Monday, December 01, 2008

Five Myths about Bert Blyelven's HOF Candidacy

1) Blylevens' record can be attributed to playing on bad teams. In fact, the teams Blyleven pitched for were above average, winning more games than they lost.

2) Blyleven's teams didn't score many runs. In fact, Blyleven's teams scored more runs than the average teams during those same seasons.

3) Blyleven had the "bad luck" of pitching when his team didn't score.

Some people have gone through the games and shown that Blyleven got below average run support when he pitched, despite the fact his teammates, on average, scored more runs than other teams. They attribute this to "bad luck".

If it wasn't for bad luck, I wouldn't have no luck at all.

Even if you attribute his teammate's offensive failure to luck, its certainly just as likely that same luck improved Blyleven's own results right along with his opponents. What are the likely explanations that would explain both pitchers doing better than usual?
  • A) As the staff ace, Blyleven faced other team's aces. But Blyleven wasn't always the staff ace. On the Twins, for instance, Frank Viola was the staff ace. Blyleven was the number two starter.
  • B) Blyleven pitched more often in stadiums friendly to pitchers than his teams played in on average.  Of course, that would mean Blyleven's own statistics were improved by the same factor that drove down his run support.
  • C) He pitched in games where umpires had large strike zones. Again, Blyleven's own statistics were improved by the same factor that drove down his run support.
  • D) His teammates just happened to score fewer runs when Blyleven pitched. Statistically, given the large number of games over many seasons, that is an unlikely explanation.
4) Blyleven's lack of run support could only have been luck.

The most likely explanation for Blyleven's lack of run support is that umpires widened their strike zone when Blyleven pitched and that benefited both pitchers. Given Blyleven's outstanding curve, that is not unlikely. Umpires can be fooled as easily as hitters. If they are calling Blyleven's curve off the plate a strike, they are likely to start doing the same for the other pitchers' pitches. Is that speculation? Yes. But so are all the possible explanations based on luck.

5) Blyleven's low results in the Cy Young award voting should not be a factor.

This is the Hall of Fame. Is there some reason why voters whose peers never recognized Blyleven as one of the premiere pitchers during his career, should now vote him in to the Hall of Fame? Moreover, it wasn't just sports writers who didn't recognize Blyleven. The managers of the allstar teams rarely recognized him as a premiere pitcher either.

Blyleven had a very long and successful career that put him among the leaders in a number of categories.  But there are many people that attribute his presence in the leadership more to the length of his career, than to any outstanding success he had.  When proponents of Blyleven's candidacy are reduced to siting  "win shares" and "ERA+", as the StarTribune did in today's paper,  you know they have been reduced to grasping at straws. 

I am an agnostic on Blyleven as a Hall of Famer. It would certainly not be an outrage if he was elected, but its not an outrage that he hasn't been. He is, at best, a borderline candidate.

15 comments:

Anonymous said...

Yeah, it's "grasping at straws" to look at how Blyleven's ERA compared to the league average throughout his career. That new-fangled "ERA" stat really is a strange one, isn't it?

Blyleven finished in the top 5 in ERA 7 times. He led the league in shutouts 3 times and finished 2nd 3 more times. Are shutouts a stat for "pseudo-sophisticated" fans?

The bottom line is, Blyleven has 287 wins, had one of the best curveballs ever, was one of the best strikeout pitchers ever, had excellent ERA's, and had incredible longevity- which is a plus, not a minus.

TT said...

For those unfamiliar with it, ERA+ is not a comparison of a pitcher's ERA to the league average. It is the result of a variety of statistical calculations.

Nick N. said...

Your claim that Blyleven's proponents are being "reduced" to using statistics like win shares and ERA+ when literally your ENTIRE argument is based around his win total is really quite laughable. Why are you so stuck on that number?

Regardless of how what reasoning you think is behind what you view as an unsatisfactory win total (by the way, if he had 13 more wins would you support his candidacy? Why is this arbitrary 300-win mark so important?), why are you so fine with ignoring all of his other undeniably Hall-worthy numbers? He was one of the league's all-time best strikeout pitchers, he frequently was among the league leaders in ERA, he threw242 complete games and 60 shutouts in his career. All that, apparently, is irrelevant to you because his total of wins -- the statistic which he has the least control over out of all that have been mentioned here -- does not satisfy your personal requirement.

I wouldn't be accusing others of grasping for straws.

JW said...

ERA+ is not the result of a "variety" of inputs, it adjusts for park factors and otherwise compares a pitcher's ERA to the league average. It's a shorthand stat, much like ERA itself, which overvalues pitchers who allow a bunch of unearned runs. But, of course, TT opposes the more complicated stats that eliminate those errors, conveniently arriving at a position where pure subjectivity rules and his own bizarre opinions are accorded equal weight with those of rational fans.

TT said...

our ENTIRE argument is based around his win total

The term "win" doesn't appear anywhere in MY argument. But, yes, most of the myths have developed to explain Blyleven's failure to win more games.

