It's not that he wouldn't be trying to get them out, it's just that he'd be more willing to let the hitters put the ball in play,

Andy, that just makes no sense. We see what happens when pitchers "are willing to let the hitters put the ball in play": they do worse. Such as when the bases are loaded:

2009 AL, bases empty: .263/.328/.427 (.755 OPS)
2009 AL, bases loaded: .295/.328/.459 (.787 OPS)

There, the pitcher has made a conscious decision to try not to walk the batter. The batter ends up with a higher batting average and SLG, while the OBP stays the same *even though the walks are reduced*. It didn't help the pitcher at all. (Yes, in this case the batter expects the pitcher to come in; but if pitchers are "pitching to the score," the batters will expect them to come in then as well.)

Or:

2009 AL, 0-0 count, first pitch: .346/.351/.578 (929 OPS)
2009 AL, 3-0 count, next pitch: .425/.966/.802 (1.768 OPS)

(Yes, we can toss out the .966 OBP there given the count situation; but the point is that batters hit .425 when they were expecting a strike.)

See how batters hit better when they're expecting a strike and/or when they get a strike? (Note also how batters hit better when they get into hitting counts, and worse when the pitcher has the advantage. You're essentially advocating that a pitcher "pitch to the score" and thereby automatically put the batters into hitting counts.) Why would a pitcher want to turn the batter into a better hitter by feeding him pitches in the strike zone? It's not logical at all. It doesn't help the pitcher at all.

Andy, that just makes no sense. We see what happens when pitchers "are willing to let the hitters put the ball in play": they do worse. Such as when the bases are loaded:

2009 AL, bases empty: .263/.328/.427 (.755 OPS)
2009 AL, bases loaded: .295/.328/.459 (.787 OPS)

Sac flies. Correcting for those, the bases loaded line is .269/.328/.417.

Ray, none of those scenarios are comparable to the one I've outlined.

First, I'm talking about situations where for the most part the bases ARE empty, because that's precisely when a pitcher with a big lead wouldn't feel that a solo home run would be that hurtful.

Second, WRT the pitch count, that's also irrelevant, since I'm talking about pitchers who would be trying to throw strikes right from the first pitch, rather than nibbling as they might otherwise do in a tight game.

But most important, your wholly logical argument rests on an assumption that pitchers and batters have always had this sort of statistical breakdown on hand. While that might be true in the last decade or two, baseball strategy didn't begin with Earl Weaver or Project Scoresheet. It may even be true that Jack Morris didn't know this!

"And it's also logical to think that this would have affected more pitchers in past eras when complete games were more prevalent than they are today."

And as noted before, then the ERA+ is balanced out.
Since they're all letting up in the 9th with a 6-0 lead, they all give up some extra runs - so no disadvantage.

I find the starting concept silly, but even if true, it wouldn't mean anything.

"And it's also logical to think that this would have affected more pitchers in past eras when complete games were more prevalent than they are today."

And as noted before, then the ERA+ is balanced out.
Since they're all letting up in the 9th with a 6-0 lead, they all give up some extra runs - so no disadvantage.

Sure, if all pitchers had adopted the same strategy. If.

I find the starting concept silly, but even if true, it wouldn't mean anything.

Of course it would, if it were true. It would mean that at least some of the pitchers who claimed to be "pitching to the score" might be telling the truth.

And again, I'm not claiming this as a fact. But it's not something I'd dismiss without further knowledge. Unlike you and Ray, I don't claim to be a complete mindreader of what's gone on inside every pitcher's head, past and present.

And guess what? You all can now have the last word on this, because if this sub-topic went on for another 8300 posts, it wouldn't change a thing. People will keep comparing apples to oranges just to amuse themselves, and there's nothing anyone can do to prevent them.

Ray, none of those scenarios are comparable to the one I've outlined.

Are you not in effect putting the hitters into hitter's counts by pitching to the score?

The entire explanation for pitching to the score is "he has decided to give up more runs rather than less." But that makes absolutely no sense.

And as noted before, then the ERA+ is balanced out.
Since they're all letting up in the 9th with a 6-0 lead, they all give up some extra runs - so no disadvantage.

