, 1,084 RBI — while above average for a second baseman, were nowhere near the numbers generally required of hitters who played the corner positions or the outfield. And while he was known as a smooth fielder, few voters were probably aware this would translate into 15.4 defensive WAR.

Given what’s happened to Trammell, who has been treated slightly better but has earned nowhere near the level of support the quantitative analysis would suggest, maybe it’s some strange prejudice against Detroit.

In any case, this is one of the more remarkable divergences between quantitative and qualitative analysis in the history of the Hall, and I admit to being totally on the quants’ side on this one. Whitaker was a wonderful player who belongs in the Hall and with any luck will be installed by some iteration of the veterans committee. But it is not his absence that is outrageous; it is the failure of the BBWAA to consider him even worthy of consideration.

Let’s return for a moment to our discussion in Part 1 of the bias of WAR for longevity. The career stat, as I mentioned, is basically an adding machine. A vastly complicated series of calculations and adjustments reduces everything to a single number per season — let’s say these inscrutable calculations produce four wins above your average replacement player for our guy in a particular season. And let’s say his number is somewhere around there for much of his career. His total WAR will be largely a function of how many years he plays.

JAWS attempts to mitigate the longevity bias by averaging the career WAR total with the seven-year peak WAR total, but the poor standing of Sandy Koufax in the JAWS rankings demonstrates that peak WAR’s mitigation of the longevity bias is insufficient. So let’s try something else. Let’s eliminate longevity as a factor altogether and see what happens. It’s easy enough to do. Take a player’s career WAR total and divide it by the number of years he played, yielding his average single-season WAR. And let’s establish a minimum 10 seasons, since that’s required for consideration for the Hall.

Here are the top 20 second basemen in history based on JAWS, the number in parentheses representing the average of that player’s career WAR and peak WAR:

  1. Rogers Hornsby (100.2)
  2. Eddie Collins (94.1)
  3. Nap Lajoie (83.8)
  4. Joe Morgan (79.7)
  5. Charlie Gehringer (65.6)
  6. Rod Carew (65.4)
  7. Bobby Grich (58.6)
  8. Frankie Frisch (57.4)
  9. Ryne Sandberg (57.2)
  10. Jackie Robinson (56.8)
  11. Lou Whitaker (56.4)
  12. Chase Utley (55.3)
  13. Roberto Alomar (54.8)
  14. Craig Biggio (53.4)
  15. Joe Gordon (51.4)
  16. Willie Randolph (50.8)
  17. Robinson Cano (49.4)
  18. Jeff Kent (45.4)
  19. Billy Herman (45.1)
  20. Bobby Doerr (43.8)

Now let’s see how that list changes if we order the players by annual WAR average:

  1. Jackie Robinson (6.2)
  2. Rogers Hornsby (5.5)
  3. Joe Gordon (5.2)
  4. Robinson Cano (5.2)
  5. Chase Utley (5.1)
  6. Nap Lajoie (5.1)
  7. Eddie Collins (5.0)
  8. Dustin Pedroia (4.8)
  9. Joe Morgan (4.6)
  10. Rod Carew (4.3)
  11. Charlie Gehringer (4.2)
  12. Ryne Sandberg (4.2)
  13. Bobby Grich (4.2)
  14. Lou Whitaker (3.9)
  15. Roberto Alomar (3.9)
  16. Frankie Frisch (3.7)
  17. Bobby Doerr (3.7)
  18. Billy Herman (3.6)
  19. Willie Randolph (3.6)
  20. Tony Lazzeri (3.6)
  21. Craig Biggio (3.3)
  22. Jeff Kent (3.2)

This list is quite different. In effect, we have gone from asking “Who was responsible for the most career wins,” a volume stat, to “Who was responsible for the most wins per year,” a pure performance stat.

Suddenly, all the best second basemen in history didn’t play 100 years ago. Why is that? Well, the guys who played 100 years ago played longer and benefited more from WAR’s longevity bias. Hornsby played 23 seasons; Collins, 25; Lajoie, 21. The principal reason Morgan gets up there with these golden oldies is that he played 22. By failing to adequately mitigate this bias, JAWS reinforces  it.

You will note that Biggio and Kent both benefit from the longevity bias. Their per-year averages drop them from 14th to 20th and 18th to 21st, respectively. Whitaker is a better player than both by both measures.

A key factor here is defense, where both Biggio and Kent are rated below average. That’s a pretty important fact to know about a second baseman. Whitaker’s career defensive WAR, as mentioned, is 15.4. Kent’s is -0.7, Biggio’s -3.9.

The result of this analysis is that I changed my mind about both Biggio and Kent. I voted for Biggio in each of his first two years on the ballot. I wanted to vote for Kent last year, his first, but, like this year, ran into the problem of more worthy candidates than votes, so I didn’t. This year, I won’t be voting for either, which reduces my list of candidates to 13.

That doesn’t mean I won’t go back to voting for them at some point in the future when the ballot backlog clears. This year, having more worthy candidates than votes, I don’t have to reach the ultimate question of whether they belong. I only have to reach the conclusion that there are 10 candidates more deserving of my votes.

Anomaly No. 2: John Smoltz.

With a 7.6 deficit to the average JAWS score of starting pitchers in the Hall of Fame, Smoltz would be eliminated from consideration fairly quickly if you accept this result. So I go back to my litmus test for this tool as it applies to starting pitchers. Koufax has a deficit of 14.3 to the average JAWS score of starting pitchers in the Hall of Fame. This fact by itself is enough to marginalize JAWS for me in the examination of pitchers.