TT opposes the more complicated stats that eliminate those errors

TT opposes stats that fail to measure what they claim to measure. They are often used by people who don't understand them and don't know how they are calculated as shorthand to "prove" their pre-conceived ideas are true. Just as they are here.

There is little doubt that parks have effects on players' results. There is also little doubt that those effects vary from player to player and year to year. There are a variety of systems for calculating "park effects" and none of them are appropriately applied to any individual player's statistics. Doing so is far more likely to distort results than clarify their meaning or "eliminate errors".

JW said...

There is little doubt that parks have effects on players' results. There is also little doubt that those effects vary from player to player and year to year. There are a variety of systems for calculating "park effects" and none of them are appropriately applied to any individual player's statistics. Doing so is far more likely to distort results than clarify their meaning or "eliminate errors".

OK, fine, let's say park factors are too difficult to account for. It doesn't matter- Blyleven's numbers are Hall-worthy without any adjustment whatsoever. Top 10 in ERA 10 times. Top 5 in strikeouts 12 times. Top 5 in shutouts 8 times, leader in 3 of them. Argue against that, not straw men.

Your "analysis" boils down to the argument that, if Blyleven really was a solid Hall of Famer, he would be in the Hall already because baseball observers would have realized he belonged.

But this argument is specious for two reasons. First, Hall voters have erred in the past (both being over-inclusive and under-inclusive). Second, voters change their minds over time. In fact, the voting system provides for exactly that scenario- election of a candidate who is underrated early in his candidacy.

Heck, the system addresses that problem twice over: first, candidates over a certain threshold remain on the ballot. This is totally illogical unless you allow the possibility other voters will change their minds (or new voters will be favorable). Second, the Veterans Committee exists in case the writers overlook a deserving candidate.

Blyleven is a classic case of a player underrated during his career. Maybe it's because he didn't play in big markets for the most part. Maybe it's because people think 300 is a magic number. Whatever the reason, they're wrong. Of course, you avoid taking a real position, since "traditional baseball values" apparently involve critiquing others without offering your own opinions.

TT said...

John -

You seem to be confusing how Blyleven was evaluated during his career with how he should be evaluated for the HOF.

My argument is that it is perfectly appropriate for today's HOF voters to look at how he was evaluated while he was playing the game. The contemporary people in the game (managers choosing the allstar pitching staff) and sports writers (Cy Young voters) did not put him high on their list of stars compared to his contemporaries.

In 22 seasons, he was named to the allstar team twice. He got even a single vote for the Cy Young award only 4 times, finishing no higher than 3rd. Even his hometown sports writers apparently didn't vote for him.

JW said...

You seem to be confusing how Blyleven was evaluated during his career with how he should be evaluated for the HOF.

My argument is that it is perfectly appropriate for today's HOF voters to look at how he was evaluated while he was playing the game.


Well sure, they can consider it. But I think you are confusing how Hall voting is intended to work, with how you think it should work.

If a player should be judged on the basis of contemporaneous accounts (relative to his career), as you suggest, the voting procedures should be much different. A player should either get in early or not at all, because the number of All-Star games or Cy Young awards isn't going to change.

But nonetheless, the Hall voting system explicitly allows a candidate whose worthiness is a minority view to remain on the ballot (so long as the minority is a decent size). This is inexplicable under your theory because fewer and fewer voters were around for the bulk of a player's career, especially if it was a long one.

In other words, the Hall voting system (including the Veterans Committee) permits voters to functionally overrule the contemporary observers, who generally did not confer many awards on Blyleven.

Now, maybe you think they should choose not to do this, but that's how the system is set up and how it often works, with many Hall members having to wait some time for induction. Your opinion is not invalid per se, as it's a subjective issue, but the people that made or oversee the Hall rules for induction must disagree with you.

Nick N. said...

TT opposes stats that fail to measure what they claim to measure.

If that is the case, then why are you so caught up on wins? That statistic really doesn't measure how many games a pitcher was able to "win," it measures how many times he left a game when his team was leading and his team was able to hold onto that lead. A lot can happen that is out of a pitcher's control to determine whether or not he receives a win in any particular game.

Even if the term "win" doesn't appear in your argument, it seems pretty clear that his win total is sole basis of your lack of support for his candidacy. As best I can tell, the only points you've brought up in his opposition are his win total and his Cy Young and All-Star appearances (which I think we can all agree are both tied directly to his win totals).

If there's a legitimate argument to be made against Blyleven on the basis of anything other than his win/loss record, I'd love to hear. I just don't think that single statistic is the be-all end-all when it comes to his worthiness for the Hall of Fame, and thus I find all these "myths" rather irrelevant.

Anonymous said...

"A player should either get in early or not at all, because the number of All-Star games or Cy Young awards isn't going to change."

Unlike his number of strikeouts or ERA which do change?

TT said...

it seems pretty clear that his win total is sole basis of your lack of support for his candidacy.

