The only problem with this view about anti-moneyball is that the A's got into contention because their offense got red hot and now they are losing again because they aren't scoring any runs. If they had maintain their offensive production of June and July all the way through to the end of the season they would have a 4.31 RPG average for the season which would be good for 8th in the league.

the ability to outperform their pythagorean record

So, last I checked no one had shown this to be an actual ability rather than just random chance. Has this changed?

So, last I checked no one had shown this to be an actual ability rather than just random chance. Has this changed?

There are probably minor factors that could create a team destined to outperform Pythag. Say your team has a league average bullpen, starting with three of the leagues best relievers, and filled out with pure garbage. Your team would outperform the league average in close games, but might lead the league in blowout losses.

Also possibly if you have 3 super stud starting pitchers, and the worst 4th and 5th starters in the league. You win a lot of games with your "big three" by a typical amount of runs, then the back end of your rotation consistently gets the stuffing kicked out of it, losing their starts by a much higher than typical run spread. But in that scenario it's still hard to have a winning team.

But would you consciously design a team like this? No.

There are probably minor factors that could create a team destined to outperform Pythag. Say your team has a league average bullpen, starting with three of the leagues best relievers, and filled out with pure garbage. Your team would outperform the league average in close games, but might lead the league in blowout losses.

Also possibly if you have 3 super stud starting pitchers, and the worst 4th and 5th starters in the league. You win a lot of games with your "big three" by a typical amount of runs, then the back end of your rotation consistently gets the stuffing kicked out of it, losing their starts by a much higher than typical run spread. But in that scenario it's still hard to have a winning team.

But would you consciously design a team like this? No.

I think there's one thing a team could try which is to never deploy their best short relievers except in high leverage situations. So once the game gets "out of hand" (for however that team defines it), a long man goes in.

I think there's one thing a team could try which is to never deploy their best short relievers except in high leverage situations. So once the game gets "out of hand" (for however that team defines it), a long man goes in.

But both this and what VA posted above are basically about yielding more runs than you could, not getting more wins out of your RS/RA mix.

In 2002, Oakland had an above-average team OBP (.339), but they failed to win the AL pennant, let alone the World Series. The Anaheim Angels went on to win it all that year, and in fact had a higher team OBP (.341) than Oakland.

I believe this is the type of factoid that took the Freakonomics author's analysis awry when evaluating moneyball. Beane was able to produce an above average OBP team (well above average considering park factors) on a much smaller budget than his rivals, something he could never have done with Batting Average, which the market valued highly (too much so). The Angels had the budget to buy better athletes with better batting averages, and got high OBP as part of the package.

2. 2007 Arizona Diamondbacks (95.54 OBP+, 90 wins, won NL West):

Only Conor Jackson and Orlando Hudson had high OBPs for this team. That's right, Conor Jackson and Orlando Hudson. Their offense was awful, despite playing in the hitters' haven that is Chase Field. Their pitching staff was good, though: they were tied for 2nd in ERA- (88) and tied for 6th (95) in FIP-.

The '07 D-Backs beat out both the Padres and Rockies (who played that incredible Wild Card play-in game) by a game in an extremely close division race. Arizona's pythagorean record (79 wins) was 11 games worse than their actual record, a significant difference. Both the Padres and Rockies had much better run differentials, but for whatever reason the Diamondbacks were able to beat both of those teams out for the division.

My memory is that the DBacks sort of fit the bad back end of bullpen/rotation model. They were a pretty good team 90% of the time, but got spanked in some huge blowouts where they were forced to let the short bus pitchers have prominent roles on the mound.

They were 32-20 in 1 run games (+12 runs), and 20-26 in blowouts (-72 runs). They actually had 5 relievers with ERA+ (bad measure, I know) over 146, and the rest didn't have many innings, but it appears they had the worst innings.

But both this and what VA posted above are basically about yielding more runs than you could, not getting more wins out of your RS/RA mix.

I think a team doing that would then be more likely to deploy top relievers in the 5th/6th inning in close or tie games which might help marginally.

They were 32-20 in 1 run games (+12 runs), and 20-26 in blowouts (-72 runs). They actually had 5 relievers with ERA+ (bad measure, I know) over 146, and the rest didn't have many innings, but it appears they had the worst innings.

