But....but.....after game 2 coolstandings.com promised me it was over! And the Supreme Court threw out my lawsuit!

They're lucky Clemente didn't decide to play for Baltimore in the series.

It doesn't look like he takes the difference in league quality into account

Breaking news:

Baseball games occur on baseball fields.

The Pirates were lucky to win the 1960 World Series.

It doesn't look like he takes the difference in league quality into account

I'm not sure 1971 was all that great a year for the NL. Reds weren't good, lost the All Star Game, Giants were weakest team in the playoffs.

The Pirates were lucky to win the 1960 World Series.

Not as lucky as some think.

The Pirates were lucky to win the 1960 World Series.

Not as lucky as some think.

The Pirates were 7 1/2 games ahead of the NL pack by mid-August and were never fewer than 5 1/2 games ahead until the pennant had been clinched. The Yankees were tied with Baltimore on September 14th, and it took a 15 game winning streak accompanied by an Oriole collapse to give them their final 8 game margin. The Yankees were playing in a demonstrably inferior league, and those three blowouts in the World Series were a hell of a lot more flukish than the fact that the Pirates eventually won. One bad hop turned that final game around, but it wasn't one bad hop that had left Casey Stengel with the likes of Jim Coates to call on with the game on the line. Run differentials over 154 or 162 games in the same league are meaningful; run differentials in a 7 game series aren't.

Luck? It was Willie Stargell willing them to victory!

I know he didn't do much in the 71 series. But still, Willie was there, menacing the pitchers from the bench and the on deck circle. Not so much when he was at the plate...but definately scaring them from afar.

Murtaugh was marvelous -- how many managers would be willing to use 6 different starters in the first 6 games?

Having the losing team score twice as many runs as the winning team in a postseason series is pretty crazy and very lucky.

They all are Cyril. They all are.

Cy, using your calc for WPCT, how many WS do you come up with where the team with the lower WPCT ends up winning? Seems as though this figure has risen since the introduction of divisional play.

And using the method, what WS is the greatest upset? (That type of data, while more time-consuming to produce, would be interesting to look at en masse--and would probably produce less wise-assery...though we might need to calculate the odds on that!)

what WS is the greatest upset?

I'll put my money on 1906.

1914 might be up there too, Andy. Not so much in terms of WPCT differential, but in terms of contemporary perception. The A's had won three of the last four WS.

W-L gaps over 10 games where the lower WPCT team won:

1906, 1945, 1974, 1985, 1987, 1990, 1995, 2003, 2006.

What about 1954 (the 97-win Giants over the 111-win Indians)?

I don't see anything wrong with those odds- Orioles at 58% and Pirate at 42%. That was probably pretty accurate going into the series. The Orioles were the favored team, and the Pirates won. It happens. See also 1969 (Mets over Orioles), 1972 (A's over Reds), 1974 (A's over Dodgers) and so on...

W-L gaps over 10 games where the lower WPCT team won:

1906, 1945, 1974, 1985, 1987, 1990, 1995, 2003, 2006.

wow--I had forgotten those records from 74 (Dodgers 102 wins, A's 90) yet the A's went into the series as a heavy favorite.

1987--the odd thing about those two teams is that each were predicted to have losing records based on Cy's OPS differential formula

Coke to #16. Eyeball cramp!

1906, 1945, 1954, 1974, 1985, 1987, 1990, 1995, 2003, 2006.

Teams suffering upsets in these Series: Cubs, Cubs, Indians, Dodgers (see #18 for clarification), Cardinals, Cardinals, Indians, Yankees, Tigers.

1914 might be up there too, Andy. Not so much in terms of WPCT differential, but in terms of contemporary perception. The A's had won three of the last four WS.

In terms of contemporary perception, I'm not so sure. On the day of the first game, the NY Times reported that the odds had been rapidly dropping to the point where the A's were only "lukewarm" favorites, from a previous indication of odds of 2 to 1.

In terms of reality, the Braves had gone 68 and 19 after July 4th. No World Series win by a team with a three month record like that should be seen as any sort of historic upset, certainly not on the level of the 1906 Hitless Wonders vs. a team that had gone 116 and 36 in what was thought to be a superior league. The league strength factor and the results of 1910 and 1911 would have made the 1914 A's a legitimate favorite, but to the extent that anyone thought of it as some sort of a shock (meaning the win, not the fact that it was a sweep), I'd say that they'd simply been misreading the true situation.

---------------------------------

What about 1954 (the 97-win Giants over the 111-win Indians)?

This example pops up all the time, but it's a total mirage. The Indians' 111 wins were entirely a product of an historically weak 1954 American League. They'd gone 89-21 against the hapless A's, Senators, Orioles, Tigers and Red Sox, but only 22-22 against the Yankees and White Sox. The Giants had compiled their 97 wins against vastly superior competition. Study those five rosters, count the number of genuine stars in their primes, note that you'll still have a full hand to spare when you're finished, and the significance of those 89 wins shrinks from a fat plum to a shriveled prune.

The Indians also played the Giants 21 times in Spring Training, and won but 8 of those games. That's "only" Spring Training, but that's not exactly a small sample size, unless you're simply throwing it all out on the spurious ground that only the Giants were trying to win those games.

I could look up the contemporary opinions if I had more time, but John Drebinger of the Times did a three part preview analysis, and thought that the Giants had a slight edge between two evenly matched teams. But perhaps that's because he'd seen the Indians' competition in action, and knew just how crapitudinous it really was.

As for the actual betting odds: The Indians were 8 to 5 favorites overall, and 11 to 10 in game 1. That's perfectly in line with the sort of odds you'll see in many World Series, nothing particularly unusual. And the odds after game 1 quickly switched to establish the Giants as the favorites.

On the day of the first game, the NY Times reported that the odds had been rapidly dropping to the point where the A's were only "lukewarm" favorites, from a previous indication of odds of 2 to 1.

Interesting stuff. Not sure I'd want to take it to the bank, given the way the gamblers were operating in those pre-Black Sox days.

Clearly the Hitless Wonders are in a class by themselves...the record of WS teams with standings leads of 12+ games over their opponent is 12-4, and they are the greatest outliers.

The Giants had compiled their 97 wins against vastly superior competition.

True. Interestingly, though, there were only two other teams over .500 in the NL that year, too. The Giants went 25-19 against them.

People probably underestimate the Giants' pitching staff. Their ERA against .500+ teams that year was 3.19, as compared to the Indians' 3.78. That type of breakout might be a better handicapping tool than just about any other...

also--I don't think anyone REALLY thought the 1995 series was an upset...so, winning 10 games more than your opponent did in the other league really doesn't mean that much

Did anybody really watch the 1995 World Series?

Teams with a 10+ game standings advantage over their WS opponent have gone 20-9. In all other instances it's a tossup: the team with fewer wins has a 40-36 record (essentially a coin flip).

On the day of the first game, the NY Times reported that the odds had been rapidly dropping to the point where the A's were only "lukewarm" favorites, from a previous indication of odds of 2 to 1.

Interesting stuff. Not sure I'd want to take it to the bank, given the way the gamblers were operating in those pre-Black Sox days.

I'd agree with that. My point was only to say that you can't reasonably put the 1914 World Series in the same category of upset as 1906, which as we both agree was in a class by itself.

Clearly the Hitless Wonders are in a class by themselves...the record of WS teams with standings leads of 12+ games over their opponent is 12-4, and they are the greatest outliers.

Especially when it seems to be the case that the NL of that time was a slightly superior league.

People probably underestimate the Giants' pitching staff. Their ERA against .500+ teams that year was 3.19, as compared to the Indians' 3.78. That type of breakout might be a better handicapping tool than just about any other...

I think what also throws many people off is that the 1954 Indians had so many Big Names on their staff: Lemon, Wynn, Garcia, Feller, Newhouser, Houtteman, Narleski and Mossi, all of whom contributed to a mind-boggling 132 team ERA+. That's a truly amazing performance, but of course the problem is that it was padded by their record against those five horrible teams at the bottom,. Against the Yankees and White Sox, that Indians' team ERA of 3.78 was actually HIGHER than the overall AL ERA of 3.72, a figure which of course included all those pitchers from the A's, Nats, etc. It's amazing how much those 111 wins can be inflated in their significance by totally ignoring the context in which they were achieved.

The Pirates were 7 1/2 games ahead of the NL pack by mid-August and were never fewer than 5 1/2 games ahead until the pennant had been clinched. The Yankees were tied with Baltimore on September 14th, and it took a 15 game winning streak accompanied by an Oriole collapse to give them their final 8 game margin. The Yankees were playing in a demonstrably inferior league, and those three blowouts in the World Series were a hell of a lot more flukish than the fact that the Pirates eventually won. One bad hop turned that final game around, but it wasn't one bad hop that had left Casey Stengel with the likes of Jim Coates to call on with the game on the line.

What is really odd is that Coates came into the game with lefty hitters Skinner and Nelson due up. Luis Arroyo, who had been Stengel's closer down the stretch and who had to that point pitched only 2/3 of an inning in the Series, wasn't even warming up. Stengel had failed to pinch-hit for Shantz, who had not thrown five innings in a game all year and who had gone as many as four innings only once before, with a chance to break Game 7 wide open in the top half of the eighth, in part because the Pirates had lefty hitters bunched at the top of the order (Virdon leading off with Skinner and Nelson 3/4).

If you weren't going to bring in a lefty, however, Coates at that point was probably as good a choice to relieve as any. Duren, Maas, and Terry were the other relievers (can't imagine that Stengel would have considered Ditmar).

-- MWE

Especially when it seems to be the case that the NL of that time was a slightly superior league.

