There can't possibly be anything left to say on this subject :)

let's please not have another instant replay of Jim Rice on this thread

All's I gotta say is Bert is one guy who should get a "Santo letter." I think he would be happy. It might even get mentioned on the Twins network unless Bert ##### it up so they have to do the ####### thing over.

We're going to have to do this whole ######' thread over because I just ###### it up.

Does he have the record for most no-decisions (in games started) in a season with 20 in 1979?

OK, what does the RA+ PythPat system say? It gives him a career record of 322-230. That's a little behind Warren Spahn (340-242) but a litle ahead of both Gaylord Perry (337-358) and Steve Carlton (328-252). And I've got a "big years" score in my system - it's just cumulative year-by-year FWP above 15 - in which he scores 53 to Perry's 52 and Carlton's 54 (and Spahn's 58.)

I've got his best year as an equivalent 24-12 in 1973, but there are other big years scattered all over his career: 20-13 in1976; 20-13 again in 1985; 17-9, 18-9, and 18-9 in 1977, 1984, and 1989.

So people argue that he was worse than his ERA (or in my case, his RA) for this, that, or the other reason. Fine - if he was the worst pitcher in the history if the game for being worse than his RA and IP, then he's still easily a HoMer.

Does he have the record for most no-decisions (in games started) in a season with 20 in 1979?

He does, yes. I believe the next most is Odalis Perez with 18 in 2004.

I would strongly recommend you guys elect Blyleven, not just because he deserves it, but because if you don't, Rich Lederer will haunt you like the Ghost of Christmas Past.

Here are the 50 pitchers with the least number of decisions while registering 30 or more starts (and no games in relief).

Just glancing at it, the list appears dominated by more recent pitchers, with only a few sticking out from bygone eras. A result of lower innings-per-start, no doubt.

I would strongly recommend you guys elect Blyleven

Blyleven is a shoo-in. There may be a few dissenters, but he's going to be in an elect-me spot on a majority of ballots, I expect.

As OCF's analysis shows, he is comparable to the 1970s stalwarts. That's an obvious HoMer.

case against Bert: he did win fewer games than 'expected', given his run support

but that merely means he isn't AS qualified as his most ardent supporters claim him to be; he's still a clear HoMer.

let's please not have another instant replay of Jim Rice on this thread

Is there a history of Blyleven flame wars at BTF? I figured he would fly into the HOM with very little dissent. Guys with 5000 IP and 118 ERA+ are hard to find.

case against Bert: he did win fewer games than 'expected', given his run support. but that merely means he isn't AS qualified as his most ardent supporters claim him to be; he's still a clear HoMer.

I wouldn't mind discussing some of this. It is indeed an interesting curiosity how he managed to perennially underperform his run support in the 70s. Bad luck? An unusual mixture of shutouts and shellings in that skews his runs-per-outing distribution? I haven't seen a thorough analysis of this since his career became fully documented by retrosheet. If that is indeed so, what anecdotally caused that odd mixture of shutouts and shellings, some days the curve had more snap than others? Did he have personality issues? I've heard murmurs that the playful immaturity that made him a loveable clubhouse prankster in his younger days wasn't so playful when he was younger, but I'm not old enough to remember firsthand.

With the caveat that he'll easily get an elect-me vote from me, a lot of this would make for interesting discussion... but if this will rekindle old BTF flame wars, then we could just skip it and just say he did have my favorite Bermanism: "Be Home" Blyleven.

Too bad this election didn't take place 2 weeks ago. I agree Bert would eat this up. He'd probably also be all over the other electees from his era that haven't made the Hall (Grich, Santo, Evans, Evans..). Bert's self promotion should be used to our benefit.

At the very least we'll get a link here: http://www.bertblyleven.com/hall_of_fame.shtml

Oh yeah - 5-1 with a 2.47 ERA in the postseason and 2 rings.