What about my scenario, no "pitching to the score".

But, elite SP's used to pitch the 200 IPs modern guys do, say at a 125 ERA+ rate. But then, they had to pitch another 50 IP as their own "mop-up" men, and because they were tired, pitched at a less effective rate, say a 100 ERA+. So, they end up at a 120 ERA+.

Today that 50 IP is transfered, either directly in blowouts, or via chaining, to 6th and 7th RPs, who used to not exist, and who pitch to 70 ERA+. By replacing those bad RPs, the elite SPs of yesterday actual reduced league ERA, b/c their tired 100 ERA+ is still better than the bad RPs 70.

So, the old-time SPs pitched more, hurting their ERA, and "replaced" bad pitchers, helping the league ERA, both of which lowered their own ERA+.

The entire explanation for pitching to the score is "he has decided to give up more runs rather than less." But that makes absolutely no sense.

No, it's not. The concept is that the pitcher takes an increased risk of giving up a small number of runs in exchange for a decreased risk of allowing a large number of runs. Say the pitcher's team is up by 4 in the bottom of the 9th, and he has the choice of two pitching methods that would result in the following run expectancies:

Method 1: 0 - 50%; 1 - 10%; 2 - 10%; 3 - 10%; 4 - 10%; 5 - 10%
Method 2: 0 - 30%; 1 - 20%; 2 - 20%; 3 - 20%; 4 - 5%; 5 - 5%

The run expectancy of the first approach is 1.5, whereas for the second, it's 1.65. But with a 4-run lead, the second approach has a better chance of winning the game.

All numbers are for illustration only (obviously, since nobody's run expectancy at the beginning of an inning is going to be 1.5).

There is no information in W-L record that isn't (a) swamped by noise and (b) captured in other statistics.

How do you know that (b) is true, if you can't find the information because of (a)?

Because I know the rules for assigning Ws and Ls to pitchers.

For the record, I don't believe that (a) is true, anyway. W-L record (not so much for a season, but for a career) gives you useful information concerning the "shape" of a player's performance, which when trying to distinguish between two players with similar bulk is IMO pretty valuable information.

I'm fine with looking at "shape" rather than just career totals. What information about a player's career "shape" is revealed by a player's career W-L record, other than showing that he's a starter rather than a reliever?

(To pick an example completely at random, Jack Morris is 254-186. What information does that tell you about his career shape?)

The whole thing about pitching to the score, to me, is that it sounds like a way for a pitcher to pitch the way he should have been pitching all along.

I have no problem with the idea that a pitcher, who has a good lead, might decide to be more aggressive, work quickly, go after the batters, and throw more strikes, instead of nibbling. But this is the same advice I keep hearing, over and over, for ALL pitchers to be effective -- throw strikes, work quickly, and go after the hitter. It keeps your pitch count down, avoids walks, keeps your fielders on their toes, and helps the pitcher establish a rhythm.

There are some guys -- a very few -- like Glavine, who can nibble and use strategic walks to their advantage, but most guys can't. The best pitchers generally don't give up a lot of walks, and throw a lot of strikes.

So if a guy was doing what we would expect by "pitching to the score", why wouldn't that make him more effective, and not less?

I have no problem with the idea that a pitcher, who has a good lead, might decide to be more aggressive, work quickly, go after the batters, and throw more strikes, instead of nibbling. But this is the same advice I keep hearing, over and over, for ALL pitchers to be effective -- throw strikes, work quickly, and go after the hitter. It keeps your pitch count down, avoids walks, keeps your fielders on their toes, and helps the pitcher establish a rhythm.

Not to mention that it stops you from wanting to shoot live ammunition and dead cats at the television screen.

There are some guys -- a very few -- like Glavine, who can nibble and use strategic walks to their advantage, but most guys can't. The best pitchers generally don't give up a lot of walks, and throw a lot of strikes.

And then there's Dr. Joba and Mr. Chamberlain, who often drinks his secret formula in between innings just to keep Scotland Yard on its toes.