But the NL was far more top-heavy. The Pirates, Cubs, and Giants were superior teams, but the rest of the league was mush. In the AL, Washington and Boston were the only truly bad teams.

-- MWE

What is really odd is that Coates came into the game with lefty hitters Skinner and Nelson due up. Luis Arroyo, who had been Stengel's closer down the stretch and who had to that point pitched only 2/3 of an inning in the Series, wasn't even warming up. Stengel had failed to pinch-hit for Shantz, who had not thrown five innings in a game all year and who had gone as many as four innings only once before, with a chance to break Game 7 wide open in the top half of the eighth, in part because the Pirates had lefty hitters bunched at the top of the order (Virdon leading off with Skinner and Nelson 3/4).

Stengel's entire pitching strategy in that World Series was highly questionable, a point which has been beaten into the ground but remains true. Ford could have started games 1, 3 and 7, but instead was held back until game 3 while Ditmar lasted less than an inning in game 1, and less than 2 innings in game 5. Stengel was undone by his obsession with platooning in game 1, and his ignoring the platoon factor in game 7.

So of course we can say that the Pirates were lucky to win game 7, since they seemingly had both Stengel and God** on their side, and without either the Yankees would certainly have won. But in the context of what those two teams did in the regular season, and in the context of the strength of the two leagues, that was no upset.

**in the form of whatever pebble or random piece of hard dirt caused Virdon's DP ball to take a flukish bounce and send Kubek to the hospital, while keeping the inning alive

Stengel was undone by his obsession with platooning in game 1, and his ignoring the platoon factor in game 7.

Stengel didn't want to use his lefties in Forbes Field, preferring to save them for Yankee Stadium. *Why* he would do that is really open to question, especially because Forbes Field was a lefty hitter's park. As it turned out, his lefties did much better than did his righties in Forbes.

Looking at how Stengel used his pitchers down the stretch, what strikes me is that he used Ford for two innings in the season finale against Boston. That looks for all the world as though he'd already decided not to start Ford in Pittsburgh three days later.

-- MWE

wow--I had forgotten those records from 74 (Dodgers 102 wins, A's 90) yet the A's went into the series as a heavy favorite.

The As had won in 1972 and 1973, that's why.

Basically in the past people looked at league quality by assuming that the WS winner's league was tougher, later they added the All Star game outcome to the mix. Both methods rely upon too small sample sizes to be useful. People also will do dumb things like look at triple crown numbers- Joe Schmoe lead the Al hitting .350-49-156, Jack Flack lead the NL by hitting just .312-34-117, obviously the AL is superior- of course that shows nothing of the sort- all that shows is that the AL likely had a higher offensive environment, it might show that Joe was better than Jack- it might even show that the NL is tougher than the AL because it is tougher to dominate the NL...

In 1919 the WhiteSox were overwhelming favorites because... because the AL was regarded as the better league because it had won most recent bunch of WS and had the bigger "stars." the fact that the Reds had won more games was considered meaningless.

I know people, who think the 1950s AL was stronger because of all those Yankee WS wins, and of course the Al was stronger in the 1970s because of all those A's, Yankee wins...

Of course when people started actually examining the issue, what happened when a .250 hitter moves from one league to another does his average go up or down, does a pitcher's EAR go up or down... well that 1980s AL was really really weak.

Stengel had failed to pinch-hit for Shantz, who had not thrown five innings in a game all year and who had gone as many as four innings only once before, with a chance to break Game 7 wide open in the top half of the eighth, in part because the Pirates had lefty hitters bunched at the top of the order (Virdon leading off with Skinner and Nelson 3/4).

The funny thing is that Shantz almost got a hit anyway. I think it was a ball that barely went foul, but I don't actually remember the details. (But I've got it on the DVR at home, so I can check tonight.)

Having the losing team score twice as many runs as the winning team in a postseason series is pretty crazy and very lucky.

Going into the bottom of the ninth in Game 7, the about-to-lose 2001 Diamondbacks had outscored the Yankees 35-14.

This example pops up all the time, but it's a total mirage. The Indians' 111 wins were entirely a product of an historically weak 1954 American League. They'd gone 89-21 against the hapless A's, Senators, Orioles, Tigers and Red Sox, but only 22-22 against the Yankees and White Sox.

everytime that year is mentioned you bring this up, and I don't see the evidence. Good teams historically go roughly 50/50 against other good teams.

Having the losing team score twice as many runs as the winning team in a postseason series is pretty crazy and very lucky.

Going into the bottom of the ninth in Game 7, the about-to-lose 2001 Diamondbacks had outscored the Yankees 35-14.

Others have noted this often, and it would have made for a nice NY counterpoint to 1960 if they'd held on.

That 2001 WS matchup was an interesting mix. One team (the Yanks) had a slightly better W-L record, and the other team (Arizona) had a six game Pythag advantage. But the one big difference between them (the strength and stamina of the top two starting pitchers) was precisely the most important advantage a team can have in a short series, other factors being roughly equal. Over a full season those two teams might have gone 81-81 with normal rotation schedules, but if you replicated a 7-game series with the usual postseason off days, with four days off between one series and the next, the Diamondbacks would like have won a good majority of the time.

This example pops up all the time, but it's a total mirage. The Indians' 111 wins were entirely a product of an historically weak 1954 American League. They'd gone 89-21 against the hapless A's, Senators, Orioles, Tigers and Red Sox, but only 22-22 against the Yankees and White Sox.

everytime that year is mentioned you bring this up, and I don't see the evidence. Good teams historically go 50/50 against other good teams.

Good teams do this all the time, but not teams with claims to being historically great, and not if they fail to post a winning record against any other winning team. The Indians' claim to historic greatness is built exclusively on the backs of five teams that collectively had a far greater claim to being historically awful than the Indians had to being historically great.

I'm not saying that the 54 Indians weren't a very good team, but don't embarrass yourself by comparing them to teams like the 06 Cubs, the 75 Reds, the 98 Yanks, or any one of several dozen other teams far more worthy of the title of greatness. Hell, if you look at the competition, you can make a decent case that they weren't even as good as the Indians teams of 1948 or 1995.

As someone who has studied the 1971 World Series intently, I don't see how the Pirates were particularly "lucky." Perhaps the outcome was "random" and yes, it was definitely "unexpected," but lucky? Milt May's game-winning bloop single in Game Four was the one lucky play that I remember, but it was somewhat counteracted by the big hop that Vic Davalillo's throw took in Game Six, allowing Frank Robinson to score the winning run.

On paper, the Orioles were the better team than the Pirates, but that Pirates' team was also very, very good. They had a deep bench, a deep bullpen, two big stars in Clemente and Stargell, two outstanding starters in Blass and Ellis, and a very strong supporting cast (Oliver, Hebner, Sanguillen, Briles, Giusti). They weren't the '69 Mets or the 1988 Dodgers.

Good teams do this all the time, but not teams with claims to being historically great, and not if they fail to post a winning record against any other winning team. The Indians' claim to historic greatness is built exclusively on the backs of five teams that collectively had a far greater claim to being historically awful than the Indians had to being historically great.

I'm not saying that the 54 Indians weren't a very good team, but don't embarrass yourself by comparing them to teams like the 06 Cubs, the 75 Reds, the 98 Yanks, or any one of several dozen other teams far more worthy of the title of greatness. Hell, if you look at the competition, you can make a decent case that they weren't even as good as the Indians teams of 1948 or 1995.

I wouldn't call them one of the five greatest teams of all time, but any team that wins 111 in the majors is a great team. The way you act, they are a team that in any other year in any other league they would have been lucky to win 85 games.

Jeezus, he just got done saying that the Indians were very good, but not historically great. You are way off.

In one of the '71 series games, I think it was game 4, Clemente hit a ball down the right field line, that bounced back into the field of play. The ump ruled it a foul ball! I watched this on a grainy replay show a few years ago, I dont think you could see if it hit the foul pole or what but the way it bounced back into RF it almost had to be a fair ball. Did anyone else see that. I feel bad Clemente got robbed of a post season HR; he had a very good post season batting record.

Another thing that sort of sticks is that they always talk about the Big Red Machine but in that decade both the Pirates and Reds won 6 divisional championships and 2 world series; the reds won 4 pennants I guess that is the difference. And the mid 70s Reds were probably the best team, but hell the Pirates won 2 championships too.

I wouldn't call them one of the five greatest teams of all time, but any team that wins 111 in the majors is a great team. The way you act, they are a team that in any other year in any other league they would have been lucky to win 85 games.

Well, there's plenty of room between "one of the five greatest teams of all time" and "a team that in any other year in any other league they would have been lucky to win 85 games." I doubt very seriously if that 1954 team would have won many division titles in the recent versions of the ALE, or many pennants since the 1970's, or many pennants in the NL of the 1960s. I doubt that it was as good as any of Stengel's previous 5 championship teams, when they had the Reynolds-Raschi-Lopat rotation in order, especially with Ford as an addition, and a much better lineup, particularly defensively. Those 1949-53 Yankee teams took on and beat all comers from both leagues, every year, not just five quasi-AAAA teams. I'd maybe put them in the top 100, but not in the top 50. And I definitely don't think they'd have beaten the vastly underrated 1948 Indians, a much better all-around team that had a staff ERA over a full run below the overall AL, an accomplishment that's been duplicated only once or twice in ML history.

Winning 111 games by itself is an historically great accomplishment , but that in itself doesn't automatically make them an historically great team. Teams that rate that label have to jump through hoops a lot higher than merely pummeling teams that won 51 to 69 games. That's really all that the Indians did, and as impressive as it appears on the surface, the more you look the details, the more it shrinks in size.