But guys! Gene Collier of the Pittsburgh Post Gazette has solid evidence that Blyleven is a poor choice:

Here's my formula. If I'm managing a decent club that's going into Pittsburgh for a weekend series in July of 1979 and the Pirates are sending Bert Blyleven, John Candelaria and Bruce Kison to the mound, is there a pitcher among them that I think I might not be able to beat?

Yes, and it's Candelaria, who is not a Hall of Famer.

Now that we know that July 1979 is the key determinent factor in questions of greatness, we'd better tread carefully with The Dutchman...who posted a 2.64 ERA in July of 1979 (second best month of the season). He was trumped by Candelaria (3.64) and Kison (3.86) who both had more of those ERAs that month than Bert. See how easily his case falls apart!

To put it another way--does the fact that he should have won 322 games and didn't make him a lousy pitcher? Or is it the fact that he won 288 (or whatever), is that what makes him a lousy pitcher. Either way is just silly if you ask me.

If it's July of 1948 and the Indians are coming to town with Bob Feller, Gene Bearden, and Bob Lemon in rotation, the guy I figure I can't beat is probably Bearden -- who is not a Hall of Famer!

To put it another way--does the fact that he should have won 322 games and didn't make him a lousy pitcher? Or is it the fact that he won 288 (or whatever), is that what makes him a lousy pitcher. Either way is just silly if you ask me.

Well, you can acknowledge was a great pitcher and still be curious about how he managed to win 30 fewer games than he should have. No one is claiming he was lousy. It could be that this debate will be colored by the fact that Blyleven has not made the HOF yet... and people will focus their debate on the remaining BBWAA holdouts to maximize leverage for the next election. Except in this forum it will be preaching to the choir.

He got my elect-me vote years ago, but I wouldn't mind looking at some of the odd or colorful aspects of his career. After all, we put guys like Koufax & Clemente through the ringer, knowing full well that they would going to fly in with little resistance.

Typo in my post #6, above: I should have said Perry 337-258. The overall point of the post remains unchanged.

Re. #14, Doc, great find. What a freakin' moron. But why would we expect BBWAA guys to vote any differently than the way they write their columns. Here's the way things SEEM to me.

And I love it BTW where the guy says he has voted "possibly for Jack Morris."

From a post I wrote on SABR-L in 2004:

Blyleven's no-decision starts (data courtesy of Retrosheet):

4/6: Blyleven faced off against Montreal's Steve Rogers on Opening Day
in Pittsburgh, pitching seven solid innings and allowing two runs, but
left trailing 2-1. The Pirates tied it on Omar Moreno's single in the
bottom of the 8th, but lost 3-2 in the tenth on errors by Kent Tekulve
and Dale Berra.

4/21: The first of six consecutive no-decisions, and eight in nine
starts. In Houston, the Pirates came back from an early 3-0 deficit to
take a 4-3 lead into the bottom of the sixth, but Blyleven gave up
two-out singles to Ken Forsch, Terry Puhl, and Craig Reynolds to tie it,
then was lifted after walking Cesar Cedeno to load the bases. Grant
Jackson shut the door, but the Astros won it in 10 as Berra again
committed a key error to hang a loss on Tekulve.

4/25: In Cincinnati, Blyleven gave up two runs in the third, and left in
the seventh with a 2-1 deficit. Moreno once again drove in the tying run
in the eighth with a single, and this time the Bucs got a win for
Tekulve in the 11th.

5/1: Against Atlanta's Phil Niekro at home, Blyleven left for a
pinch-hitter in the bottom of the eighth, with the game tied at 2. The
Braves jumped on Tekulve and Enrique Romo for a three-spot in the top of
the ninth to take a 5-2 win.

5/5: In St. Louis, Blyleven was rocked in a 3-run fourth inning and was
removed trailing 4-0. The Pirates got two in the sixth, one in the
eighth, and then tallied three unearned runs in the top of the ninth to
take a 6-5 comeback win.