SBB, I'm going to try to follow your logic to see where we get if we evaluate players by their number of wins. I like to credit a "win" to the starting first baseman provided he plays at least 5 innings in a game, his team wins, the player exits or completes the game with the lead, and the lead does not change hands after the first baseman leaves the game. Albert Pujols is a fine player certainly, but he really didn't do much last year where it counts to differentiate himself from James Loney. The Brewers went from playoff contenders in 2008 to pretenders in 2009 due to a big fall off in wins from the slightly below average Prince Fielder. Adrian Gonzalez had a catastrophic season and was lucky not to have been cut by the Padres. The 1939 Yankees were very fortunate that Babe Dahlgren could step up to vastly exceed Lou Gehrig's typical production.

Today that 50 IP is transfered, either directly in blowouts, or via chaining, to 6th and 7th RPs, who used to not exist, and who pitch to 70 ERA+. By replacing those bad RPs, the elite SPs of yesterday actual reduced league ERA, b/c their tired 100 ERA+ is still better than the bad RPs 70.

Repeating this on a new page doesn't make it any more true.

Then aren't the support neutral stats that Wolverton came up with a decade ago -- expected wins and losses -- better than wins?

Yeah, and they'd apply here because you can use them with Morris. But, prior to the Retrosheet era, you run into a problem. Over the years, that should change. They just added pbp info for 1871. We'll find out whether or not Rynie Wolters was clutch.

Would you say, GGC, that Malcolm's use of "neo-sabe" is similar to the way Backlasher refers to certain sabermetricians as "saberists" in, say, Moneyball & DIPS discussions? That's the way I've always thought of it, but thought I'd get another's view since it's being discussed a little.

BL showed up after Don's heyday but, IIRC, he once said he would have either loved him or hated him. I can't speak for them, but they are coming from two different perspectives. Don was a sabermetrician of sorts; published a few books under the rubric of Big Bad Baseball Annual. Backlasher, OTOH, was exposed to advanced mathematics somewhere (he could discuss abstract algebra; #### where 1+1=3) and saw the limits of STATS 101 type analysis (Hey! I took quantitative courses in college. I am a better person than Joel Sherman.), but had difficulty expressing it so that a layman like me could understand it.

They both did have an acid tongue, though. That might be where they are similar.

Today that 50 IP is transfered, either directly in blowouts, or via chaining, to 6th and 7th RPs, who used to not exist, and who pitch to 70 ERA+. By replacing those bad RPs, the elite SPs of yesterday actual reduced league ERA, b/c their tired 100 ERA+ is still better than the bad RPs 70.


Repeating this on a new page doesn't make it any more true.

How is it not true? I'm looking for a counter-argument if you have one.

If the top-2 starters per team used to pitch 250 IP and now pitch 200, where do those extra IP come from? It has to be the 6th/7th RP. 30 years ago teams only carried 5 RPs.

"Today that 50 IP is transfered, either directly in blowouts, or via chaining, to 6th and 7th RPs, who used to not exist, and who pitch to 70 ERA+."

Can you give me an example of this?

How is it not true? I'm looking for a counter-argument if you have one.

As I said several pages ago, teams have radically re-structured their thinking about this question and now make a concentrated effort to cultivate relief pitchers in a way that did not exist back in the day.

I'm not saying this theory is definitely wrong. I am saying that I'd like to see some actual data which confirms that starter ERA dipped significantly in innings 8 and 9 back in the day, data telling us what the numbers look like for relievers in the late innings of games thrown by superstar pitchers, and some systemic effort to put together an idea of what starter ERA+ numbers would have looked like in, say, 1963 if you treated them like modern 7-inning pitchers and gave them the same kind of relief support.

Obviously, the fact that they used to throw 300 inning seasons means those individual seasons have more value. And obviously, there is a reasonable case to be made that it's easier to have an outlier ERA+ in less innings. What I'm disputing is that there is some third effect that shows up in starter ERAs but is not captured by ERA+.

Or, what 419 said.

"Today that 50 IP is transfered, either directly in blowouts, or via chaining, to 6th and 7th RPs, who used to not exist, and who pitch to 70 ERA+."