Checked the Shantz at-bat, and it was about what I thought. He tried to chop one over the third baseman's head, and it worked, but landed foul by about a foot. Game of inches, y'know.

Winning 111 games by itself is an historically great accomplishment , but that in itself doesn't automatically make them an historically great team. Teams that rate that label have to jump through hoops a lot higher than merely pummeling teams that won 51 to 69 games. That's really all that the Indians did, and as impressive as it appears on the surface, the more you look the details, the more it shrinks in size.

The problem is you keep dipping into this silly argument. The record of the teams in the 1954 American League was the exact same as it was in AL in 1948, in the NL in the 1960s or 1970s or any other year before 1997. Citing the paltry win totals of the Red Sox, Senators and A's tells us nothing about their quality or the quality of the league as a whole. Nothing.

There are legitimate arguments, many of which you've made, about why the Indians of 1954 aren't as good as other famed clubs. But when you keep drifting back to the W-L records of the inferior clubs of the league, you're taking a giant dump on those sound reasons.

As someone who has studied the 1971 World Series intently, I don't see how the Pirates were particularly "lucky." Perhaps the outcome was "random" and yes, it was definitely "unexpected," but lucky? Milt May's game-winning bloop single in Game Four was the one lucky play that I remember, but it was somewhat counteracted by the big hop that Vic Davalillo's throw took in Game Six, allowing Frank Robinson to score the winning run.

On paper, the Orioles were the better team than the Pirates, but that Pirates' team was also very, very good. They had a deep bench, a deep bullpen, two big stars in Clemente and Stargell, two outstanding starters in Blass and Ellis, and a very strong supporting cast (Oliver, Hebner, Sanguillen, Briles, Giusti). They weren't the '69 Mets or the 1988 Dodgers.

Blass, Briles, and Kison gave up a combined 10 hits in 33 and a third innings. None of those guys was Sandy Koufax. It was in general a disappointing performance by the Orioles hitters. Coincidently or not, about a half dozen of the Orioles had big falloffs in their batting in 1972, enough to cost them a winnable division despite no dropoff in pitching and adding Grich and Baylor. Whereas the Pirates had another strong year. Maybe the Orioles got old a week too early.

Winning 111 games by itself is an historically great accomplishment , but that in itself doesn't automatically make them an historically great team. Teams that rate that label have to jump through hoops a lot higher than merely pummeling teams that won 51 to 69 games. That's really all that the Indians did, and as impressive as it appears on the surface, the more you look the details, the more it shrinks in size.

The problem is you keep dipping into this silly argument. The record of the teams in the 1954 American League was the exact same as it was in AL in 1948, in the NL in the 1960s or 1970s or any other year before 1997. Citing the paltry win totals of the Red Sox, Senators and A's tells us nothing about their quality or the quality of the league as a whole. Nothing.

There are legitimate arguments, many of which you've made, about why the Indians of 1954 aren't as good as other famed clubs. But when you keep drifting back to the W-L records of the inferior clubs of the league, you're taking a giant dump on those sound reasons.

Except that every time I've cited that 89-21 record, I've also made reference---and in #20 above even provided individual links to the BB-Ref team pages---to the weakness of the rosters of those five teams, which is a point that seems to be ignored. Study those rosters for more than a minute or two, and note just how few superstars and lesser stars were on them, and in their primes. I could prompt you, but I'd rather you look for yourself. I was only half kidding when I applied that "quasi-AAAA" term to them, and these teams provided over 70% of the Indians' competition. For all the current complaints about "competitive imbalance" we hear from the MSM, those of us who were around back then know what real competitive imbalance looks like, and those bottom 5 AL teams of 1954 would have fit rather nicely on one of those 1954 polio stamps. (Sorry, I couldn't resist that one.)

Blass, Briles, and Kison gave up a combined 10 hits in 33 and a third innings. None of those guys was Sandy Koufax. It was in general a disappointing performance by the Orioles hitters. Coincidently or not, about a half dozen of the Orioles had big falloffs in their batting in 1972, enough to cost them a winnable division despite no dropoff in pitching and adding Grich and Baylor. Whereas the Pirates had another strong year. Maybe the Orioles got old a week too early.

After the Orioles swept the first 2 games of that Series, I got an incredibly stupid friend of mine in Berkeley to double down on his bet on the Pirates to win the whole thing. Or at least it seemed at the time that he was stupid---I never should have had that anticipatory dream about coolstandings.com.

Except that every time I've cited that 89-21 record, I've also made reference---and in #20 above even provided individual links to the BB-Ref team pages---to the weakness of the rosters of those five teams, which is a point that seems to be ignored.

And pointing out the weaknesses of the rosters is fine. But you seem incapable of understanding that their records tell us nothing, otherwise you'd realize the folly in noting them. The AL teams in 1954 had a .500 record. Since the Indians won so many games, the wins had to come from somewhere. When you stop writing nonsense like "teams that rate that label have to jump through hoops a lot higher than merely pummeling teams that won 51 to 69 games," you won't find as much opposition to your position.

Blass, Briles, and Kison gave up a combined 10 hits in 33 and a third innings. None of those guys was Sandy Koufax.

I'll just throw this in the face of the sample size: the three combined were 13-3 in postseason play, with ERAs ranging from 1.98 to 3.10.

Edit: I misread something, they were a combined 10-3.

#28 It wasn't so much avoiding lefties in Forbes as wanting them for Yankee Stadium.

As I've mentioned before it wasn't so much his handling of Ford (normal by Stengel's standards -- he never used Ford 1/4/7) as the fact that he was always going to use Ditmar twice and Ditmar was terrible (getting only 1.2 innings total in his two starts).

Yeah, he lost his job over the way he used Ford (at least that was probably the largest single factor), but Ditmar had pitched at least as well as Ford in the 1959-60 period (and Stengel was the ultimate what have you done for me lately kind of guy). It was scapegoating to a very large extent.

1971 was the first season I purchased Strat-O-Matic Baseball. (Man...how good was Aaron that year !)

I replayed that world series half a dozen times and it seemed that the Pirates always beat the Orioles. I learned at 11 years old that win totals on that card didn't mean squat. The Orioles starters were very beatable. Their 3.40 FIP was only 18th best in a 24 team MLB. The Pirates at 3.33 ranked 15th.

Meanwhile the Pirates had a 113 wRC+ and the Orioles 112 wRC+

So with a slightly better team FIP and wRC+, the Pirates would probably not be considered an underdog by today's standards, at least not before a league adjustment.

(I can't open the article from China....it's blocked...sorry if any of this was covered in the article)

Except that every time I've cited that 89-21 record, I've also made reference---and in #20 above even provided individual links to the BB-Ref team pages---to the weakness of the rosters of those five teams, which is a point that seems to be ignored.

And pointing out the weaknesses of the rosters is fine. But you seem incapable of understanding that their records tell us nothing, otherwise you'd realize the folly in noting them. The AL teams in 1954 had a .500 record. Since the Indians won so many games, the wins had to come from somewhere.

I suppose you could call it a chicken vs. egg situation, but I've pointed out the weakness of those rosters every time this topic has come up. If I'd neglected to do that; if I'd based my case solely on the W-L records; then, but only then, would your mathematical point be compelling.

But the reason that the Indians went 89-21 against those five teams was precisely because the rosters of those five teams were so weak. The Indians' terrific pitching staff enabled them to take advantage of the situation, but the fact that those same pitchers couldn't do any better against their only real competition** gives us a pretty good reason to conclude that the significance of those 89 wins isn't all that it's cracked up to be.

**Again, against the only teams with winning records, the Indians either went 22-26 or 30-39, depending on whether you want to throw out 21 Spring Training games against the Giants. But even without those ST games, a 22-26 record against the only winning teams you've played isn't exactly symptomatic of an "historically great" team. Or even a "great" one.

Isnt it true that in complete isolation (w/ no world series and no interleague play) It would be completely impossible to estimate how strong any team is? Other than I suppose using a radar gun to measure pitchers and something similar for batters.Something we don't have.

Thus I would have to agree that using w/l record in intra league play is not helpful. Jolly St Nick (Andy) basic contention then has to do with the distribution of those wins, wins bunched up against cellar dweller teams. I guess that is a legitimate pt. but if the Cleveland Indians were somehow uniquely able to take better advantage of cellar dwellers than say the Yankees, why should that be used against them ? Perhaps great teams are better able to take massive advantage of cellar dwellers. And if so then they are better than the Yankees for that year.

Also if you're going to make this win distribution contention, then don't you have to do the same for the 75 Reds or the 52 Yanks or 39 Yanks or whomever? Until we see that then you're not even making comparables.

I can sort of see your pt. Andy, but then again, I think it is not at all the only way to look at it. Taking advantage of cellar dwellers may be evidence of greatness. Going .500 vs the Yanks in small sample size maybe random fluctuation.

I'd leave out the spring training games at least until we see who was starting and also pitcher usage. If only to focus on the crux of the problem which I think is w/l distribution thing for now. Not saying you dont have a pt. just might be better to work out one issue first.

Also, do games (such as World series, all star) played year previous or year after make any difference in all this?

THis then leaves this puzzling quote:

And pointing out the weaknesses of the rosters is fine

How is it "fine?" If someone batted .350 or threw an ERA of 2.4 in totally isolated fish pond, then it has no more relevance than the w/l record. Yes? So it is hard to see what part of the roster are you using? All star appearance, Hall of Famers on the roster? Batting average? What?

Dont any of these methods suffer from the same problem as w/l records?

In defense of the Indians, the Red Sox that season were 69-85 - 2-20 vs. the Indians and 67-65 vs. everyone else.