I think that is your belief about the "sole basis" for anyone's lack of support for his candidacy. Just as you believe that is the only reason he didn't make the allstar team or get Cy Young votes.

The only reason I mentioned was the belief that his overall numbers are more the result of longevity than outstanding performance.

For instance, for 10 years he was one of the top 10 pitchers in a league. That is a lot less impressive when you acknowledge that he pitched 22 seasons. That means in 12 seasons there were at least 10 pitchers better than him in just the one league he pitched in.

Almost half full? Or more than half empty? That is a judgment that has absolutely nothing to do with how many wins he got.

Nick N. said...

The only reason I mentioned was the belief that his overall numbers are more the result of longevity than outstanding performance.

Looking at contextualized rate stats helps solve this problem. Unfortunately, you refuse to concede that any metric developed within the last half-century has any merit.

Also, since when is longevity a bad thing?

JW said...

Unlike his number of strikeouts or ERA which do change?

Cy Youngs and All-Star selections are subjective. Strikeouts and ERA are not. Subjective opinions can change, statistical records cannot.

The voters can support a candidate who was unfairly overlooked for subjective awards. The voters cannot change a players statistics if they are inadequate.

Of course, my point was that voters should be able to change their minds, so I don't know what your beef is anyway.


For instance, for 10 years he was one of the top 10 pitchers in a league. That is a lot less impressive when you acknowledge that he pitched 22 seasons. That means in 12 seasons there were at least 10 pitchers better than him in just the one league he pitched in.


Longevity is always a factor in Hall admissions. Even players with relatively short careers by Hall standards have to reach certain minimum levels.

Now, clearly a player must also have "peak" seasons in which he was an elite player. Pure longevity is not enough. Three dominant seasons, and nothing else, is not enough.

Blyleven had a large number of seasons where he was one of the best pitchers in the league. He satisfies the requirement a Hall candidate have impressive overall numbers and also seasons in which he was dominant.

Nick N. said...

One other thing I'd like to note about Blyleven's "longevity" giving him an unfair advantage...

Seasons played by Blyleven: 22

Seasons played by the five dudes ahead of him on all-time strikeout list (all in the HoF or locks to go there):

Steve Carlton: 24
Roger Clemens: 24
Randy Johnson: 21
Nolan Ryan: 27

Seasons played by the five dudes below him on all-time strikeout list (all in HoF or locks to go there):

Tom Seaver: 20
Don Sutton: 23
Gaylord Perry: 22
Walter Johnson: 21
Greg Maddux: 23

We can play the same game with shutouts. Blyleven ranks ninth on this list, each of the guys ahead of him is in the Hall of Fame and each of them played for 20-plus years (with the exceptions of Eddie Plank and Christy Mathewson, who both played for 17 years and both retired before 1920).

Each of the 13 guys behind Blylevel on the shutout list is also in the Hall. Nearly all of them racked up those big totals either by pitching in the early 1900s/late 1800s or by having very long careers.

This longevity argument is really, really weak.

TT said...

Nick -

I think you are demonstrating why people should be suspicious of "new" stats. Because the people who use them don't really understand the limits of their actual meaning. What does the number of seasons pitched have to do with the number of opportunities for strikeouts or shutouts? Its a very crude comparison.

Blyleven is 11th on the career list of games started. Each of those was an opportunity to pitch a shutout. His ranking of 6th is better than expected, but its not obviously outstanding beyond what you would expect for a pitcher with that many opportunities. On the negative side, he is 10th in career losses which is slightly more than you would expect. And has been pointed out, he is lower in the rankings for wins than he should be given the number of starts he has.

In terms of strikeouts, the measure of opportunities could be either batters faced or innings pitched (number of outs). And Blyleven is 14th in innings pitched and 13th in batters faced. On the negative side, he is 10th in earned runs allowed while he got those strikeouts. He is 8th in home runs allowed and 15th in hits allowed.

When you look at his career ERA, he is at 300. Now that isn't fair since ERA's vary a lot by era. But neither is pointing to strikeouts which have increased dramatically since expansion in 1960. He is also 126th in career WHIP (one of the few new-fangled stats that actually means what it says). And it would be getting used here if it didn't help make the case against Blyleven.Thus, the comment about grasping at straws.

The fact is that Blyleven is among the career leaders in opportunities - whether games started or batters faced. As a result, he is among the career leaders in a lot of categories, good and bad.



And here is a new one of those myths "it (wins) measures how many times he left a game when his team was leading and his team was able to hold onto that lead."

That may be true for pitchers today, but not in the era when Blyleven pitched. Blyleven has only about 40 fewer complete games than he has wins. He got a lot of losses as well and he left every one of those games with the winning run at least on base, if not already having scored.

The point of my article, was that, in their enthusiasm for Blyleven, some of his supporters were creating a whole mythology to explain his lack of wins and high number of losses. While continuing to deny that they have any meaning.

MLB Twins Updates