Can you give me an example of this?

Sure, just picking two Yankee teams.

The 1980 NYY had Tommy John and Ron Guidry as the #1-2 SP. They combined for 484 IP, 21 CG, posted ERA+ of 114 & 110. The team had 5 RP's who pitched >10 IP all of whom had an ERA+>100.

The 2009 NYY had Sabathia and Burnett as #1-2. They combined for 437 IP, 3 CG and ERA+ of 127 & 106. The team also had 5 RP's with an ERA+>100 but used a total of 12 RPs (>10 IP). 5 of those RP's had ERA+ <80.

John and Guidry pitched an extra 47 IP, even though they weren't quite as good as Sabathia/Burnett, and Sabathia is the most durable SP in MLB today. Given the, CG's and the fact the team had Gossage as an elite closer (33 SVs/173 ERA+), those extra IP were probably amassed mostly by finishing games with decent size leads. Sabathia and Burnett would almost always be pulled in those games after 7 or 8.

In 2009 those IP have to eventually be filled, and since the good (top 5) RPs pitch fewer IPs than their 1980 counterparts, those IP eventually go to the bad (6th,7th,8th) RPs. These IP will have poor performance. It is quite likely, though you'd have to do the research, that John/Guidry pitched worse in their late innings than their average, given fatigue, but not as bad as the bad RPs of today.

The combined effect is probably to hurt the SP ERA, but improve the team ERA.

I don't know how to test this in any systematic fashion w/o going through hundreds of box scores for pitchers of the past.

The 2009 NYY had Sabathia and Burnett as #1-2. They combined for 437 IP, 3 CG and ERA+ of 127 & 106. The team also had 5 RP's with an ERA+>100 but used a total of 12 RPs (>10 IP). 5 of those RP's had ERA+ <80.

Not true. They had 3, pitching 25, 22, and 13 innings. You can't count Mitre and Wang, as they amassed the vast majority of their innings as starters, both starting 9 of 12 games.

edit: Mitre pitched 7.2 in relief, Wang 8.0

Tommy John (1980)
Split G PA AB R H 2B 3B HR SB CS BB SO SO/BB BA OBP SLG OPS TB GDP HBP SH SF IBB BAbip tOPS+ sOPS+
Innings 1-3 36 439 400 41 107 23 6 8 1 6 29 36 1.24 .268 .318 .415 .733 166 16 1 8 1 1 .277 112 103
Innings 4-6 35 412 383 51 113 19 3 3 1 5 20 24 1.20 .295 .334 .384 .718 147 5 3 5 1 0 .308 109 100
Innings 7-9 26 233 219 17 49 7 3 2 0 1 7 18 2.57 .224 .251 .311 .562 68 12 2 2 3 0 .233 63 59

New York Yankees (all pitchers) (2009)
Split G PA AB R H 2B 3B HR SB CS BB SO SO/BB BA OBP SLG OPS TB GDP HBP SH SF IBB BAbip tOPS+ sOPS+
Innings 1-3 162 2096 1848 256 482 102 8 58 56 21 202 406 2.01 .261 .338 .419 .756 774 46 21 8 17 3 .303 106 102
Innings 4-6 162 2148 1905 286 506 102 10 66 41 20 191 412 2.16 .266 .338 .434 .772 826 34 27 7 18 8 .304 110 99
Innings 7-9 162 1910 1690 208 388 69 6 55 25 10 172 420 2.44 .230 .307 .375 .682 634 30 23 13 12 16 .271 86 89

It's not really what I expected.

[Data from BB-Ref.]

Not true. They had 3, pitching 25, 22, and 13 innings. You can't count Mitre and Wang, as they amassed the vast majority of their innings as starters, both starting 9 of 12 games.

Sorry, I meant to say < 85 ERA+: Abaladejo, Veras, Ramirez, Tomko and Marte. Wasn't counting Mitre and Wang.

It's not really what I expected.

So, John got much better the later the game went? Wow!

Can you do that search automatically for a large number of pitchers/years, or is it manual? I'll have to check out B-Ref.

You'd have to look at something like the best 25 SPs per year for 1955-1980 and see what their 8th/9th IP performance was like.