Koufax suffers, of course, from the longevity bias in both WAR and JAWS. He pitched only 12 seasons, and half of those were unremarkable. Kevin Appier, who pitched 16, is 13 places ahead of Koufax in the JAWS rankings. If I had been Jaffe while he was developing this system, I would have looked at this result alone and seen that I was failing to adequately mitigate WAR’s longevity bias. But quantitatively-oriented minds may simply see Koufax as an outlier who cannot be accounted for by any formula.

Even when you eliminate longevity as a factor, the result in Koufax’s case is puzzling. He has a career WAR of 49, which gives him an average for each of his 12 seasons of 4.1. Forget all the categories in the WAR formula, forget all the math, and just ask yourself: If you put Joe Blow average pitcher out there every fourth day in Koufax’s place, he’d win four fewer games over the course of a season? Really?

Well, yes, because Koufax doesn’t begin to assemble WAR of any kind until his career is half over. We forget that he was a quite forgettable young hurler in the late 1950s. Koufax’s peak, among the most brilliant in the history of the game, was only six years long. He was an all-star in all six after never having been one before. He won three Cy Young awards and finished third for a fourth. He received MVP votes after all six campaigns, winning it once and finishing second twice.

Now get this: Koufax loses 4.2 career WAR — or the equivalent of a full season — because he was a lousy hitter. Seriously. But even eliminating that silliness, he accumulated 88 percent of his career WAR over the final six seasons of a 12-year stay in the big leagues. If you calculate his average annual WAR over those six seasons, it’s 7.8, which is more like it, although I’m still not sure the formula is adequately valuing him. In 1963, when he wins both the Cy Young award and the MVP, he’s 25-5 with an ERA of 1.88. His WAR that year is 10.7. So if the Dodgers had sent Joe Blow average pitcher out there in his place that year, Joe Blow wins 14 games? Really? Jim O’Toole and Bob Friend tied for ninth winningest pitcher in the National League that year with 17. Fourteen wins is a pretty good year for Joe Blow.

Anyway, when you take the skepticism for JAWS you have learned from the Koufax case and apply it to Smoltz, you see quite quickly why the formula values him so much lower than it values his former teammates, Maddux and Glavine. Smoltz had a five-year stretch right in the middle of his career in which he started only five games. He missed all of the 2000 season following Tommy John surgery and returned in 2001 as a reliever, although he got all five of his starts between 2000 and 2004 that year. In 2002, he saved a league-leading 55 games and finished third in Cy Young voting and eighth in MVP voting. His WAR that year was 1.2. So presumably Joe Blow saves 54 of those games.

I tell you one thing I’m learning here: I’m grabbing Joe Blow in the Rule 5 draft at the first opportunity.

Because of his unusual double as one of the best starters and one of the best closers of his time, Smoltz really has only one comparable, and that’s Dennis Eckersley. Eck spent about half his career in each role. He was a pretty good starter, getting double-figure wins 10 times and winning 20 once, but not as good as Smoltz. He was a better closer by volume because he did it so much longer, but it’s hard to beat what Smoltz did in those three seasons on a qualitative basis. Eck, who is in the Hall as a closer, trails Smoltz in career WAR, 66.5-62.5. Smoltz is the only player in baseball history to amass at least 200 wins (213) and 150 saves (154). Eckersley fell three wins short of 200. I have not done exhaustive research on this, but so far as I can tell, Smoltz is the only player in history to have led his league in both wins (twice) and saves (once).

For the 11 seasons before his injury, Smoltz averaged 3.95 WAR per year, including 7.3 in 1996, when he won the Cy Young award and went 24-8. So presumably Joe Blow wins 17 that year; I totally want to sign this guy. Smoltz’s annual WAR as one of the best closers around are 1.2, 3.3 and 2.2. If you bump these up to the average he’d established prior to that, he jumps up the JAWS rankings to the Bob Feller, Roy Halladay neighborhood, and now he’s a serious candidate even by JAWS standards, although he still has a deficit to the average JAWS score of pitching Hall-of-Famers. When he returned to starting at the age of 38, he put up consecutive annual WARs of 4.9, 5.9 and 4.6. If you give him that annual average during his three years as a closer — 5.1 — he improves his standing further. At some point, you get lost in this hypothetical math, so I’m going to stop.

Bottom line: In my opinion, WAR and JAWS are inadequate to account for an anomaly like Smoltz. To my subjective eye, he was every bit Glavine’s equal, if not quite Maddux’s, and held his own very nicely in that select company for many years. He was a wonderful starter, a wonderful reliever and a superb postseason competitor (15-4, 2.67). I’m voting for him.

Anomaly No. 3: Larry Walker.

I admit I bring a clear bias to this part of the conversation, but I contend that my bias is largely a reaction to the bias of most baseball writers and analysts not located in the Rocky Mountains — meaning all but about a dozen of them — against performances put up at Coors Field. Everybody knows it is the park that produces the biggest offensive numbers most of the time, therefore those numbers have to be discounted.

Which is where all these advanced metrics come in, right? They do that. They adjust for ballparks. So that should take care of it. According to JAWS, Walker is the 10th best right fielder in baseball history. The nine ahead of him are all in the Hall of Fame. Three of the four immediately behind him are in the Hall. If he were inducted tomorrow, his JAWS score would be slightly above the average of his new peer group. And yet, he’s getting barely cursory consideration from Hall voters. Last year, he finished 19th of 36 candidates. One in 10 voters checked him off.