How is it "fine?" If someone batted .350 or threw an ERA of 2.4 in totally isolated fish pond, then it has no more relevance than the w/l record. Yes? So it is hard to see what part of the roster are you using? All star appearance, Hall of Famers on the roster? Batting average? What?

It's fine because you can use other data to help know what the general skill level of the players populating those rosters (using other seasons, switching leagues, etc.) But yes, you do run into the same problem if you simply try to identify that roster's strength through 1954 numbers alone.

Okay I decided to run some numbers to see how "overrated" those Indians were (I left out the 98 Yanks only because the interleague play thing meant a little extra back/forth)

team/year....winning perc. vs lower half of league.....win per. vs two closest teams

1927 Yanks........ .798 .636
1932 Yanks........ .762 .568
1939 Yanks........ .738 .550
1961 Yanks........ .715 .528
1975 Reds......... .736 .462

So there's five great teams. Just a couple of observations.

1. Even the great Red Machine, that beloved team had a sub .500 record vs it's two closest rivals Dodgers/Pirates. For once the Pirates actually managed to stay .500 with them (the Reds gave them fits).

2. Obviously one could make adjustments for just how bad the bottom division is, these calculations were done quickly and in my head, but IIRC the 1975 NL was the most even distribution of w/l.

3. Statistically, Andy's argument hardly seems very significant. The 54 Indians went .809 vs the bottom 5, but the 27 Yanks are virtually there and a couple other teams not far off. is this statistically significant? There's only one team that seems to stand out via Andy's reasoning: the 27 Yanks and .636 vs close rivals; but it's still a chicken or egg question: if the 1927 AL is top heavy in Yankee talent, then we still dont know if this makes them better than the Indians. Aside from that you've got 4/5 teams whose winning profile doesnt look all that different from Clevelands.

4 Why take the bottom FIVE teams vs Indians, wouldnt make more sense to take the bottom 4? I think you cherry picked that a little.

5. It is not that statistically out of bounds to be .500 vs your close rivals. We already have the 75 Reds, out of 5 sample teamas. But The 39 Yanks actually went 8-11 vs Boston; and the 32 Yanks were .500 vs Washington. I could cherry pick those games out as well and make the same argument agains these teams as against the Indians.

6. it really all comes down to a matter of taste. The bottom line is how many games you win VERSUS THE LEAGUE; it makes not difference if you win against bottom feeders to close rivals. To say that the Indians only went .500 vs close rivals, that's just your preference, it obviously didnt keep them from piling up a huge number of wins vs the league in toto.

Looking at the overall seasonal record of the teams has its limits in what it can tell you about which is the best team in that specific short series.. The team in October may be better or worse than what that overall season record tells you. A team's success fluctuates throughout the season. This suggests that the team's quality does so, too. A team may be better in April or June or August than it was in May, July or September. It may have peaked and gone beyond that or it may just be peaking now. You may be looking at that team at its crest or at the trough. That great team at such and such a time in the season, or that great team per the indications of its overall seasonal record, may not be the team that will be playing at the time of that World Series. (Not to mention that the regular season and the World Series are played differently--you mostly only need your best and better players.)

Teams like players have stretches where they are better are worse than their rates for the year. A team with a great record for the season may at a given moment not be as good, or as bad, as they are at another time.

Jolly St Nick (Andy) basic contention then has to do with the distribution of those wins, wins bunched up against cellar dweller teams. I guess that is a legitimate pt. but if the Cleveland Indians were somehow uniquely able to take better advantage of cellar dwellers than say the Yankees, why should that be used against them ? Perhaps great teams are better able to take massive advantage of cellar dwellers. And if so then they are better than the Yankees for that year.

I'm getting the feeling that this is getting to be an argument about whether or not the 1954 Indians were even a very good team, or not. To the extent that my rhetoric has added to the confusion, I apologize, but let me summarize exactly what I'm saying.

---The 1954 Indians were a very good team. They demonstrated this by winning 111 out of 158 games. That is a "great" accomplishment in itself, regardless of the splits against winning or losing teams.

---OTOH they were not an "historically great" team, because to say that, then you have to place the bar much higher. You have to ask whether or not they consistently beat other "very good" teams, and if they didn't (which the Indians obviously didn't, over a "sample size" of 48 games), then you have to note that other teams with claims to "historical greatness" were able to do so, often quite emphatically.

---To establish "historic greatness" for the 1954 Indians, you have to consider many factors: Their W-L record (111-43), their Pythagorean record (104-50, which in itself brings them down a notch), the level of the competition, the relative strength of the leagues at the time**, the postseason record, the overall makeup of the roster (# of superstars or other HoFers/HoVGers in their primes vs. the # journeymen of roster fillers), etc.

Unfortunately, there's no one magic bullet formula that can incorporate all these factors and spit out a "Certificate of Authentic Greatness", with a "Mint/PSA-10" number attached. There's bound to be a certain amount of subjectivity that goes into the assessment. But when you consider ALL the factors, rather than just those 111 wins stripped out of context, I can't see any way to call that team "historically great" in any meaningful sense of the word.

"Historic accomplishment", yes. "Historically great", no.

**Regardless of the lack of any precise metric, I think it's pretty safe to say that the 1954 American League was significantly weaker (and far more topheavy) than the 1954 National League. To deny that would mean that you'd have to throw out or dismiss such things as the greater number of HoFers in their primes, the number of emerging black stars, the roster makeups of the bottom 5/8 of the AL, and so on.

I'd leave out the spring training games at least until we see who was starting and also pitcher usage.

That's a fair point, and if this topic ever comes up again, I'll dig out my 1954 Sporting News copies (I've got em all) and answer that question. Unfortunately, there are about 8 or 9 surrounding years on top of them in the back of an overcrowded closet, behind another similar stack and draped by hanging clothes, and so that bit of mega-research is for another day. I suppose that one could use ProQuest's Paper of Record files to do the research, but that's even more time consuming than closet digging. But given the veteran makeup of the Indians' rotation, I'd be very surprised if the Indians' starters in Arizona that year weren't pretty much the same ones they used throughout the regular season.

Morty (#53) - Good points. I weigh a team's late season roster makeup disproportionately, and this would be reflected in a team's postseason record.

team/year....winning perc. vs lower half of league.....win per. vs two closest teams

1927 Yanks .798 .636
1932 Yanks .762 .568
1939 Yanks .738 .550
1961 Yanks .715 .528
1975 Reds .736 .462

Without getting into the details, I think that as a general rule, the explosion of available talent in recent decades has resulted in a wider distribution of that talent throughout the game. This is reflected in the overpreponderance of "historically dreadful" teams in the years prior to 1980, when the influx of Latinos and other "foreigners" really took off. To use a variant of Gould's observation about the leveling of individual accomplishments relative to one's peers, I'd say that it's also a lot harder for modern teams to dominate today than it was when the talent pool was restricted, the draft didn't exist, and the rich teams were willing and able to pay for more scouting and more farm teams than their brethren on the bottom. All of which makes the accomplishments of the more recent teams like the 98 Yankees even more impressive, but it also partly explains why in discussions of "greatest" teams, you get so many pre-1980 teams listed.

Just to make the comparison more fair, let's run the same method (bottom half, not bottom 5) and toss in a couple of Connie Mack teams:

1954 Indians........ .776 .500
1931 Phil........... .754 .500
1929 Phil.......... .776 .651

Put the whole thing together:



1927 Yanks........ .798 .636
1932 Yanks........ .762 .568
1939 Yanks........ .738 .550
1961 Yanks........ .715 .528
1975 Reds......... .736 .462
1954 Indians...... .776 .500
1931 Phil.......... .754 .500
1929 Phil.......... .776 .651

One note: Those 1929 and 1931 AL are actually quite well distributed for talent, the bottom 4 teams in 1931 went .395 overall and the bottom 4 in 1929 went .422. That's very good as far as I can recall, makes the Al Simmons teams look quite good.

Now look at all 8 teams. Do the 54 Indians really stand out to you as some sort of abomination? Hell Im not a statistician but I'd wager you that those Indian numbers are well within the random fluctuations of that group. Do you agree or not? They are like a notch above average vs bottom dwellers and they are pretty much right about average vs close rivals.

Do you see it that way or not?

This is kinda fun anyone want to run this for NL teams?

Willie Mays made the point in this 1966 autobiography that after playing the Indians so much in spring training the Giants were not particularly impressed with them and were very confident going into the World Series.

I'm getting the feeling that this is getting to be an argument about whether or not the 1954 Indians were even a very good team, or not.

I hope nothing I said. My very first post in this thread defended you specifically on this basis.

I think there is a serious issue with respect to how can we define very good versus great when we are dealing with two talent pools that live in virtual isolation. Only the all star games and the world series are the only in season competition prior to inter league play. This makes it very difficult.

The spring training games brings up another problem. That is to be fair you would really have to find spring training games for all the great teams in question. The 1931 Athletics vs the Gas House gang. Are we really prepared to do that? I would expect the results to vary greatly but have at it...

I can certainly concede that overall NL talent could very well be better in 1954. However this doesnt obviate the problems with teams. The Oakland As seemed very tough teams back in the 70s even thought I was quite convinced the NL had better overall talent.