If they did get worse, it call into question the comparability of ERA+ across eras, if they didn't get worse, it really calls into question today's approach to managing SP.

It's not really what I expected.

If managers have any clue about how their starter is pitching, there should be a selection bias at work.


Ron Guidry (1980)
Split G PA AB R H 2B 3B HR SB CS BB SO SO/BB BA OBP SLG OPS TB GDP HBP SH SF IBB BAbip tOPS+ sOPS+
Innings 1-3 30 361 320 31 79 14 3 8 3 8 35 74 2.11 .247 .319 .384 .704 123 8 1 1 4 1 .293 96 95
Innings 4-6 31 362 319 43 91 18 3 6 2 3 32 50 1.56 .285 .344 .417 .761 133 8 0 4 7 0 .315 113 111
Innings 7-9 30 206 188 18 45 6 2 5 0 2 13 42 3.23 .239 .287 .372 .659 70 2 0 4 1 0 .282 84 86


EDIT: Just for completeness.

If managers have any clue about how their starter is pitching, there should be a selection bias at work.

Well, yes and no. The point is that in 1980 we think the manager was more likely to leave John and Guidry in. The hypothesis seems to be that you don't go to the bullpen if you've got a reasonable lead and the starter is staying out of trouble. The stats seem to indicate that it was the right thing to do, anyway.

They both did have an acid tongue, though.

Yeah, "Jack Crust" was just brilliantly witty.

I have been out all day doing good work, so I'm trying to catch up. Great baseball discussion, so you can consider me a hopefully-not-derailing outlier.

FIRST OF ALL:

Precisely the point, even if in our hearts we know that Bach is the GOAT.

Andy, I really hope this is some reference I don't get and not some kind of dismissal of Bach. Seriously. (And, if you weren't aware of the tale, it was Bach who walked 100 miles in the winter to hear Buxtehude.)


Snapper, if you're willing to fully attest to the greatness of Celine Dion as a composer, Dan Brown as a writer, and a velvet elvis as artwork, your point is made.

Your dismissal of modern art is kind of sad. You've said that great art brings joy, and a LOT modern art has brought me a lot of joy. Much has also made me scoff and roll my eyes. I find beauty at the Met and the MOMA. There was show called TANGLED ALPHABETS at the latter recently that was breathtaking.

Salieri wrote excellent music. His music, however, was not as good as Mozart's. You want the line? There's the line. It definitely exists. I could show you this, I could play you the music, and you would hear it. And I don't even like Mozart. Of course there's a level of subjectivity, and of course musical training clarifies the line to the ear. But it isn't a up-turned nose line, it's a line of the ear, a line of construction, a line of melody. As a metaphor - just because someone can't detect when someone's out of tune doesn't mean they somehow aren't actually out of tune.

Andy, you bring up Gershwin and Stax/Volt. I can't answer that, that's tin pan alley vs. Stax/Volt. But there were a TON of fungible Gershwin contemporaries, as well as a ton of acts making records in the Stax/Volt era who were pure crap. For every lesser tin pan alley guy or Stax/Volt guy who didn't get the audience he rightfully deserved, there were endless replacement-level or worse creators who RIGHTFULLY didn't get any audience. Because they were below the line. SOMEONE liked them, but they were still below the line.

There is a lot of subjectivity. I ADORE Journey and Robert Ludlum. It's not a line I throw about with any kind of definition or unquestioning debate. Subjectivity is a given, but again, so is the line, to say it doesn't exist within acceptable comparable pieces or genres is not realistic.


Penderecki is an interesting one. Which Rodriguez is he? Some would say Alex, others Ivan. I'm quite fond of him, myself.

Oh, I mean Alex Rodriguez. Carter is rightfully given the honors, he deserves every one of them. But Penderecki is top-level as well, and for a long time hasn't gotten the "Yes, he is THAT good" that Alex has also been denied for various reasons.