I’ll be honest: I don’t know how to compensate for the effects of Coors Field. I’m not that into the mathematics of baseball. So I accept the ballpark adjustments the sabermetricians make. I don’t know exactly what they are, but this is what they do, right? They take the above-average offensive output at Coors and they multiply performances there by some factor to normalize them. I assume this is what they do.

So once Walker’s traditional stats have been put through these various wringers to produce his WAR and JAWS numbers, why don’t they count? Why don’t most of the voters take him as seriously as his JAWS score suggests they should?

Having spent quite a bit of time in press boxes during my days as an active BBWAA member, I would say this is because of a sort of sneering, smirking prejudice against Coors Field and Colorado and anything done here — yes, I’m writing from elevation, and no, I can’t seem to throw a breaking ball — on the part of most baseball writers. A lot of them honestly don’t think, even after 22 years, that Denver is a suitable place for big league baseball. The game is too weird here, not real, not legit.

If you cheer for the Rockies, you may feel the same way. Based on that 22-year data set, there are statistical anomalies here that are really discouraging with respect to the home team’s chances of sustained success. Pitchers blow out here, both physically and mentally. No good pitcher in his right mind wants to pitch half his games here. The few pitchers who experience success here can’t sustain it. No pitcher has won 100 games in a Rockies uniform. No pitcher has won 80. So elevation is a big issue in terms of the team’s ability to compete.

But if the math can adjust for the outsized numbers put up here, why should a player of Walker’s quality be penalized for playing much of his career in a place the snobs of baseball don’t take seriously? Here’s something the discounters don’t pay enough attention to know: Rockies hitters traditionally struggle when they head out on the road after a homestand because they have to readjust to the sharper bite of breaking balls and sinkers again. So while it is true that their offensive output at home is exaggerated by the Coors Field effect, their offensive output on the road is also depressed by it; hence the dramatic home/road splits that critics attribute entirely to the magnification at home. The road numbers are the “real” numbers, they say. They consistently fail to acknowledge the established phenomenon of the destructive effect each and every time hitters return to sea level. Back in the day, Dante Bichette brought a “curve ball machine” on the road with him in an attempt to reacclimate. An inventor in Greeley, Colorado has proposed a pressurized batting cage in which the Rocks could work against pitching in normal (read: sea level) atmospheric conditions in preparation for each road trip. There is no adjustment for this phenomenon in the advanced metrics that I know of. Walker’s numbers at Dodger Stadium would be adjusted for the park in the same way as everybody else’s, even if he’s coming from a mile-high elevation and experiencing a phenomenon that only his teammates face.

Walker got MVP votes after eight seasons, winning the award in 1997. He batted .384 at Coors that year, but he also batted .346 on the road. He hit 29 of his 49 homers on the road. In “late and close” situations, baseball’s measurement for clutch, he hit .352. Heck, he batted .322 in 1994 playing his home games in Montreal’s ghastly Olympic Stadium. He was a marvelous outfielder with a terrific arm and one of the best baserunners of his generation. He played an all-out style that produced various injuries that hurt his volume numbers, but he still overcomes the longevity bias of both WAR and JAWS. He and Harry Heilmann played the fewest seasons (17) of anyone in JAWS’s top 10 right fielders. His JAWS score is better than that of Hall-of-Famers Paul Waner, Sam Crawford, Tony Gwynn, Dave Winfield and a bunch of old-timers you may not have heard of. Gwynn, you may recall, was elected to the Hall on his first try, with 97.6 percent of the vote.

According to the eye test, Walker was an obvious Hall-of-Famer when healthy, a five-tool player who simply played the game at a higher level than everybody else. But there was always that sneaking suspicion about the enhancing effect of Coors Field on his numbers, that those three batting championships, that incredible run of six seasons — .366, .363, .379, .309, .350, .338 — was some sort of mirage. I voted for him his first two years on the ballot based on my subjective view of his talent, but I left him off my ballot last year when there were so many deserving candidates.

I’m grateful for the sabermetrics here because I didn’t know how to adjust for the Coors Field factor and they do, or at least they think they do. The quantitative analysis says Walker was one of the best right fielders ever to play the game. And here’s the important point: The quant analysis says he is the best right fielder in history who is not in the Hall of Fame.

The privilege of watching him every day produced the same conclusion from the eye test. In this case, it seems to me, the quantitative analysis finds a great player and the qualitative analysis of most writers is buried in prejudice and ignorance about Colorado and Coors Field. With my subjective view reinforced by the quants, I will not only vote for Walker this year, I will reserve one of my votes for him for as long as he is on the ballot and I have the privilege of voting.

These are the most interesting and/or difficult issues I wrestled with this year. According to JAWS, neither Tom Gordon nor Lee Smith meets the standard of an average Hall of Fame relief pitcher, and subjectively, I agree. As we’ve seen in Smoltz’s case, advanced metrics have difficulty valuing closers. It’s ridiculous, in my opinion, to suggest that a top closer means only one or two wins a year over an average closer, but I sympathize with the difficulty of the task. It’s such a specialty, it’s a little like a kicker in football, and those guys don’t generally make the Hall of Fame, either.

Eliminating Biggio, Kent, Gordon and Smith gets me down to 11, and there I’m stuck. I would vote for all remaining 11 if I could. And who knows, if I had more votes I might vote for Biggio, Kent and Gary Sheffield, too. As I mentioned earlier, I never reached that point because I knew I didn’t have enough votes.