5/11: In a return engagement against the Reds in Pittsburgh, Blyleven
pitched six indifferent innings, allowing four runs. The Bucs tied the
score in the sixth as Blyleven left for a pinch-hitter, and the Reds got
one in the eighth and three in the ninth for an 8-4 win.

5/16: In Three Rivers against the Mets, Blyleven took a 3-2 lead into
the ninth inning. Steve Henderson led off with a single, and then Berra
booted John Stearns's grounder for an error. Tekulve came on for
Blyleven, giving up a sacrifice and an intentional walk, and then
Jackosn relieved and allowed Gil Flores to hit a sacrifice fly which
plated the tying run. Mike Easler's pinch-hit HR won the game for
Pittsburgh in the 13th.

5/26: After beating the Expos in Montreal on the 21st for his first win
of 1979, Blyleven was back in the no-decision rut in Shea Stadium. The
Mets cranked out four consecutive two-out singles in the third which led
to a five-run inning and an early shower for Blyleven. Facing a 5-1
deficit, the Pirates quickly responded with a run in the fourth off Mike
Scott and three in the fifth off Scott and Jesse Orosco to take Blyleven
off the hook, but they eventually lost the game 10-8.

5/31: Back home against the Cubs, Blylven left for a pinch-hitter in the
bottom of the sixth inning of a 2-2 tie. The Pirates got a run in that
frame to give Bert a 3-2 lead, but the Cubs tied it off Romo in the
eighth, and the Pirates were forced into extra frames before winning 4-3
in the 10th.

6/10: Blyleven had a 2-1 lead at Three Rivers entering the top of the
sixth against the Giants, but Bill Madlock's leadoff triple and Darrell
Evans's single quickly tied the game, and after an additional single by
John Tamargo, Romo came into the game. Aided by yet another Berra error,
the Giants took a 3-2 lead, but the Bucs got Blyleven off the hook yet
again in the bottom half with two runs, although San Francisco
eventually won 7-4.

7/8: In the second game of a doubleheader at Riverfront Stadium,
Blyleven allowed just one run in seven innings, but the Pirates could
only muster a single tally themselves off Tom Hume. Blyleven left in the
top of the eighth for a pinch-hitter, and Willie Stargell's homer off
Dave Tomlin in the ninth gave Pittsburgh a 2-1 win.

7/19: Against the Astros in the opener of a home twinbill, Blyleven gave
up three in the first and two in the fourth before leaving in favor of
Dave Roberts, tied at 5. Phil Garner's two-run shot in the sixth gave
Roberts his first victory as a Pirate, as the Bucs eventually won 9-5.

8/5: Pitching the first game of a double dip at home against the
Phillies, Blyleven was handed an early 3-2 lead, but departed in the
middle of a 6-run Philadelphia 5th inning outburst punctuated by Greg
Luzinski's grand slam. The Pirates fought back to tie it at 8 after
eight innings, and John Milner hit a gran slam in the bottom of the
ninth to salvage a 12-8 win.

8/10: At Philadelphia, again in the lidlifter of a doubleheader,
Blyleven lasted seven innings, allowing three runs. Tin Foli's two-run
homer in the top of the eighth tied it at 3 as Blyleven left for a
pinch-hitter, and the Phils eventually won 4-3 in 12.

8/20: In a wild one against the Giants in Pittsburgh, Blyleven pitched
six innings, allowing five runs and leaving with a 5-5 tie. An
eighth-inning passed ball by Mike Sadek allowed pinch-runner Alberto
Lois to score the winning run in a 6-5 victory.

8/25: In San Diego, Blyleven allowed the Padres two runs in the third,
then shut down the Pads through seven before leaving for a pinch-hitter,
down 2-0. The Pirates scored twice in the ninth to tie it (aided again
by a passed ball that allowed the tying run to score). Both teams plated
single runs in the 12th, and the Bucs eventually outlasted the Padres
4-3 in the 19th.