Are you talking solely about classical compositions here? I'd have to assume that you are, since if you delve into popular music the exercise is almost comical in its futility. How on Earth can you possibly compare Gershwin with the best songwriters of the Stax-Volt complex? What would you be measuring, and who's an "objective" judge? Live audiences, dead audiences, live musicians / critics, or dead musicians / critics? And don't even begin to get into what vocalist's rendition of any particular song is the "best."

On this, I agree with everything Lassus says just above, and would add that Gershwin vs. Stax-Volt artists is directly analogous to the question of Bach vs. Penderecki with regard to the line. They're not getting compared to each other, but to the others who wrote in sufficiently similar styles as to be directly comparable. (Which is just another way of phrasing what he says.)

The discussion about art above is a good one. Art seems both more and less commoditized than it was during the Renaissance. Instead of noble and/or ecclesiastical patrons, artists now have the market and universities, which means that on one hand, they don't have to do anything that anyone likes as long as they do enough of the right sort of thing to get hired and get tenure. On the other hand, they do have to do those things, which are determined by other academic artists and by critics. And there is always going to be the pressure to sell the artworks, which will push the art in certain aesthetic directions according to what private patrons want to buy.

That's very much like music, except that music today is very much divided into the "popular" world, in which musicians do for the most part rely on the market, and will tend to be stylistically shaped by market forces, and the "academic" (or "classical") world, in which the musicians still like to sell scores and recordings and such, but in which most of them have (or are trying to get) academic jobs that put food on the table regardless of what any sort of public beyond academia wants them to do.

Lassus, I'm pretty sure that "GOAT" stands for "Greatest of All Time," so Andy isn't denigrating Bach.

And to everyone who likes Bach and/or choral music: take it from two choristers that Kryzstof Penderecki (b. 1933) is an A-Rod level of composer.

Lassus, I'm pretty sure that "GOAT" stands for "Greatest of All Time," so Andy isn't denigrating Bach.

Ah. That makes sense. Thank you. Let me unload my luger.

FIRST OF ALL:

Precisely the point, even if in our hearts we know that Bach is the GOAT.

Andy, I really hope this is some reference I don't get and not some kind of dismissal of Bach.

Sorry for slipping in a reference to Muhammad Ali in the context of a discussion of classical music composers, but the "GOAT" here means "Greatest Of All Time."

(Jeez, glad I straightened that one out! How rumors get started, etc......)


Andy, you bring up Gershwin and Stax/Volt. I can't answer that, that's tin pan alley vs. Stax/Volt. But there were a TON of fungible Gershwin contemporaries, as well as a ton of acts making records in the Stax/Volt era who were pure crap. For every lesser tin pan alley guy or Stax/Volt guy who didn't get the audience he rightfully deserved, there were endless replacement-level or worse creators who RIGHTFULLY didn't get any audience. Because they were below the line. SOMEONE liked them, but they were still below the line.

Well, I'd certainly agree with that on a subjective level, but how are we to determine which of those neglected artists deserved their neglect and which ones didn't? I could certainly name many cases of underappreciated R&B;vocalists in the 60's whose talent was easily as good or better than those who became mainstream household names, but who lacked a key factor that had nothing to do with musical talent. In the broader "arts" world as well as in life, it often boils down to "who you know, not what you know." Is Annie Leibovitz really a better" portrait photographer than Robert Bergman, or is she little more than a great networker who's always known which way the wind is blowing? (if not the real estate market)

All I'm saying is that popular consensus is a very imperfect way of determining quality when it comes to music or art of any type, and critical consensus only a bit less so.*** I see your point about a "line," but it's a lot easier to sort out the greats from the shlubs than it is to differentiate among those near the border.

***This is especially so when critics fall into one "political" camp or another, perfect example being Pauline Kael, perhaps the most pretentious anti-pretentious film critic who ever lived. There are a zillion other examples like this all over the spectrum in every field of the arts.

EDIT: I see that while I was off doing other things before hitting the "SUBMIT" button, I should have bought a six pack of coke while I was at it. Anyway, the first one to Vaux.

It's not really what I expected.

If managers have any clue about how their starter is pitching, there should be a selection bias at work.

There was.