Because of the logjam when Maddux, Glavine and Thomas hit the ballot last year, I didn’t vote for either Curt Schilling or Mike Mussina, having voted for Schilling the year before. But after studying their numbers, both advanced and basic, my subjective view that they are qualified was confirmed, Schilling, frankly, more so than Mussina because of the unbelievable postseason record. But, in examining my eye test on these guys, I had to admit to a personal bias against Mussina because he left the Orioles as a mercenary and signed with the Yankees. I hate players who do that within a division because it makes the smaller market team look like part of a feeder system to the bigger market team and undermines the illusion of competitive balance.

Mussina went for the money and the improved chances of a championship on a team willing to field the best players money could buy. Do you believe in karma? The Yankees won a championship the year before he arrived, in 2000, and the year after he retired, in 2009. But not during the eight years he was there. As Mel Allen used to say, how ’bout that. Still, he gave the O’s 10 good years and they had losing records his last three seasons there. I still don’t forgive him for the treachery of signing with the Yankees. He could have gone anywhere else without seeming to throw over his girlfriend to go date the richest girl in school. I still haven’t forgiven Reggie Jackson, either, and he spent only one season with the Birds, biding his time until free agency. If anyone was ever meant to play in New York, it was Jackson, but I still have the SI cover with him in an Orioles uniform. He looked good. But I digress. I will vote for Mussina and his 270 wins and 3.68 ERA and JAWS score slightly above average for Hall of Fame pitchers.

Schilling, of course, was obtained by the Orioles in a great trade — from the Red Sox, who drafted him, along with Brady Anderson, for Mike Boddicker — and then shipped out in one of the worst trades of all time — with Steve Finley and Pete Harnisch, for Glenn Davis. This is not relevant to the discussion in any way; just thought I’d mention it. Davis played 185 games over the next three seasons and retired. Schilling, Finley and Harnisch all went on to have long, productive careers. JAWS says Schilling is better than the average pitcher now in the Hall, and I know he was better in the postseason, when he went 11-2 with a 2.23 ERA. In the World Series, he was 4-1, 2.06.

Randy Johnson and Pedro Martinez are no-brainers I won’t dwell on. Johnson is one of 24 pitchers in history to win at least 300 games (303). He led the league in strikeouts nine times, won five Cy Young awards and finished second for three more. He had records like 18-2 in 1995 and 20-4 in 1997 for the Mariners, 21-6 in 2001 and 24-5 in 2002 for the Diamondbacks. At 6-10, the Big Unit and his unwinding windup were unique.

Martinez won three Cy Young awards, finished second for two more and won five ERA titles with these numbers: 1.90 (Montreal, 1997), 2.07 (Boston, 1999), 1.74 (Boston, 2000), 2.26 (Boston, 2002) and 2.22 (Boston, 2003). That year in between titles in Boston? He sagged to 2.39. From 1997 to 2000, there was nobody better, not even Johnson.

I’m voting for Piazza and Bagwell on the same basis I voted for them last year — their numbers obviously qualify them and I don’t have enough evidence of cheating, despite the rumors that followed them, to disqualify them on that basis. Piazza is the fifth-best catcher in history, according to JAWS, better than Yogi Berra and Bill Dickey and Mickey Cochrane. The only catchers who score better are Johnny Bench, Gary Carter, Ivan Rodriguez and Carlton Fisk. Bagwell is the sixth-best first baseman of all time, JAWS says, better than Frank Thomas, elected on his first try last year, Eddie Murray and Willie McCovey, Hall-of-Famers all.

I’m voting for Raines and Trammell, as I have consistently, in declining hope that advanced metrics will make more voters wake up to how good these guys were. The persistent and determined undervaluation of Trammell and Whitaker remains a mystery to me. Maybe it’s a Detroit thing. Been a while since Al Kaline and Sparky Anderson. Trammell has a better JAWS score than Derek Freaking Jeter, who will be serenaded into the Hall on a bed of rose petals at the first opportunity. Trammell was not as good an offensive player, although a lifetime average of .285 with 185 homers ain’t too shabby for a shortstop, but he was a much better defensive player at arguably the most difficult defensive position in the game. Trammell’s career defensive WAR is +22. Jeter’s is -9.7.

Raines’s JAWS score justifies his election, too. He’s the eighth-best left fielder in history, according to JAWS, and the only ones above him not in the Hall are Bonds and Pete Rose. He was also the best base-stealer I ever saw. He finished fifth all time with 808, but the amazing part was his percentage. He was successful 85.7 percent of the time. Only Carlos Beltran, who tried less than half as often, has a better rate. As with Trammell, I am encouraged that advanced metrics bolster Raines’s case, but doubtful that my colleagues in the BBWAA will see the light in time. The new rule allowing candidates to remain on the ballot only 10 years instead of 15 will hurt Raines more than anyone else. He has three years left instead of eight, in the midst of the current wave of superstars becoming eligible. Trammell was grandfathered into the 15-year stay, but has only two of them remaining. The both deserve inclusion, as the advanced metrics confirm, but they will have to rely on the kindness of the veterans committee.