9/8: In Shea Stadium, Blyleven again allowed just two runs in seven
innings, only to find himself down 2-0. This time, the Bucs tied the
game in the eighth as Stargell batted for Blyleven and delivered a
pinch-single, but the Bucs eventually lost to old shipmate Dock Ellis in
the 15th.

9/15: Facing the Mets again in Pittsburgh, Blyleven gave up a pair of
runs in the first, but took a 4-2 lead into the seventh. Blyleven loaded
the bases with two outs, then yielded a two-run single to Lee Mazzilli
to tie the score and was replaced by Roberts. Roberts got the 5-4 win
when Bill Robinson drove in the tiebreaking run in the bottom of the 7th.

9/19: In the first game of a doubleheader at Veterans Stadium, Blyleven
allowed single runs in the first and third, and departed for a
pinch-hitter in the top of the sixth, trailing 2-1. Mike Schmidt hit a
grand-slam off Romo in the seventh to extend the lead to 6-1. The Bucs
got three off Tug McGraw in the eighth, then completed the comeback with
five off McGraw and Rawly Eastwick in the ninth, keyed by Manny
Sanguillen's two-run triple to take a 9-6 win.

9/29: Against the Cubs at home, Blyleven led 3-1 through five but saw
the lead evaporate on a Dave Kingman single and Steve Ontiveros's
two-run double. Romo came on and yielded two additional runs as the Cubs
took a 6-3 lead. The Bucs got one in the sixth and two off Bruce Sutter
in the seventh to tie it. Stargell's 13th-inning error gave Cub rookie
Bill Caudill his first major league win after seven straight losses.

The thing that struck me about these 20 no-decisions is that Blyleven
left with a lead only twice, while departing with a deficit nine times
and in tie games nine other times (including a couple of cases where the
Pirates tied the game as he was leaving for a pinch-hitter). Although
Blyleven did tend to receive less run support in his starts than his
teammates (receiving less-than-expected run support in 23 of 37 starts
in 1979), poor run support was a hindrance in only about a third of his
no-decisions; the rest of it was generally Blyleven's own inability to
hold the opposition down. Looking at his pattern of runs allowed, my
best estimate of his likely won-lost record, had the Pirates scored for
him as they did for his teammates, would be about 14-11.

-- MWE

And 14-11 is exactly what he needed for the Coop! : )

Let's talk a minute about Blyleven versus Carter for the 1-2 slots. It's academic really, they'll fly in, but they represent a challenge.

I originally thought it would be BB then GC on my ballot, but the different between them is slightly in favor of Carter. I don't know if my system is right though.

I run everyone through the big Keltner system. That gives me

Carter 45
Bly 47

Then assign 10 points for being the best available at the position.

Carter 55
Bly 57

Then I assign a floating ranking that shows a guy's standing relative to positional peers, it's simply a floating constant (representing ideal positional balance) minus his ranking. For pitchers that number is divided by 3 to approximate the relative number of pitchers versus players at each position. I add that to each:

Carter 67
Bly 71

Then I adjust catchers and pitchers since they tend not to earn as many points in my system as (other) hitters. Pitchers go up 16% and catchers 24%. Which leads to a final score of

Carter 83.4
Bly 81.5

That's wicked close, and it's an arbitrary rating. It's close enough to wonder if there's any real good reason that I should reverse their orders or keep it this way.

And 14-11 is exactly what he needed for the Coop! : )

:-)

Yeah, Blyleven's no-decisions in 1979 are unrelated to his large ERA/W-L discrepancies from 1971-77. Bert's 108 ERA+ in 1979 is unremarkable in the context of the rest of his career and the .706 is "lucky".

thanks Mike

"I strongly believe Bert Blyleven should be in the Hall of Merit."

- Bert Blyleven

Although
Blyleven did tend to receive less run support in his starts than his
teammates (receiving less-than-expected run support in 23 of 37 starts
in 1979)

How are you defining the bit in bold? I ask because it's key to distinguish between average and median. In a league that average 4.00 runs per game, a sizable majority of starts will receive less than the average. A team can always score five more runs than the average, but they can't score five fewer in that circumstance. If the above quote doesn't take that into account, Blyleven may actually have recieved less-than-expected no more often (or perhaps even less often) than he should've.