Tommy John (1980)

Split        G ... tOPS+ sOPS+
Innings 1-3 36 ... 112    103
Innings 4-6 35 ... 109    100
Innings 7-9 26 ... 63     59

Splits by Length of Appearance
<= 3 Innings: 1 G, 0-1, 22.5 ERA
3.1-6 Innings: 10 G, 2-4, 8.82 ERA (started the 7th inning once without recording an out)
6.1- Innings: 25 G, 20-4, 1.98 ERA

Splits by Decision
22 Wins: 8.29 IP/G, 1.58 ERA
9 Losses: 6.22 IP/G, 7.23 ERA
5 ND (Team Record 4-1): 5.40 IP/G, 8.00 ERA

"If they did get worse, it call into question the comparability of ERA+ across eras, if they didn't get worse, it really calls into question today's approach to managing SP."

Unless it didn't matter so much if a tired SP kept going back then, because fewer hitters were HR threats, for example, as I mentioned 150 posts ago.
And assuming managers today don't have RPs specialized to be ready to pitch an effective inning every other day.

You seem to be assuming, for some reason, identical SP environment conditions to now.

I could certainly name many cases of underappreciated R&B;vocalists in the 60's whose talent was easily as good or better than those who became mainstream household names, but who lacked a key factor that had nothing to do with musical talent. In the broader "arts" world as well as in life, it often boils down to "who you know, not what you know."

Well, I do think there's a bell-curve. Something is always going to get wrongfully left behind, and there is always subjectivity, but you need to have a belief that the majority of excellence is recognized or you'll just be angry and bitter all the time. And I agree with you about popular consensus not being a great way or the only way to judge, hence my point to snapper. But it is at least A place to start, and a valid one.

And then, when we know what the popular consensus appears to have been, we have a framework by which to make our additional subjective judgments of the remaining repertory stand a degree of scrutiny. They'll still be subjective, but by comparing lesser-known works to those with established critical consensus, it can be argued that they fall into a first rank, or second rank, or what have you.

Our entire understanding of some periods is based on our knowledge of apparent consensus, because when we go back far enough, our knowledge is determined by what that consensus was. In the Renaissance, the more popular a piece was--the more highly regarded it was--the more distinct sources into which it was copied and/or the more copies of the more editions in which it was printed. And, recognizing attrition, the more likely it is to have survived at all. Almost by definition, works that were less well-known were less likely to be capable of becoming known to us in the future.

I convinced myself long ago that it was virtually impossible for a body of great works by a great composer to have been lost, because so much survives, and so much of what survives are redundant sources for the same works. It's certainly possible that a few individual great works have been lost, but our picture of the music of that time, and the group of composers then active, can't be distorted in any significant way. As great as Beethoven's Fifth Symphony is, Beethoven with no trace of the Fifth Symphony would be regarded the same way by us in relation to his contemporaries as he is, on the basis of all his other work. Most likely, at least one complete J.S. Bach cantata cycle is lost. If we somehow recovered those works, we could hardly think more highly of him than we do already.

I could certainly name many cases of underappreciated R&B;vocalists in the 60's whose talent was easily as good or better than those who became mainstream household names, but who lacked a key factor that had nothing to do with musical talent. In the broader "arts" world as well as in life, it often boils down to "who you know, not what you know."

Well, I do think there's a bell-curve. Something is always going to get wrongfully left behind, and there is always subjectivity, but you need to have a belief that the majority of excellence is recognized or you'll just be angry and bitter all the time.

No real argument with that, and from a consumer's POV, there's always the upside of thinking that you're a true connoisseur who appreciates talent that the great unwashed masses don't. Anyone who denies that he sometimes engages in that sort of thinking is a flat out liar.

But "angry and bitter"? Only if I were one of those neglected artists. Fortunately I can't sing a note or draw a straight line.

And I agree with you about popular consensus not being a great way or the only way to judge, hence my point to snapper. But it is at least A place to start, and a valid one.

I agree with that, too, as long as when we get into discussions like this, we all spell out our standards of quality, and our personal reasons for putting any given work or artist/musician, etc., over your line. Merely citing critical consensus doesn't cut it when subjectivity is such an important factor.