So that leaves Edgar Martinez as the odd man out. I wish I could vote for him. I voted for him last year. He was a lifetime .312 hitter, a two-time batting champion, one of those guys who was born to hit a baseball. Unfortunately, there is also a major weakness to his game, which is defense. He was a sub-par defender when he played third for the Mariners and then he played no defense at all in 68 percent of his games, which is not an insult but a fact. He became a designated hitter. That didn’t keep me from voting for him last year, and it won’t in the future when the logjam clears, but it’s the distinguishing factor among worthies that takes him off my ballot this year.

In the end, even with the criticisms I’ve expressed about JAWS, I end up with only one change from the JAWS top 10 of this year’s class of eligibles after my exclusions for obvious cheating. I remove Edgar Martinez and add John Smoltz.

My 2015 ballot:

1B Jeff Bagwell

P   Randy Johnson

P   Pedro Martinez

P   Mike Mussina

C   Mike Piazza

LF Tim Raines

P   Curt Schilling

P   John Smoltz

SS Alan Trammell

RF Larry Walker

Happy new year.


2015 Baseball Hall of Fame ballot, Part 1

This makes two years in a row, consecutively, that the U.S. Postal Service has delivered my Baseball Hall of Fame ballot to the correct box and not inserted it into a 300-page holiday catalog. That’s a good sign for 2015 and appropriate celebratory measures were taken.

Any year now I’m confident the Hall will transition to electronic voting, but this is baseball, so one step at a time. This year’s innovation: For the first time, they will confirm receipt of ballots.

Before the ballot arrived, I knew I would want to vote for more candidates than the maximum of 10 the Hall permits. My first scan of the ballot, with no benefit of statistical analysis or divine wisdom, produced 14 players that would get my vote. So this year’s consideration was less about the usual issue — Hall worthiness — and more about distinguishing among worthies to cull the list to 10.

Cutting is something newspaper people know about, and we don’t like it. It simplifies, it misleads, it sometimes make you look like an idiot. So I decided to protect myself. For the first time, I used sabermetrician Jay Jaffe’s JAWS system (the humble acronym for Jaffe WAR Score) as my initial screen to get down to the difficult qualitative choices.

In previous years, I looked at lots of numbers to confirm or dispute my impressions, but mostly I relied on the eye test, my totally subjective judgment of whether a player looked like a Hall of Famer. This year, it turned out there wasn’t that much difference between the two, and there were areas of agreement that I found encouraging, not so much for the reliability of my vision as for the chances of certain players I’ve been supporting to no avail up until now.

This happy coincidence is not always the case. For example, when I did my JAWS work last year, I found Jack Morris down so low among the starting pitchers I had to add memory to my laptop to find him. His deficit to the JAWS average of starting pitchers already in the Hall of Fame was massive. So I looked at some of the names above his and asked myself if I would rather have them start a big game. Guys like Mel Stottlemyre, Bartolo Colon, Camilo Pascual, Brad Radke, Jamie Moyer, Carlos Zambrano, Jimmy Key, and so on. The answer in all those cases was no.

The calculation of wins above replacement (WAR), whichever version you subscribe to, is too complex to permit easy diagnostics of the anomalies, but in Morris’s case, I imagined earned-run average was a major culprit. He ranks No. 394 at 3.90, putting him between Joe Nuxhall and Ben McDonald.

On the other hand, if you sort the list by wins, he bounds up to No. 44 all-time with 256. His neighbors here are Red Faber and Carl Hubbell, Hall of Famers both. Now, wins have become a deeply unfashionable stat for pitchers, owing to the many extraneous factors that affect them, but this is also slightly ironic because wins are what WAR allegedly measures.

Anyway, I’m not trying to re-litigate the Morris matter. His eligibility expired last year after he received 351 votes out of 571 ballots cast, or 61.5 percent. I think he has a pretty good shot to make it eventually through some iteration of the veterans committee because he was a horse and a hell of a competitor and for a while there, the dominant pitcher of the 1980s. But again, I digress. He’s a good example of the difference between quantitative and qualitative analytical methods in certain cases.

Let’s back up and explain the basics of the quantitative screen I used this year to get down to the serious calls. Here’s a link to the Baseball Reference explanation of JAWS; here’s one to the BR explanation of its version of WAR.

If you’re reading a 4,000-word post on this year’s Baseball Hall of Fame ballot marked “Part 1,” you probably know that WAR is a complicated formula — so complicated that it’s effectively impenetrable for non-math geeks — that attempts to boil down to a single number the overall value of a baseball player. It speaks to the combination of two human needs — certainty and simplicity. So long as the evaluation of players is based on many different categories, as it was for generations, it is messy, and the emphasis any one judge puts on each of those various categories can and often does change the outcome of the inquiry. That’s neither certain nor simple.

WAR’s many calculations and adjustments introduce subjectivity themselves — BR recently changed its definition of an “average” replacement player — or contribute subjectivity in the weight they are accorded in the overall formula. Since most fans like the sport a lot more than they like equations, especially complicated equations, they stop paying attention to the details pretty quickly. Like the passer rating in football, it’s a nice, simple number that many fans use and few can explain in detail. We have two main versions of WAR at present, which make slightly different subjective judgments along the way, and produce slightly different results.

At its most basic, WAR is an adding machine. Player A was X amount better than the average guy at his position, plug that into a formula and you get 2.4 extra wins for his team that year. Then you add up all those little premiums over the course of his career. Even though its modifications and adjustments are considered ameliorating factors, it rewards longevity in a big way — the more years of little numbers like that, the bigger the career total is going to be.