Another very serious issue is random fluctuation which I do not think has been addressed yet. As a base starting pt. My guess it is about 5% plus/minus. Which is to say that any statistical measure of a team can only predict it's w/l record to within +/- 5%. So far I have seen this in two ways, expected winning percentage for teams in one run ball games, good teams consistently fall off 5% from expected in one run games (vice versa with bad teams). Also starting to use OPS+ and ERA+ and if you find a team with say 20 pts above average in OPS+ and ERA+, their wins can vary from 100-110. So there is a lot of noise. WHich means this statement:

for the 1954 Indians, you have to consider many factors: Their W-L record (111-43), their Pythagorean record (104-50, which in itself brings them down a notch)

I have to disagree, this 7 wins is still within random fluctuation or statistical noise as far as I can tell. Again I am not a statistician and have not really crunched much baseball data but just a starting pt.

Do you have any more evidence or theory of how much random fluctuation we expect in baseball? My guess is 5% but jut a guess.

Without getting into the details, I think that as a general rule, the explosion of available talent in recent decades has resulted in a wider distribution of that talent throughout the game.

I think it all depends. See my note in post 56. There are very fairly well distributed leagues in the old days and of course, there was expansion in 1961. Those seasons seem quite mis distributed. Can't recall 1969.

I think my general recollection is sort of what you are saying, in the days of Honus Wagner and Ty Cobb I do recall a lot of mis distributed w/l records. But going from memory.

Your pt about the 1998 Yankees merely begs the question doesnt it? Are they great or is talent mis distributed. The use of free agency and a large bankroll would creates a significant counter to all this. Yes?

Sunday Silence,

All good questions that I'll take up at some point later in the day. Have to pick up Mrs. JOSN at BWI and have to cut it short for now.

Here are winning perc. for the bottom half of teams/leagues for seasons we are dicussing:


1929 .422
1931 .395
1927 .399
1939 .399
1954 .388
1961 .408
1975 .446
1998 .444

Just for good measure I did the same method (bottom half/close rivals) for 1906/1907 Cubs:


1906 Chic..... .722 .758
1907 Chic..... .636 .755

Yeah, the old Cubs teams really beat the hell out of their close rivals: Pirates both years, and NYG in 06 and Phil in '07. Outstanding.

Winning perct bottom half of league:

1906 .383
1907 .400

So I guess a little unbalanced not as bad as I thought.

Have to pick up Mrs. JOSN at BWI and have to cut it short for now.

Good luck; I have only a few rules I live by in my life (learned from grandpa) and one of them is:

"Never pick up or deliver relatives to the airport while stoned."

Grandpa knew what he was he talkin about.

But the reason that the Indians went 89-21 against those five teams was precisely because the rosters of those five teams were so weak. The Indians' terrific pitching staff enabled them to take advantage of the situation, but the fact that those same pitchers couldn't do any better against their only real competition** gives us a pretty good reason to conclude that the significance of those 89 wins isn't all that it's cracked up to be.

I don't really care about their record against those teams and equally don't care about their record against the two good teams, It doesn't matter at all. The Phillies were a better team than the Cardinals this year and posted a 6-10 record against them. Things like that happen. Argue that the 1954 league was weak and you have and given reasons. Although I don't know if you are arguing that just 1954 was historically weak(which would be a moronic argument) or that a spread of 5 years or so centered around 1954 was historically weak(which could be fathomable.)

But the Indians record against a couple of teams is not an indication of anything.

edit:Great posts Sunday silence

Teams like players have stretches where they are better are worse than their rates for the year. A team with a great record for the season may at a given moment not be as good, or as bad, as they are at another time.

That is true however problem with the Indians is that they were playing .800 ball or almost for the last 3 months of season. Giants were playing well too, but only .630/.600 in Aug/Sept. So the issue of seasonal fluctuation still doesnt really help to resolve the 54 Indians specifically.

Well, not really. The Indians in '54 played at very high level throughout the season (except for the truncated April), which you would expect of a team that wins 111 out of 154 games. Nevertheless, they still played better at some times than at other times, if W-L means anything. There are still month to month differences. The team's winning percentage for September was .696, which is high, but it ain't the .813 of August.

Even if they would have been winning in the last three months, this doesn't defeat my overall general point, which was not meant to just apply to the '54 Indians. It was that teams have streaks, just like players--probably because players have streaks, teams have streaks.

Yes I agree, I didnt really respond to your overall pt. I was more thinking about the upset of Giants over Indians and the issue of whether the Indians were really any better; or whether it was a statistical illusion. Which I thought might be related to what you were saying about streakiness. If anything the Indians were playing about 100 pts better than the Giants in Sept. So basically I wouldnt ascribe the upset to streakiness.

But you werent saying that, so OK.

On the larger pt. Yeah I see that, however then we would really have to break this stuff down on a week by week or series by series breakdown and at that pt. I think the random noise would just over whelm it.

However as you suggest, I think it could very well have been that the 54 Indians caught the Yankees when the Yanks were on a hot streak and so didnt do so well. Or perhaps the 39 Yankees caught the Red Sox when the Yankees were cold (see post 52). So yeah I definitely think that could be at work.

I was thinking another reason could have been that Al LOpez could have simply been saving Feller et al vs a series with Boston and thrown his lesser guys against the Yankees. On the reasoning that he might still be 50-50 vs Yanks but might .800 w/ Feller vs a cellar dweller. It might just be a better way to use talent. I thin the Indians pitching staff was so good though that this probably not what happened.

I think it would make more sense to look at the overall record against .500+ teams than to simply use the next two closest teams. Bigger sample size in almost all cases.

It's probably not a coincidence that 71 of the 76 teams with a .600+ WPCT against .500+ teams were in the post-season. The exceptions: 1951 Dodgers (#3), 1945 Cardinals (#14), 1930 Senators (#39), 2010 Cardinals (#42), 1994 Royals (#45; questionable due to the strike), 1989 Royals (#55).

Here's the rank of those teams from the list, with a few extras:

CHC 1906 #1
SEA 2001 #4
NYY 1932 #13
BAL 1969 #18
NYY 1927 #25

NYY 1939 #30
PHA 1929 #56
PHA 1931 #58
NYY 1998 #85
NYY 1961 #156

CIN 1975 #205

CLE 1954 #606

Now that #606 is still around the 70th percentile in terms of road WPCT against winning teams: the average for this over the last 111 years is .437. You need to get the overall performance context. Great teams usually perform extremely well in this area--upwards of the 90th percentile. The 54 Indians were down in the 70th percentile.

On the opposite side of the coin, those 89 wins against <.500 teams that the '54 Indians racked up is the highest total in baseball history. (That same year, the Yankees were 77-33 in such games, and the White Sox were 76-34. It was a great year in the AL for beating up on doormats, and I'd suggest that this boosts Andy's premise about the league strength.) The '06 Cubs were second (85-24). The Indians had the sixth highest WPCT in baseball history against <.500 teams (.806). Six of the seven games they gained over their PWP came from the extra games they won against these teams.

I think it would help a premise about DISPARITY in league talent, as opposed to total talent or whatever is meant by term "League strength."

For this context, total talent is hard to measure: you can't really do it with WAR or Win Shares. Concentration of talent might come from the percentage of wins in the top three teams, the way you were doing before, vs. total league wins. Disparity of talent might be arrived at by summing B-R's SRS value for the top three teams, then subtracting from it the SRS value for the bottom five teams (we're working with pre-expansion only at this point: it gets more complicated in later incarnations). Call that the "SRS differential."

So, concentration of talent and disparity (dispersion) of talent, AL 1952-55:

Year, Wins, % W, SRS diff
1952, 266, 43.2, 4.1
1953, 280, 45.5, 6.0
1954, 308, 50.0, 7.7
1955, 280, 45.5, 6.1

Knowing the deviation in number of games played against <.500 teams might be useful, too. For example, in AL 1954 62.5% of all games were played against <.500 teams; that percentage was only 37.5% in AL 1955.

well you've certainly made Andy's point in dramatic fashion.

606th! My head splode.

Can you apply this same reasoning to say Bob Feller and claim he wasnt so good?

I think there is a serious issue with respect to how can we define very good versus great when we are dealing with two talent pools that live in virtual isolation. Only the all star games and the world series are the only in season competition prior to inter league play. This makes it very difficult.

I guess to put it in HoF terms, I'm a "Small Hall" guy when it comes to teams. I doubt if more than about 25 or 30 teams at most would qualify for "historically great"**, and I doubt if more than 4 or 5 of those could really be considered for #1. To meet my standard, a "great" team has to demonstrate several levels of greatness:

---Dominance of its own league, as represented by W-L and Pythagorean records. A .667 WP would be a good starting point, but I wouldn't require it in all circumstances. I'd look to see how soon their pennant (or division) was clinched, and what their record was then, rather than focus on the garbage time after that. And the overall league strength would definitely be another factor.

---At the very least, an overall winning head-to-head record against whatever other very good teams they competed against, in either league, counting both regular season and postseason.*** I say "overall" to enlarge the sample size, and I wouldn't require a winning record against every other team, but a team that did accomplish that (like the 2001 Mariners, at least in the regular season) would certainly get an extra point or two for consistency.

---A core of superstar and star players in their primes, as opposed to a lot of one year wonders having a coincident series of career years. Certainly as a tiebreaker, a one year wonder team would lose points compared to other great teams with roughly comparable surrounding years, or at least surrounding years where they were in solid contention.

---A reasonably strong bench and bullpen to complement their everyday lineup and starting rotation.

---Postseason success. Can be overridden in the case of teams like the 06 Cubs or the 01 Mariners, which were truly dominant from start to finish in the regular season, but when you're looking for the best of the best, you really need for a team to run the table in October to be considered (IMO) for #1.

---And then as another tiebreaker, I'd look at the rosters of all teams in the league from the POV of demonstrated career talent. All things being equal, a team stocked with current stars (in their primes) and future HoFers rates over a team without them, and a league with such players also provides a much better test of greatness than a league without them.