The Orioles have taken this to another level this year:

21-6 in 1 run games (+15 runs)
15-26 in blowouts (-74 runs)
11-2 in extra inning games

They have 5 relievers with IP> 25 and ERA+ >148. Their closer has only a 124 ERA+, but he’s allowed over half of his runs in two games, one of which was already a blowout loss.

Oriole pitching by WPA leverage:

High Leverage .644 OPS
Medium Leverage .737 OPS
Low Leverage .770 OPS

The league as a whole has shown almost no difference in performance by leverage.

You want to know a franchise that sucks at winning a one run game? The Chicago Cubs, that's who. Since 1914 their best spread was +10 wins in 2003. You have to go back to deadball era and the great Cub teams to find really good 1 run game records.

So is +10 wins the least most in the liveball era?

The Angels had the budget to buy better athletes with better batting averages, and got high OBP as part of the package.

Let's not confuse the current Yankee-West version of the Angels with the team that won it all in 2002. They did outspend the A's that year, but were outspent by the Mariners, and greatly outspent by the Rangers. The AL west that year actually finished in inverse order of team payroll.

A's 40
Angels 62
Mariners 80
Rangers 106

The lineup contained no big ticket free agents. The lineup was full of home grown products (Salmon, Erstad, Glaus, Anderson, Molina) or a cheap pickups who turned into contributors (Spiezio, Eckstein, Fullmer). Only Salmon had a big contract at that point.

On the pitching side, Washburn, Ortiz, and Lackey were in their cheap years. They paid big bucks only for Appier (who was decent) and Sele (who was replaceable, and got replaced). The outstanding bullpen was full of cheap players other than closer Percival.

That's true that they were either home grown or picked up on the cheap but it is also true that the Angels could spend more back then to keep their players. Tim Salmon was in the middle of a long term contract that paid almost 10 million a year. Garrett as well was probably too much money for him to be on the A's.

Beane was able to produce an above average OBP team (well above average considering park factors) on a much smaller budget than his rivals, something he could never have done with Batting Average, which the market valued highly

I'm surprised this myth has proven so durable. There is very little evidence that OBP was undervalued enough to make a practical difference in building a team. It may be true that many teams didn't pay enough attention to BBs when evaluating players. However, what kind of players draw a lot of walks despite a modest BA? The answer, mainly, is power hitters. And there is no reason whatsoever to think that teams didn't value HR hitters--quite the contrary. So practically speaking, there was no way to gain any significant edge from this alleged market inefficiency. And in fact, while Oakland did have a good BB rate, it also had a high ISO (4th in the AL in 2002), one right in line with it's BB rate.

A's did pay Eric Chavez more than Salmon just 2 years later, so the could have afforded one big contract like that if they wanted to. But I have no doubt that they would have let Salmon walk and take the comp picks, with a key reasoning being that Chavez was mid 20's and Salmon early 30's. Though neither of those contracts turned out to be wise investments. Anderson? Had he come up with the A's he would have been traded for a package of prospects right after the 1995 season. He was most definitely not their type of player.

Garret Anderson was the sabermetric whipping boy in the late 90's. A's would have never tolerated a corner OF with no walks and only moderate power. Angels liked him because he got RBI. He was actually a decent player, because he was a strong defender, well above average in a corner and capable in center. But pretty much nobody noticed that in the late 90's.

A's in 1999 got over 70 homers and 200 RBI from John Jawa and Matt Stairs (who was as tall as a jawa.) Total cost was under 2.5 million for that season. Stairs was a good pickup in his arb years who was not given a chance by other teams. Jaha was a guy they took a chance on after 2 unproductive, injury marred seasons. They hit big on that year, but didn't really come out ahead looking at the whole picture. They gave Jaha two more years and 6 million dollars, during which he gave them 21 hits and 1 homer.

Most of the A's who were bargains during that time were bargains because they were in the cheap free agency years. They got 6 superstar talents (Mulder, Hudson, Zito, Tejada, Chavez, Giambi) out of the draft (+1 international signing) in a very short time. That was quite an impressive feat.

So practically speaking, there was no way to gain any significant edge from this alleged market inefficiency.