So, for example, Sandy Koufax, one of the best starting pitchers of all time according to pretty much anyone paying attention to the game when he was pitching, ranks No. 85 in the JAWS system, with a considerable deficit to the average JAWS score of starting pitchers already in the Hall of Fame. He’s down around a bunch of guys who aren’t in the Hall and aren’t likely to get there — Kevin Appier, Chuck Finley, Orel Hershiser, Tommy John, Frank Tanana, Wilbur Wood, et al. The JAWS system attempts to compensate for the longevity bias of WAR by averaging a player’s career WAR with his “peak” WAR — the seven highest annual scores of his career.  This is supposed to cover for players like Koufax, who were brilliant, but whose careers were cut short.

But, clearly, it doesn’t. Koufax’s peak WAR is higher than any of the guys I mentioned in the previous paragraph, but not by enough to make much difference in the overall calculation. Luckily, there was no WAR when Koufax became eligible for the first time in 1972 and he was elected on the first ballot with 87 percent of the vote, leading a class that also included Yogi Berra and Early Wynn.

I bring this up because it also contributes to this year’s biggest anomaly when using the JAWS system, which is John Smoltz. The former Braves hurler is on the ballot for the first time, following by one year former teammates Greg Maddux and Tom Glavine, both of whom were elected on the first try. The voters and JAWS agreed on Maddux. Not so much on Glavine. Maddux ranks 10th all time on the JAWS meter, ahead of immortals like Bob Gibson and Warren Spahn, and pending immortals like Pedro Martinez. Glavine ranks 30th, behind all those guys, and behind a couple more on this year’s ballot — Curt Schilling and Mike Mussina. JAWS counts Maddux as way above the average Hall of Famer at his position, Glavine as barely above average.

Smoltz ranks 58th, well below average for the Hall of Fame, down around guys like Jim Bunning and David Cone. This is mostly because he quit starting for a while to help his team. From 2002-04, Smoltz registered 144 saves with ERAs of 3.25, 1.12 and 2.76. He received MVP votes four times in his career, and three of them were after those three remarkable years as a closer. The Braves made the playoffs in all three seasons. He’s credited with 6.7 wins above replacement for those three seasons combined. He got more than that (7.3) in 1996 alone, his Cy Young year as a starter.

So Smoltz fares poorly in the JAWS Hall of Fame analysis, in part because he’s listed with the starters and no allowance is made for all the starts he gave up to become one of the best closers in the game. For those of us not yet willing to sacrifice our own observations entirely to the gods of quantitative analysis, Smoltz requires an eye-test adjustment this year, which we will get to in Part 2. For the purpose of culling the field of 34 down to a more manageable number, JAWS was quite useful in quantifying the eye test results.

The no-hopers

1. 3B Aaron Boone

He’ll always have 2003, a moment seared into the memories of Red Sox fans who would not slay the beast for another year. Boone’s walkoff home run leading off the bottom of the 11th in Game 7 of the ALCS made him a Yankee hero, even though he played only 54 of his 1152 career games in Yankee pinstripes. His best days were in Cincinnati. He is ranked the 150th-best third baseman in big league history by his JAWS average (14.3). His deficit from the JAWS average for Hall of Famers at his position (40.7) is the greatest in this year’s class. The top comparable to him on Baseball Reference is Scott Brosius.

Every year, there are fans who squawk, “Why is he on the ballot?” I have known members of the screening committee and I agree with their bias toward inclusion. If you meet the basic requirements — a career of at least 10 seasons in the designated time period — I believe you ought to get your minimum one appearance on the ballot. A 10-year major league career is a big deal in any context other than this one. Appearing on the ballot once seems to me an appropriate reward. Boone won’t be back, and neither will the rest of the guys in this category.

2. 1B Tony Clark

Clark had a nice 15-year career with Tigers, Diamondbacks, Mets, Red Sox and Padres. He hit 251 homers with a lifetime .262 batting average and a career OPS of .824. The JAWS average for first basemen now in the Hall is 54.2. Clark’s is 14.2. That deficit (40.0) barely lost out to Boone’s for the title of No-hoper of the Year. His top comparable is Adam LaRoche.

3. SS Rich Aurilia

Aurilia played 15 seasons, mostly for the Giants, finishing with a long goodbye tour through Cincinnati, San Diego and Seattle. He was mostly a shortstop, but he played triple-figure games at each of the other infield positions as well. He had good power for an infielder, hitting double-figure home runs eight times, including an eyebrow-raising 37 in 2001, the year Barry Bonds hit 73 and a number of players looked like Popeye after the spinach. His JAWS deficit to Hall of Famers at his position is 37.3. No. 1 comparable: Brandon Phillips.

4. RF Jermaine Dye

Dye had some serious pop, hitting 44 homers for the White Sox, batting over .300 for the only time in his career and finishing fifth in MVP voting in 2006. He spent 14 years in the big leagues almost equally divided among the Royals, White Sox and A’s, with a nightcap in Atlanta. His deficit to the JAWS average of right fielders in the Hall is 36.4.

5. SP Jason Schmidt

Schmidt was an average starter for most of the first decade of his career, never pitching to an ERA under 4.00 until arriving in San Francisco in the latter part of the 2001 season. Quite suddenly, at the age of 30, he became an elite pitcher for the Giants, going 17-5 with a National League-leading ERA of 2.34. He finished second in Cy Young voting that year and fourth the following year, when he went 18-7 with a 3.20 ERA. As a result of this renaissance in San Francisco, the Dodgers gave him a big free-agent deal in 2007 — three years, $47 million. Alas, Schmidt blew out his shoulder and pitched just 43 mostly ineffective innings over two seasons for L.A. before calling it quits. He comes in at No. 284 on the starting pitchers’ JAWS list. His deficit to existing Hall of Famers is 33.4.