**And no, the 1954 Indians wouldn't be among those 25 or 30 teams. As I said above, I seriously doubt that they were even better than their own 1948 or 1995 teams, whatever the W-L record might imply.

*** For the life of me I can't see how anyone would ever throw out the postseason altogether in assessing a season, unless key injuries were a deciding factor in a loss.

Postscript to #71: While there would be plenty of numbers that would enter into all of those factors of "historic greatness", I don't believe for a second that any final ranking can be reduced to a number, or be anything other than somewhat subjective. I'd find it laughable to put the 1954 Indians on the same page as the 1975 Reds or the 1998 Yankees, but choosing between those last two teams would be a much harder task. What we need to settle that dispute is a reliable time machine.

So what I’m hearing is that the 1954 Indians are like baseball’s equivalent of the Bud Grant Viking? E.g the Vikes always come to the Superbowl with a really good record and then they play like they shouldn’t even have been invited to this game…

How is that possible? Do they have some singular ability to feast on lower pond scum at a 90% clip whilst the Dallas Cowboys of the world have to resign themselves to a mere 65% clip? Did it have to do with not wearing gloves in the cold?

These questions and more….

I mean I dont even know if Don's data helps Andy or hurts him. After all, it was Andy who said:

"No question about it folks. "No sirreeer! ". There is no way in hell you can tell me that the Indians weren't at least very good. I swear on a stack a bibles that this team was very good.....



And I believed you Andy. I fell for it.

I feel so used.

These Indians, these freakin Redmen that you Andy St Nick, said were very good and they're what hell that guy say.............? 606th.

Are you sh1tt'ng me? 606th?? I mean what hell that? That's like the 1974 Red Birds. Or say, I dunno Wally Berger and the 36 Braves would be most similar to the 1954 Cleveland Indians.

You talk some sense into the boy, ANdy. I give up....

I'd find it laughable to put the 1954 Indians on the same page as the 1975 Reds

well I mean "YEAH". I hear ya on this on. Freakin Big Red Machine went .466 vs the Pirates and Dodgers.

Hell they can't even play .500 against the best teams. At least the GOD"AM EM EFFIN CLEVELAND INDIANS CAN PLAY .500 ball against the best teams.

And I know they ain't even sh!t.

"No you'r forgettting the post season play. Those games have to count. You have to take that into consideration, that's what your forgetting."

oh you want to throw that one in? You really do? So what? Here:

1975 Reds..... .487 vs Pirates, Dodgers and Bosox.

That's the Big Red Machine for you. They still haven't broken .500 and I've given 'em two tries.

"Yeah but they had a sustained period of excellence, SS. You really have to count the two years, 1975-76."

You really....

"...because you need a larger data sample set. This will increase the standard deviation thereby minimizing the noise to obtain more accurate levels of measuring..."

You really wanna go down this road?

---A core of superstar and star players in their primes, as opposed to a lot of one year wonders having a coincident series of career years

Agreed So get this party started and toss off the '69 Mets straight out. I mean I never thought they were that good anyway. And let's face it Swoboda, Cleon Jones, Boswell and Grote never had seasons like that again. And Agee had two more seasons.

That leaves us with what Bud Harrelson and Kranepool those are your year in and year sustained, whatevere guys.

So the 69 Mets obviously: OUT!

Thanks for your support on this one, Andy.

Here's the rank of those teams from the list, with a few extras:

CHC 1906 #1
SEA 2001 #4
NYY 1932 #13
BAL 1969 #18
NYY 1927 #25

This brings up an interesting pt.

To me clearly, the 27 Yanks are in the discussion. They seem to me to be prototypical candidate for best team or one of the best teams. Now if your ranking here is of any value whatsoever, any team that can out perform the 27 Yanks on a valid reliable criterion for greatness must also be a candidate. yes?

So who are the other 20 candidates that you haven't told us about.

I wait with baited breath.

SS, I'm not sure where the last part of your #72 was coming from, since if you'll read what I wrote in #71, I listed six factors, not just one, that I take into consideration in thinking about "historically great" teams. That #606 ranking for the '54 Indians' record against .500+ teams that Don mentions is both interesting and amusing, but it's only one piece of the puzzle. It helps knock them off any pretensions to be considered "historically great", but it obviously doesn't mean that there were 605 teams better than they were.

Ironically, there is a certain similarity between the 50's Indians and the 70's Vikings, though not one that's all that precise: They were both very good teams with key flaws that kept them from reaching the top. They also had the misfortune to be in an era (1947-56; 1969-78) where the competition above them (the Yankees in one case, the top AFC teams in the other case) were simply better than they were. If you want a one year comparison, you might compare the '54 Indians in particular to the '69 Vikings, since in both cases the thought that their final loss was a "big upset" was largely due to an underestimation of the opponent (and its league) and a corresponding overevaluation of the team with the gaudy prior record.

As for #74, I'd suggest you try looking again at all the factors I listed in #71, rather than simply one of them. You do this in #75 as a way of tossing out the '69 Mets from consideration, and I'd agree with you there. That was a very good team for one year, but looking at what happened afterwards, it's hard to think of them as "great" by comparison to the other teams under consideration.

That #606 ranking for the '54 Indians' record against .500+ teams that Don mentions is both interesting and amusing, but it's only one piece of the puzzle. It helps knock them off any pretensions to be considered "historically great",

Well look you've already said it's a valid measuring criterion. If not, it wouldn't knock the Indians off anything.

If it's valid and the Indians are 606th, then clearly your def'n of "Very good" is very expansive. Right?

ANd if w/l record vs +500 teams is valid, then so is w/l record vs top three teams. I mean both of these measure play against other top teams or very good teams or pretty good teams or something. I dont know which one is more valid, but I do know if I buy into the reasoning that record against good teams is important, then the record vs the top three teams must also be important. Right?

So Indians: .500 Big Red Machine .486. Indians Very Good; Reds something less than.

What am I missing?

For the life of me I can't see how anyone would ever throw out the postseason altogether in assessing a season, unless key injuries were a deciding factor in a loss.

OK that seems fair. What about regular season when someone gets hurt. That's every bit as devastating right? And regular season counts for something. SO what about the Dogers w/o Koufax? And the Tigers w/o MIckey Cochrane? Yankees w/o Munson? All those teams get some sort of boost as well?

Postseason success. Can be overridden in the case of teams like the 06 Cubs or the 01 Mariners, which were truly dominant from start to finish in the regular season...

Now I confused. Didnt you just say we can't throw out data pts? But now we can? We can throw out data pts. if the rest of the season was "truly dominant.?"

E.g. the Indians 1954 season? Was that truly dominant?

What about the 2004 Cards? They played .605 vs +500 teams (46-30) and we can toss out the world series sweep?

The 04 Cards played .648 ball. Are they an all time great team?

That #606 ranking for the '54 Indians' record against .500+ teams that Don mentions is both interesting and amusing, but it's only one piece of the puzzle. It helps knock them off any pretensions to be considered "historically great", but it obviously doesn't mean that there were 605 teams better than they were.

Well look you've already said it's a valid measuring criterion. If not, it wouldn't knock the Indians off anything.

Again, it's one factor out of six, not the sole determinant---there is no sole determinant. And it certainly doesn't mean that they'd drop down to #606. What you need to do to make a coherent ranking (IMO, anyway) is to consider ALL of the factors that I listed in #71, weigh them accordingly, and come up with some sort of a ranking.

And even with all that, any comparison between (say) the 1927 Yankees and the 1998 Yankees is going to be inherently subjective, because some people are going to weigh relative dominance more heavily, while others (like myself) are going to give much more weight to the overall competitive nature of the game (meaning the size and the quality of the talent pool) within the two teams' respective eras. Some people are going to factor in postseason performances more than others. Some people are going to look at one year wonder teams with more of a skeptical eye than others, as you did with the '69 Mets. And so on.

None of these preferences is necessarily "right" or "wrong", but when you think about all the possible permutations among six (or more) factors, you can see why there's never going to be a consensus about which team is # 1.

For the life of me I can't see how anyone would ever throw out the postseason altogether in assessing a season, unless key injuries were a deciding factor in a loss.

OK that seems fair. What about regular season when someone gets hurt. That's every bit as devastating right? And regular season counts for something. SO what about the Dogers w/o Koufax? And the Tigers w/o MIckey Cochrane? Yankees w/o Munson? All those teams get some sort of boost as well?

I can think of plenty of seven game series where the absence of a key player could cause a team to lose. I can't think of any team that went from "historically great" to merely "very good" due to the loss of any player for a few months of the regular season. There's no one-size-fits-all answer to this, but I don't think that (for instance) the 1925 Yankees get elevated to "great" simply by imagining what they might have been like with a healthy Babe Ruth, whereas if Ruth had missed an entire World Series that the Yanks lost in 7 games, I would take that sort of injury into consideration before docking them for not winning it.

But OTOH, while I wouldn't call the '72 A's an all-time great team, the fact that they still won the World Series without Reggie Jackson and Darold Knowles shows that they have to be given credit for a fair amount of depth and resilience. Not that you can assign a number to something like that, but it's still a point worth mentioning in any discussion of that team.

Postseason success. Can be overridden in the case of teams like the 06 Cubs or the 01 Mariners, which were truly dominant from start to finish in the regular season...

Now I'm confused. Didnt you just say we can't throw out data pts? But now we can? We can throw out data pts. if the rest of the season was "truly dominant.?"

E.g. the Indians 1954 season? Was that truly dominant?

In terms of regular season overall W-L records, obviously. By other measurements, not at all.

What about the 2004 Cards? They played .605 vs +500 teams (46-30) and we can toss out the world series sweep?