BB rate was one thing they were looking at. They were also looking at unattractive players, and slow players, and players that had not lived up to their MLEs. That's how you ended up getting Geronimo Berroa, Matt Stairs, John Jaha, Rich Becker, Olmedo Saenz, Billy McMillon, Frank Menechino, Mark Bellhorn and Jeremy Giambi all on the team around the same time. The A's of the late 90s looked like a softball team, an image they gradually left behind. Of course lots of these guys were role players or never contributed anything - that the A's success was mostly due to the more traditional talents of Chavez, Tejada, Zito, Hudson, Mulder is something we are all aware of.

The answer, mainly, is power hitters. And there is no reason whatsoever to think that teams didn't value HR hitters--quite the contrary.

The theory isn't that nobody valued 40 home run guys with low batting average but high OBP guys but the A's did and went out and got them. They couldn't afford those guys either. The theory is that the A's recognized that a guy hitting, say, .250 but with 20 homers and 70 walks was much more valuable than the markets were currently viewing them so he could pick them up on the cheap and they would provide more wins than a conventional player that the market had priced at equivalent level or higher.

They were also looking at unattractive players, and slow players, and players that had not lived up to their MLEs.

Right, and as a corollary of "slow" they got poor defensive players if I recall correctly. But what was the net value of all these players, in comparison to their salaries? It's not clear they were a plus at all, much less an important part of the A's success. I certainly don't think anyone believes that defensive skill was overvalued a decade ago.

There was an academic paper years ago that claimed to confirm that MLB had undervalued OBP prior to Moneyball, and then the market corrected. But if you use their model to evaluate how much the As players were undervalued pre-Moneyball, what you find is that the savings were miniscule, maybe 3% of payroll. It's a fun story for us stats nerds, but it just has nothing to do with the team's success.

The theory is that the A's recognized that a guy hitting, say, .250 but with 20 homers and 70 walks was much more valuable than the markets were currently viewing them so he could pick them up on the cheap and they would provide more wins than a conventional player that the market had priced at equivalent level or higher.

I understand that is the theory. But it is wrong. Never happened.

The A's of the late 90s looked like a softball team, an image they gradually left behind. Of course lots of these guys were role players or never contributed anything - that the A's success was mostly due to the more traditional talents of Chavez, Tejada, Zito, Hudson, Mulder is something we are all aware of.

But they did contribute. Just because they didn't rack up a combine 50 WAR per season doesn't mean they didn't contribute to success.

For instance Velarde contributed to the 2000 team and he was picked up from the Angels for basically nothing. The A's traded their 2nd first round pick (32nd overall) from 1997 for him plus some scrubs but he was never much of anything.

Olmedo Saenz played great in his limited role in 2000 and he was picked up for the major league minimum.

I understand that is the theory. But it is wrong. Never happened.

Uh, okay.

A's did pay Eric Chavez more than Salmon just 2 years later, so the could have afforded one big contract like that if they wanted to

Well, two years in baseball is a long time. The A's were at a different place financially and as a team on the field two years later.

As for Anderson the point was that the Angels could spend more to either get players or to keep the players they had at that point in time.

The theory is that the A's recognized that a guy hitting, say, .250 but with 20 homers and 70 walks was much more valuable than the markets were currently viewing them so he could pick them up on the cheap and they would provide more wins than a conventional player that the market had priced at equivalent level or higher.

Here's whom I find that sort of fits that description, other the high draft choices -- Chavez, Grieve, Crosby:

1999 - Tony Phillips, Olmedo Saenz
2000 - Matt Stairs, Jeremy Giambi
2002 - Jermaine Dye
2003 - Erubiel Durazo
2004 - Jermaine Dye, Bobby Kielty

These seem like useful players but nothing revolutionary.

It's a fun story for us stats nerds, but it just has nothing to do with the team's success.

I won't sign on to "nothing," but obviously the story of Beane Discovers OBP is overrated. The team's success was built on the crazy frontline talent they had. They did plug the holes around the stars with players that were arguably undervalued, and undervalued for a variety of reasons, not just BB rate. But anyone that boils the Beane story down to just BB rate is missing a lot anyway - he was also a trendsetter in a few other ways, like trusting minor league numbers and amassing compensation picks.

For instance Velarde contributed to the 2000 team and he was picked up from the Angels for basically nothing. The A's traded their 2nd first round pick (32nd overall) from 1997 for him plus some scrubs but he was never much of anything.

I'd do that trade a million times if I'm the Angels. In 1999 Angels were going nowhere and had a 36 year old 2B having a career year. They got a solid 4th outfielder (DaVannon) in return.