6. LF Cliff Floyd

Floyd was the 14th overall pick of the 1991 draft, and while he had a 17-year stay in the majors, he never became the star that early comparisons to Willie McCovey seemed to promise. He drove in more than 100 runs only once and hit as many as 30 homers twice. He was bedeviled by injuries and ended up playing for seven teams. Still, he finished with a career OPS of .840. His top comparable on Baseball Reference is Aubrey Huff. His JAWS Hall of Fame deficit is 27.9.

7. CF Darin Erstad

Erstad was a marvelous athlete whom the Angels made the first pick of the 1995 draft out of Nebraska. He was the punter for Tom Osborne’s ’94 Cornhuskers’ national championship team. Erstad arrived in the majors a year later and finished sixth in rookie-of-the-year voting despite playing only 57 games. He was an all-star at 24 and again at 26, receiving MVP votes both years. He was a five-tool player on the brink of stardom. After batting an astonishing .355 with a league-leading 240 hits in 2000, he never hit .300 again. After putting up double-figure home run totals in five of his first seven seasons, he never hit more than seven after the age of 28. His hustling style and proclivity for leaving his feet produced chronic issues with his shoulders, hamstrings and ankles. For most of the latter part of his career, he was battling one injury or another. His JAWS deficit to center fielders in the Hall is 26.7. On the bright side, he went back to Nebraska as a volunteer assistant for the baseball team in 2011 and Osborne, then the athletic director, named him head coach a year later. His top comparable at 23 was Jeff Bagwell. Ten years later, it was Mark Kotsay.

8. RP Eddie Guardado

Guardado would be a long way from Hall consideration anyway, but the fact that he’s a relief pitcher makes it impossible. He was a decent piece out of the pen for the Twins for years who suddenly became a closer, and a good one, for a four-year stretch from 2002-05. He was an all-star in ’02 and ’03, putting up a league-leading 45 saves in ’02 and following it with 41 the next year. The result was a nice free-agent deal with the Mariners and two more good years before he ran out of fuel. He’s No. 53 in career saves with 187. His JAWS deficit is 22.1. In the history of baseball, five relief pitchers have been inducted, so if you’re Eddie Guardado, you have no shot.

9. RP Troy Percival

Percival was a stud closer for a long time, so you’d think he’d have a chance, but keep in mind he’s way behind Lee Smith in the JAWS rankings and Smith got only 30 percent of the vote in his 12th season of eligibility last year. Percival is in the top 10 in career saves — No. 9 with 358. But again, the standard is the five incumbents — Dennis Eckersley, Hoyt Wilhelm, Rich Gossage, Bruce Sutter and Rollie Fingers. Mariano Rivera will make it six when he’s eligible. Percival ranks behind them all, and behind Smith, too. In fact, his JAWS average is worse than those of Steve Reed, Dick Drago and Dave Giusti. It’s just not happening. His deficit to the Hall average at his position is 18.1

My kingdom for 5 percent

1. RF Sammy Sosa

You wouldn’t think that one of only two players in major league history to hit more than 60 homers in a season twice, one of only eight to hit more than 600 in a career, would be in any danger of falling off the ballot in just his third season of eligibility, but Sosa treated his entire career like an SNL skit. Like Bonds, he used performance-enhancing drugs to such excess that he became a massive, hulking, comic book parody of his former self, then pretended he didn’t understand English when hauled before Congress to explain this amazing transformation. He got 12.5 percent of the vote in his first year and 7.2 percent last year. Half as great a loss of support this year would send him on his way. Even if he’d done it honestly, his JAWS average of 51 wouldn’t justify his election. The average of the right fielders already in the Hall is 58.1. Even through the eyes of a Cubs fan, Sammy doesn’t pass the eye test.

2. 1B Don Mattingly

It doesn’t seem fair that you could drop off the ballot for lack of support the day you’re dropping off anyway by virtue of having been there 15 years without getting elected, but that is a possibility for Donnie Baseball. His election return of 8.2 percent of the vote last year was his lowest total in 14 years of eligibility. He hasn’t been above 20 percent since his second year on the ballot. For a stretch in the 1980s, he was as good as it got, but back trouble limited him to just four years of WAR of 5 or more, and that’s nowhere near enough. He also faces competition from a bunch of other first basemen who aren’t going to make it but have better numbers than he does. On the JAWS list, he ranks behind Norm Cash, Gil Hodges, John Olerud and Will Clark, none of whom are in the Hall. He also ranks behind Carlos Delgado and Fred McGriff, who are on the current ballot and also have little or no shot. His JAWS deficit to the Hall average first sacker is 15.3.

3. 1B Carlos Delgado

Delgado isn’t a close call according to JAWS — he’s 14.8 average WAR behind the existing Hall standard at first — but he was a very productive power hitter for a long time, so you wouldn’t think he’d be in any danger of missing the minimum to remain on the ballot. He has a career OPS of .940 and drove in at least 87 runs in 13 consecutive seasons. Nevertheless, it’s a very strong ballot and Delgado’s JAWS average ranks below McGriff’s. McGriff fell to 11.7 percent of the vote last year, losing half the support he got just two years before. If there’s an argument to be made that Delgado is one of the top 10 players on this year’s ballot, I’m not sure what it is.