The 04 Cards played .648 ball. Are they an all time great team?

By certain "Big Hall" standards, they were indeed a "great" team, meaning that they might have been among the top 100 or so. And I'm pretty sure that they wouldn't have had too much trouble winning the majority of games against the '54 Indians over the course of a season.

But "all time great"? Not a chance, not after being swept in the World Series. Not when there were plenty of other teams with better than .648 records that also won the World Series. Again, you can't just focus on one particular metric. You have to look at the whole package, and unless you're talking about exceptions-that-prove-the-rule teams like the '06 Cubs or the '01 Mariners, I can't see any team that doesn't finish the job being considered for a top 10 ranking when you consider the nature of the competition.

and unless you're talking about exceptions-that-prove-the-rule teams like the '06 Cubs or the '01 Mariners,

are you coming up with a new category? You had said that you can throw out the post season if the team was dominant. The 04 Cards played .648 overall and .605 vs +500 teams. Is this not dominant for you? If not what standard?

You cant just pick and choose who you want for exceptions.

and unless you're talking about exceptions-that-prove-the-rule teams like the '06 Cubs or the '01 Mariners,

are you coming up with a new category? You had said that you can throw out the post season if the team was dominant. The 04 Cards played .648 overall and .605 vs +500 teams. Is this not dominant for you? If not what standard?

You cant just pick and choose who you want for exceptions.

The '06 Cubs played .763 overall and .721 against the other two .500+ teams in their league---coincidentally the exact percentage that the '54 Indians compiled against the entire American League, including the Five Dreadfuls.

The '01 Mariners played .716 overall and .676 against the other 8 winning teams they played in the regular season: .677 against the 5 AL winners and .667 against the 3 NL winners. And AFAIK they remain the only team in the interleague era to post a winning record against every team they played.

So yes, I'd feel justified in saying that those two teams deserve a postseason failure pass more than the 2004 Cardinals, who were indeed a dominant team in the regular season, but not on the historic level of those other two. The '06 Cubs and the '01 Mariners are the only two non-championship teams I think would qualify for an all-time top 10 list, and if you don't take either postseason or era into account, I don't see how any team could match the '06 Cubs.

I'm massively still at a loss understanding the messed up thinking involved here. If a team has a very good record overall, and a good record against other good teams, which means they have just a good record against the poor teams.. how is that an indication of their quality?

I mean if a team doesn't beat up massively on the crappy teams...then are they truly great? Ultimately the only gauge of a teams quality is their w/l record with run differential and any other normal adjustments made.... but I just can't fathom claiming one 105 win team is clearly more dominant than another 105 win team because of the fact that the one team beat up good teams, while doing just pretty good against the poor teams vs the other team held their own against good teams while dominating the poor teams.... it's a massively silly ####### argument. Ultimately you argue the teams record, their expected record and the strength of their league...their record against a select pool of teams shouldn't ####### matter.

I'm massively still at a loss understanding the messed up thinking involved here. If a team has a very good record overall, and a good record against other good teams, which means they have just a good record against the poor teams.. how is that an indication of their quality?

I mean if a team doesn't beat up massively on the crappy teams...then are they truly great? Ultimately the only gauge of a teams quality is their w/l record with run differential and any other normal adjustments made.... but I just can't fathom claiming one 105 win team is clearly more dominant than another 105 win team because of the fact that the one team beat up good teams, while doing just pretty good against the poor teams vs the other team held their own against good teams while dominating the poor teams.... it's a massively silly ####### argument. Ultimately you argue the teams record, their expected record and the strength of their league...their record against a select pool of teams shouldn't ####### matter.

What all this amounts to is that you want to reduce the determination of "best" (or "historically dominant") teams to a single metric (the regular season W-L record) with perhaps a few minor tweeks (run differential and "normal adjustments", whatever that means), all crunched into a computer and spit out on a spreadsheet. That may not be literally what you're saying, but it's the clear implication.

Not allowed: Adjustment for league strength; any examination of the makeup of the rosters of the teams and their competitors; postseason results, no matter how good or how bad; any sort of breakdown of how the team performed against the better competition during the regular season. In a word, little or no adjustment for the context in which those gaudy W-L records were achieved, or how they were achieved. It's as if you're filtering out any information that can't be programmed into a spreadsheet, or that might serve to call some of those W-L records into question.

Here's the difference: I'll make my rankings by considering as many factors as possible, as I outlined in # 71. You seem to think there's something ###### up with that, because some of those factors can't be reduced to a number. To that I plead guilty, because ultimately I don't think that a million computer sims are ever going to be able to prove whether the 1906 Cubs could beat the 1954 Indians**, or whether the 1975 Reds could beat the 1999 Braves. And that's because, to be blunt, it's just too damn complicated, and no human being can possibly process and program all the necessary information to make such simulations coherent.

Don’t get me wrong: Numbers can help us determine the best team. Obviously. But numbers alone can’t finish the job, and unlike people, all wins are not created equal.

Let me ask you this: Are you looking to rank the all-time teams merely by determining "dominance" within the closed context of one league, one year? A five year old could figure that out by simply looking at the W-L records and the run differentials, but to me that's only the starting point, and to bring any discussion of "all-time great teams" beyond the level of spredsheets you're going to have to go beyond that. You have to bring in the whole picture.

**In a matchup between those two teams, for instance, would spitballs be allowed? How far away would you place the outfield fences, and how lively would the ball be? Do you really think there's any way a computer can determine how well Three Fingered Brown's Cubs might stack up against Bob Lemon's Indians without knowing what the rules and playing conditions would be?

**In a matchup between those two teams, for instance, would spitballs be allowed? How far away would you place the outfield fences, and how lively would the ball be?

But is that really Cardsfan premise? Or any of us? A series of individual match ups to produce a single "best ever" team might very well prove ambiguous with team A > team B > C, but C > A.

When you bring up the concept of "best ever" it seems almost necessary that you are going to rank the teams along a linear scale. so the non transitive, Rock/Scissors/paper situation would likely not exist. It's probably not what happens in real life, but any sort of ranking is going to do that. This then brings up another issue:

I'll make my rankings by considering as many factors as possible, as I outlined in # 71. You seem to think there's something ###### up with that, because some of those factors can't be reduced to a number.

You've mentioned that notion, a little earlier as well. That you would use non statistical, perhaps non objective evidence. I dont' know how you are going to do that if the goal is to create a linear ranking. You can't have a ranking like:

1. Team: 25A/15B/16C
2. Team 23A/17B/12C

There would just be no way to rank such teams in a linear order if you are going to use these more or less, non objective factors (not sure how to say that).

I am not opposed to using different parameters, or different statistics. But I dont see how you can use non numerical factors in that AND still produce a linear ranking.

I'm massively still at a loss understanding the messed up thinking involved here. If a team has a very good record overall, and a good record against other good teams, which means they have just a good record against the poor teams.. how is that an indication of their quality?

I guess the thinking is that we see these sort of situations all the time, and in fact the '54 Indians could very well be an example of this. The Marv Levy Buffallo Bills, the Jim Mora Saints, the Bud Grant Vikings, these are all teams that were very good at exploiting weak teams, to the extent that they almost never lost to bad teams, but against an equally matched teams their own weaknesses became much more noticeable. At least that's how I think of it.

I've never endorsed this idea as a way to determine "best ever" although I do think it happens in sports quite a lot. But I am trying to hear Andy's pts and try to point some weaknesses in this approach. I dont think it at all "messed up" It could very well explain how the Indians could go .800 vs the weaker teams and still only go .500 against very good teams. That effect is certainly possible, I think it almost certainly exists in sports. BUT I hesitate to use it as a ranking of "best ever" because of the very problem with using non transitive values and how they mess up a linear ranking.

Another problem is that I am not sure this Rock/Scissors/Paper (mathemtically: "a non transitive set") effect happens all that much. But then again sometimes you do see it in certain pitching match ups. In the early 70s I remember one of our baseball announcers noting that it seemed the Mets always beat the Red, the Reds always beat the Pirates, but the Pirates seemed to beat the Mets. I dont know if it was really true, but it's not hard to imagine the Pirates actually being able to get hits off the Mets staff, but the Mets being strong enuf to stop the Reds, but then Johnny Bench would just shut down the Pirates on the basepaths.

Something like that. But then this goes back to the pt. I am made at the start: If we are going to allow for non transitive types situations, i.e. match up issues, then you are never going to be able to produce a linear ranking of teams. The reasons hopefully would become apparent from reading this.

What all this amounts to is that you want to reduce the determination of "best" (or "historically dominant") teams to a single metric (the regular season W-L record) with perhaps a few minor tweeks (run differential and "normal adjustments", whatever that means), all crunched into a computer and spit out on a spreadsheet

This may be repetitive but just to touch all the bases so to speak. It is okay to use more than one variable or parameter or whatever. But basically I think you would have to concede that if you are going to rank in a linear order all the best teams, then by nature it is going to come down to at the end, a single metric.

Yes? Doesn't mean the info we put is in one variable, but the final result you are going for is a linear ranking of best teams. Is that not so?

The '01 Mariners played .716 overall and .676 against the other 8 winning teams they played in the regular season: .677 against the 5 AL winners and .667 against the 3 NL winners. And AFAIK they remain the only team in the interleague era to post a winning record against every team they played.

So yes, I'd feel justified in saying that those two teams deserve a postseason failure pass more than the 2004 Cardinals,

Well okay, you are now beginning to establish some sort of objective basis for this "post season pass" concept as part of your ranking system.