Yeah, Blyleven's no-decisions in 1979 are unrelated to his large ERA/W-L discrepancies from 1971-77.

Blyleven, in the early part of his career, was left in during the late innings of close games - many of which he lost, and in a fair number of which he surrendered runs late to blow a lead and/or a tie game, hence his reputation as a pitcher who "pitched just well enough to lose", which was well-established by 1975. When he went to Pittsburgh, Chuck Tanner started out handling him the same way, but by 1979 Tanner was going to his bullpen earlier and more frequently when the starter got into trouble. Blyleven hated being taken out of games, and jumped the team early in 1980 when Tanner showed no signs of changing his approach.

-- MWE

How are you defining the bit in bold?

It was based on Pittsburgh's run-scoring pattern and the number of innings Blyleven pitched in each game. If Blyleven had 7 batting innings of support (e.g. he pitched between six and seven innings on the road, or between 7 and 8 innings on the road) I looked at the distribution of runs scored by the Pirates over seven innings in games that Blyleven did not pitch and determined where Blyleven's support for that game fell on that chart.

-- MWE

Blyleven did get some run support in the 1979 playoffs and world series ...but it turns out that he did not need much ... he pitched incredibly well in the high stakes games. He even pitched out of the bullpen in the WS. Whether or not he is HOF or HOM worthy ...who knows? He, and his 1979 teammates, has a world series ring because he rose to the occasion.

He, and his 1979 teammates, has a world series ring because he rose to the occasion.

Gene Collier of the Pbgh Post-G disagrees with you. Of course, he's sillyhead, but he gets to vote for the Coop and we don't.

Blyleven did tend to receive less run support in his starts than his teammates (receiving less-than-expected run support in 23 of 37 starts in 1979)
. . .
It was based on Pittsburgh's run-scoring pattern and the number of innings Blyleven pitched in each game. If Blyleven had 7 batting innings of support (e.g. he pitched between six and seven innings on the road, or between 7 and 8 innings on the road) I looked at the distribution of runs scored by the Pirates over seven innings in games that Blyleven did not pitch and determined where Blyleven's support for that game fell on that chart.

When in baseball history is this statistic reliable --greater than 0.5 a reliable indicator of good run support?

Innings support is discrete and always an integer, hence "three times as discrete" as innings pitched. For many to most pitchers in many to most seasons, there must be a few consecutive integer numbers of innings support that dominate the distribution.

And run support is discrete. For an extreme example of the interaction between these two discrete variables, suppose average scoring is 3.96 runs per game. Then
3.96 is average runs in nine support innings
3.52 is average runs in eight support innings
3.08 is average runs in seven support innings
So four runs is "above average" and three runs is "below average" in seven, eight, or nine support innings.

Nine support innings must alone dominate the distribution for many pitcher-seasons. It represents nine innings pitched, home or road, plus eight inning pitched, road.

One moderately mitigating fact at least through the deadball era is that several of the complete game outings in a typical season for a stereotypical frequent-complete-game pitcher were 7 or 10 innings. There were more game called for darkness and more nine-inning ties in that era.

Cross-posting from the annual Hall of Fame discussion thread:

Mike Emeigh: I need to repost the study data that I have for Blyleven, but the basic issue with him is that he did very poorly when handed a lead, especially early in the game, compared to other HOF pitchers of his era. My expectation from an HOF starter is that when you give him enough runs to put him him in front, he stays there, and Blyleven didn't do that...

My expectation of an HOF pitcher is that he hold leads at far above the rate that a typical pitcher holds leads. And what I found when I looked at it was that when Blyleven was given a lead, he lost those leads at something closer to the rate of an average pitcher than that of an elite pitcher.