4. RF Brian Giles

I’m also not sure why Giles does better on the JAWS tote board than Delgado. I’d take Delgado every time, although Giles was certainly a nice offensive player. Delgado had way more homers and RBI, a slightly better OPS+ and trailed in batting average by 10 points. It is true he also struck out a lot more than Giles, but that’s the price of that power (1512 RBI to 1078; 473 homers to 278). Both were below-average defenders, but Delgado somewhat further below average. Giles is a classic very good player who is nevertheless significantly short of the Hall standard by both quantitative and qualitative analysis. Like Delgado, you would expect him to exceed the minimum necessary to stay on for another year if the ballot weren’t so loaded. And he still should. His JAWS deficit among right fielders already in the Hall is 14.0.

Obvious cheaters

1. 1B Mark McGwire

The only thing that saves Sosa from appearing here, where he belongs, is the fact that he might fall off the ballot altogether, earning him a spot in the previous category. McGwire, the only player other than Sosa to hit more than 60 homers in a season twice, comes dangerously close to joining him there. But Mac managed 11 percent of the vote last year, so I’m guessing he survives again, although his number has come down from 23.7 percent in 2010. I realize there is a fair amount of sentiment among sabermetricians that there is no such thing as right and wrong, or we’re not worthy to assess it, or only numbers matter, or something along those lines. If you’re interested in why I don’t agree with this, I went into some depth about it in my post on last year’s ballot. McGwire, by the way, has a small deficit — 2.3 — to the average JAWS score of first basemen now in the Hall.

2. SP Roger Clemens

Clemens’ numbers are ridiculous, which, in his case, is not just an expression. At the age of 42, suddenly he was as good as he had ever been. In fact, he won the ERA title with the best mark of his career (1.87). In court, personal trainer Brian McNamee laid out his doping program in gruesome detail. Clemens’ able defense team got him off perjury charges, but that’s not the same as being innocent. According to JAWS, he’s the third-best pitcher of all time, behind Walter Johnson and Cy Young. His surplus to the average starter in the Hall is 41.5, which would be the highest on the ballot if it weren’t for . . .

3. LF Barry Bonds

Clemens redux. Thanks to the journalists Mark Fainaru-Wada and Lance Williams, and their book Game of Shadows, we know pretty much exactly what Bonds took and why. Like Clemens, he had a chronology-defying, chemistry-enabled, Greek God-like renaissance very late in his career, winning four consecutive MVP awards at the ages of 36, 37, 38 and 39. We had never seen anything like it before and, assuming Rob Manfred is slightly more on the ball than Bud Selig — a low bar — we will never see anything like it again. Like Clemens, as many have pointed out, he was probably a Hall of Famer before he started cheating, but those sportsmanship and integrity criteria established by the Hall don’t have the luxury of time travel. His surplus to the average JAWS score of Hall of Fame left fielders is 64.3, a number almost as big as his body in those days.

Just wide

1. Nomar Garciaparra

Nomar was a stud shortstop in a very impressive generation. Imagine a three-year stretch in which rookies named Alex Rodriguez, Derek Jeter and Nomar Garciaparra enter the league playing the same position. The Red Sox made Nomar a first-round pick, the 12th overall, in 1994. Three years later, he was rookie of the year, an all-star and eighth in MVP voting. At the age of 23, he hit .306 with 30 home runs and 98 RBI. The next year was even better — .323, 35, 122 — and he finished second in the MVP voting to Juan Gonzalez. He won the batting title each of the next two years, at .357 and .372. If this was not the runup to a Hall of Fame career, it was hard to imagine why not. But it probably wasn’t. Injuries would turn Nomar old before his time and his 30s were nothing like his 20s. The average of his career WAR and peak WAR — 43.6 — is 11.1 below the average of shortstops already in the Hall. You can certainly make an argument for him. His JAWS rating puts him between Luis Aparicio and Joe Tinker, both of whom are in. The improvement in the offensive output at the position since then has changed the calculus.

2. 1B Fred McGriff

You look at those career numbers and it’s hard to imagine he doesn’t belong. Seven homers shy of 500. Twenty-one consecutive seasons of double-digit homers. An OPS just short of .900. MVP votes following eight different seasons. And yet, when you add them all up, he’s a full 10 wins over replacement short of the average for first basemen already in the Hall. In fact, he ranks behind Keith Hernandez, John Olerud and Will Clark, none of whom are enshrined. Of 20 first basemen in Cooperstown, McGriff’s JAWS average is better than four, worse than 16. He’d been getting about 20 percent of the vote in his first four years on the ballot, then fell to 11.7 percent last year as the ballot strengthened. Probably won’t do a lot better this year.

3. RF Gary Sheffield

Sheff had such a long, productive career, one is tempted to think of him as the Don Sutton of hitters, a guy who eventually makes it on sheer volume. It could happen, but advanced metrics don’t help him. He accumulated surprisingly modest wins above replacement value during his career and his JAWS average of 49 falls 9.1 short of the average of right fielders already inducted. Sheff’s average is worse than those of Bobby (not Barry) Bonds, Reggie Smith and Dwight Evans, none of whom is in. With more than 500 home runs, a career batting average over .290 and an OPS over .900, Sheff passes the eye test, at least for me. He might begin to build support after the current wave of stars passes. For now, he just has to hang around.

Coming in Part 2: The final fifteen