Now that you've assigned some numbers the arguments get more interesting then. For instance I am still not sure the '75 Reds get any more respect. I dont think they would score high enough on the "total seasonal record" to avoid criticizing them on the basis of their record vs the top teams. It would be interesting to see if you can jigger the system enough so that it puts in all those teams we "think" are the best ever.

And then you're going to assign certain percentages (I think) to the various stats which still leaves us with the Indians. Obviously you want to penalize them for the world series, but then what sort of pct. do you think? How much do we have to punish them to bring them down to the realm of "very good" and will this mess up the rankings of other teams? I think you are going to have land on some sort of number at that pt. and the criticism starts again.

My own thinking in all this is that we should first agree on how much error or random fluctuation can we expect from any set of baseball stats: w/l; runs scored, batting average etc?

My guess is about +/-5%. This is based on a couple of ideas. One is the pythagorean approach that shows teams in 1 run games not winning (or losing) as much as we expect. Another method that I was using was to group all the teams that scored a cumulative total of about 220 of ERA+ and OPS+ together. So far in the last 10 years, these teams average wins from 100-110. Another group with cumulative of about 215, had wins from 95-105. So using OPS+ and ERA+ a team with the same score could still vary up or down by at least 5 maybe 6 wins. It's quite a bit.

SS, I appreciate your patience with my line of reasoning, which is nothing at all like yours, even if in the end we might agree on the best teams.

For one thing, my "linear rankings" would almost certainly be a lot less precise when it comes to differentiating certain teams that are closely bunched together, and I couldn't even imagine reducing them to any single number, other than "#1, #2, #3", and so on. And even there I'd begin by conceding the subjectivity of how I weigh my various factors.

Let me give you an example. I'd probably rank the '98 Yankees as # 1 on an all-time list, and here are some of the reasons:

---Their overall 125-50 (.714) record, by coincidence the exact equal of the 1927 Yankees' regular season record. I include the postseason in the total record, and why not? Even if you choose not to weigh it disproportionately, as I do, it's patently absurd to count 13 games against the worst teams but throw out 13 games against three of the best teams in baseball.

---Their offensive and defensive balance, expressed in identical team OPS+ and ERA+ numbers of 116---tops in the AL in both categories.

---Their Pythagorean record of 108-54, which isn't as lofty as their actual record, but is still a damn good indication of historic excellence.

---The offensive balance within their starting lineup, with the top 7 regulars having OPS+ numbers of 160, 132, 130, 127, 124, 121 and 115. With a bench that included figures of 236, 134, 116 and 107. This starting nine included 7 multiple All-Star picks, with another (Raines) on the bench, plus a 3rd baseman who had his career year while hitting in the ninth position in the batting order.

---A solid if admittedly not brilliant 5 man starting rotation, with a bullpen that featured relievers with ERA+ numbers of 233, 137, 134, and 118.

---An 11-2 postseason in which they outscored the opposition by 62-34

---And they played in an era where the talent pool was arguably as good as it's ever been, though probably no better than it's been since then, and in a year when the leagues were in rough balance---unlike the 1910's (when the AL dominated) or the mid-50's through the 60's, when the NL did.

Negative points:

---While they had a .649 winning record against other winning teams (48-26, again counting the postseason), they failed to match the 2001 Mariners' winning record against EVERY team, as they were only 5-6 against the Angels and 6-6 against the Blue Jays.

---They didn't have to play the best NL team (the Braves) in the World Series. OTOH the next year a decidedly worse Yankees' team swept a Braves' team that was statistically superior, and given the results of that World Series, and the 1998 regular season interleague tally (Yanks 3, Atlanta 1), I don't hold that too much against them.

---Their strength was in their depth of stars, but they had no offensive superstars on the level of a young A-Rod, Griffey or Bonds, and no pitching superstars like Pedro or Clemens. OTOH it might be noted that in 4 starts against the Yanks that year, the CYA winning Clemens was but 1 and 2 with a 3.86 ERA, well over a run over his league leading 2.65 ERA.

I've given you some details about this, not to argue particularly for my choice of the '98 Yanks as the #1 team of all time, but to show you how I'd approach the question, and why I don't see how it's possible to reduce the choice of "best team ever" to a single metric.

I agree there's no reason not to count the post season, we assume the teams are trying hard to win, etc. The spring training is another problem altogether.

Not sure it matters if a teams talent is evenly distributed or not. Are we trying to figure out which team would come out the best in a season played with all these other teams? Distribution may not really matter.

It's entirely possible the 1906 Cubs did not really have much interest in winning the world series as it was not the same type of thing back then. This may put their post season issue in a different standing.

There is so much error built into many of these statistics I think the best that can be done is to come up with a group that is statistically indistinguishable from one another.

The issue of time continues to be important, teams change every month, really they change every day. Are we supposed to account of this team at it's peak? Or an average yearly? Is a two year average any better?

Not sure it matters if a teams talent is evenly distributed or not. Are we trying to figure out which team would come out the best in a season played with all these other teams? Distribution may not really matter.

I see talent distribution within a lineup as a factor, because the fewer holes in the lineup the better. In that respect the 2009 Yankees were an amazing bunch: Take out Melky's 93 OPS+ and the next worst number was Johnny Damon's 118. All of the other starters, including the DH, posted numbers between 121 and 141. A lineup like that leaves opposing pitchers with little or no breathing room.

Of course all things being equal, you'd rather have a lineup with 5 Babe Ruths and 4 Melkys than one with 9 Swishers, but under more ordinary conditions, I think that balance is a positive factor.

It's entirely possible the 1906 Cubs did not really have much interest in winning the world series as it was not the same type of thing back then. This may put their post season issue in a different standing.

Given the level of citywide interest in that intracity Civil War that would be almost impossible to imagine today,** I seriously doubt if the Cubs had anything less than 100% interest in winning. OTOH when a team playing in the stronger of 2 leagues wins 116 out of 152 games and destroys its closest competitors in head-to-head matchups, the most a 6 game WS loss would do would be to drop them down a notch or two in an all-time list. Same with the 2001 Mariners.

There is so much error built into many of these statistics I think the best that can be done is to come up with a group that is statistically indistinguishable from one another.

That's more or less what I've been trying to get at all along, though I might put it differently. And it's not just the inbuilt error, it's that there's simply now way to quantify many factors with any degree of precision.

The issue of time continues to be important, teams change every month, really they change every day. Are we supposed to account of this team at it's peak? Or an average yearly? Is a two year average any better?

Yet another point where it's impossible to find a consensus among historians, which is one reason I think you have to start with the actual results, and then work your way backwards to try to determine any mitigating factors. Personally, however, when it comes to ranking teams, I see injuries as simply something that happens, and at best an "injury bonus" would merely elevate a team up a notch in a ranking of "most talented" teams, as opposed to moving it up on a list of "best" or "most dominant" teams.

**The famous tightwad Charlie Comiskey was so thrilled with his team's win that he threw his owner's share of the pot for the first four games into the players' pool, giving the White Sox players $1,874 each, more than some of them made in their regular contracts. A local railroad magnate re-named a whole string of new stations after the White Sox players. When you combine neighborhood rivalries, newspaper wars, and the centrality of baseball in American culture back then, it's impossible to think that in the absence of game fixing, either team wouldn't have been going all out to win.

Hello everyone. I hope to continue the thread at least until people get sick of it. If I dont post for a few days it doesnt mean I've given it up...

Anyhow, Andy brought up the issue of using players themselves to help fill in the gaps in our Greatest Team Ever contest. But I think there are a number of problems with this, for one thing what else can the players themselve add that we can't get from pythagorean projection and/or won/loss and/or ERA/OPS+? Not sure it adds, but it may distort.

For one thing take the 70s Reds vs the Pirates of that era. of course we mostly think of the Reds as better, and if it comes down to player I guess the Reds have about 4-5 position players that are HoF level (including Concepcion, not sure if he's in). WHereas I guess the Pirates have three (Maz Clemente, Stargell). After that in the very good column, I guess we have Oliver, Madlock, perhaps Sanguillen. The Reds have at least Griffey, Foster, maybe Driessen. So the Reds maybe have more famous career guys but does that really tell us anything? All it seays they went to have better careers...

Another there is that of timing, viz. what time period are we going to use? Monthly, weekly? Seasonal? The issues we have with players do we take Pete Rose at his peak? Do we assume the As get Reggie without crutches? I am more likely to just avoid the whole player issue as it seems that team w/l or pythagorean etc. should give us enuf data.

Then when it comes to evaluating players we almost have the same issues that we are having with teams. Do we take peak value vs career? Personally I like pitchers with great peaks because that is more likely to bring you a championship. Whereas position you can use position players that can give you 8-10 pretty good years and find some stars to fill it out. But there's other problems e.g. strength of the league. As Ty Cobb's Tigers lost 3 WS do we have to downgrade his performance some?

One thought that occurred to me in the league strength issue was that any large discrepancy between good/bad teams merely reflects mal distributed talent. It really says nothing about the talent level in terms of absolute strength. I.e. if the Yankees win the pennant by 20 games and there are some sub .400 teams in that league, all it means is talent is maldistributed. It tells us nothing about how good the Yanks are vis a vis the Dogers..

Ok so much for that, but what if there is a large maldistribution of talen, but there is like 2 or 3 competitive teams at the top? The team that comes out ahead of them might make a better case for being excellent. An example is '06 Cub who were playing .700 ball but the NY Giants in second place were actually playing e.g .617 ball and the Pirates were well above .500. Looking at those teams, the Cubs still kicked the Pirates/Giants ass.

I was wondering if we could create a metric for that and would it help any?

So anyhow, I'll be checking back, I'll let you know if I am giving up the thread.