It is not enough, in my book, to pitch well but lose or get a no-decision if your team gave you a lead. The mark of an elite pitcher, to me, is that once his team helps him enough to get ahead he does the rest of the job. Blyleven didn't do that often enough, so he falls below the elite.

My response:

Mike Emeigh, that's one peculiar way to look at things. What about pitchers who give up 1 run in 7 innings, while the opposing team scores 1 or 2 runs over that span, but then the pitcher's team scores 2 in the ninth to win the game? Did that SP do nothing for his club??

Look, the argument that the distribution of a pitcher's runs allowed is significant is obviously a valid one. Clearly, a P who gives up 91 runs in one start and 1 run in each of the other 29 starts is more valuable than a P who gives up 4 runs in each start. But focusing exclusively on situations when a P has a lead, while ignoring all other situations, seems nothing short of crazy to me.

Fortunately, we have a more sophisticated tool at our disposal than mere percentage of leads held. It's called Win Probability Added, I imagine you've heard of it, and it measures precisely what you are looking for: What impact did this pitcher have on his team's wins and losses? Unfortunately, we're missing it for the first four years of Blyleven's career, but we can extrapolate.

Just backing out from his innings and ERA+, Blyleven's teams "should" have won 29.5 games more than average from 1974-1992. In fact, his WPA for this time period was just 22.3. Precisely as Mike Emeigh and Blyleven's myriad other detractors state, Blyleven's actual contributions to his teams were less than what a simple ERA+ and IP analysis would suggest.

Let's assume that Blyleven's tendency to pitch "away from the score" was constant throughout his career. If we plot his expected wins above average against his actual WPA, we get the following equation: WPA = .815*expected WAA - .1. Applying that equation to the four years we do not have WPA data for, Blyleven's career WPA would be 33.0, instead of the 43.1 implied by his ERA+ and IP.

OK, using a replacement level of 2.4 wins below average per 200 innings, Blyleven would come out at 102.8 wins above replacement using ERA+ and IP, and only 92.7 wins above replacement using Win Probability Added. That drops him from being a peer of Steve Carlton (103.4 wins above replacement by this method) to the level of Nolan Ryan (94.1 wins above replacement using this approach).

I'm happy to grant that because of his poor distribution of runs allowed, Blyleven was not as good as Carlton, as ERA+ would suggest, but instead was one notch below, in the Nolan Ryan class of starters. But if you're going to tell me that Nolan Ryan isn't a Hall of Famer, then I think we may have to boot you off the island. :)

Two-and-a-half years ago we discussed the complex issue of a pitcher's run support and control of opponent scoring --complex because they interact, at least in the data if not in the hearts and minds of the players.
Refer to #20, 26, 28, 31. (and not to DanR's today comment)

I confess that I don't understand the rhetorical question that opens my own comment #31.

Substantially my comment now seems to me a contribution to elaborating or illustrating Mike Emeigh's method rather than criticism in the negative sense. It's clear #28 that Mike handles the substantial matter well, game by game. The only open question to me is how he aggregates and summarizes that data. Especially, when if at all do R runs in I innings count as "average support" when R is the median of the reference distribution? Is there some region for team frequencies of >4 and <4 runs support where R=4 counts as average?

Essentially this is equivalent to the question whether "less-than-expected run support in 23 of 37 starts" implies greater-than-expected run support in 14 of 37 starts.

(To tune of "Matchmaker Matchmaker")

Blyleven, Blyleven,
I cannot lie,
I hope the Hall
Once more denies
Blyleven, Blyleven
I know its wrong
But I hope you're still denied.


Blyleven, Blyleven,
wonderous curve
Still I don't think
That you deserve
To be enshrined with a plaque on the wall
With Ruth, Cobb and Mays, et al.

I know your
Stats are terrific

I know there's
A logical case

But Bert, well, it doesn't matter
You just don't have a Hall of Fame face.

Blyleven Blyleven,
They'll try once more,
E-R-A-plus
Strikeouts Galore
But they forget that you never were great
Rejection remains your fate.