I think Ellis probably counts. He was a high-OBP guy whom they paid nothing.

Ellis was paid nothing because he wasn't yet a FA (or arb-eligible).

Saenz produced little. Dye was a disaster. Velarde was useful. Obviously, you can find individual examples of players that seem to fit the mold and who seem to pan out. But the question is whether there is any evidence that the A's systematically benefitted from buying OBP (or BBs) on the cheap. As far as I can tell, there is no good evidence this is true. And the one effort I'm aware of that claims to show it is true -- the Sauer-Hakes study -- actually shows that the A's payroll was only reduced by 3% because OBP was "too cheap."

They got a solid 4th outfielder (DaVannon) in return.

5 years later. I think any reasonable GM could have found a solid 4 OF'er if given 5 years to do it in. Plus they gave away a player who at the present time was playing well and allowed the A's to be the one to trade him away for value. They got Aaron Harang for him but unfortunately they traded him away before he did much for Jose Guillen who was killing it for the Reds but didn't do much for the A's before he left for FA and they didn't offer him arbitration.

Ellis was paid nothing because he wasn't yet a FA (or arb-eligible).

Yes and the A's traded for him.

But the question is whether there is any evidence that the A's systematically benefitted from buying OBP (or BBs) on the cheap. As far as I can tell, there is no good evidence this is true. And the one effort I'm aware of that claims to show it is true -- the Sauer-Hakes study -- actually shows that the A's payroll was only reduced by 3% because OBP was "too cheap."

I have no idea what this study is, but if it's only looking at payroll it is badly flawed. You don't just buy players, you draft them, you trade for them, and those are opportunities that allowed Beane to gain an advantage.

But again, looking only at BB rate is nonsense, because there were always other factors at play.

Mark Ellis was up and down with the OBP during his time in Oakland. For his career there it was .331.

Plus they gave away a player who at the present time was playing well and allowed the A's to be the one to trade him away for value. They got Aaron Harang for him but unfortunately they traded him away before he did much for Jose Guillen who was killing it for the Reds but didn't do much for the A's before he left for FA and they didn't offer him arbitration.

Yup, Billy Beane makes a lot of trades.

I have no idea what this study is, but if it's only looking at payroll it is badly flawed. You don't just buy players, you draft them, you trade for them, and those are opportunities that allowed Beane to gain an advantage.

I'm not sure how else one would try to document that MLB teams were undervaluing OBP. But it's fine with me if you want to throw out this study. Then the question is: do you have any evidence at all that teams were undervaluing OBP, and/or that the A's were able to exploit this inefficiency? I don't know of any.

"But again, looking only at BB rate is nonsense, because there were always other factors at play."

Why is looking at BB rate "nonsense?" It's one of the claims made in Moneyball, and it's widely believed. What's wrong with examining it? You seem to want to ask some other question (maybe "was Beane right about anything at all?"), but that's not the question I'm addressing.

Mark Ellis was up and down with the OBP during his time in Oakland. For his career there it was .331.

Well, 2011 was horrible for him. Up until that point his OBP was .336. AL secondbasemen were averaging a .329 OBP from 2002 to 2011. Fangraphs views him as a good value even as his contract got into the mid single digit millions.

Does anyone know how much the average secondbasemen cost overall from 2002 to 2011? The A's spent about 25 million for his production.

I'm not sure how else one would try to document that MLB teams were undervaluing OBP. But it's fine with me if you want to throw out this study. Then the question is: do you have any evidence at all that teams were undervaluing OBP, and/or that the A's were able to exploit this inefficiency? I don't know of any.

I don't know of a better study, and I don't have any evidence beyond anecdotes. But it seems like the study has to be thrown out, if you've represented it properly. All pre-arbitration players have the same payroll costs in such a study, but the costs of acquisition differ wildly, and in ways that are not easily measured. (For example, Beane got Jeremy Giambi for Brett Laxton, but maybe today he would be worth a a better player? I don't know if it's possible to measure things like this)

Why is looking at BB rate "nonsense?" It's one of the claims made in Moneyball, and it's widely believed. What's wrong with examining it?

There's nothing wrong with examining it. You're right, it's not "nonsense." But it is reductive.

John Jaha signed a minor league contract and ended up playing for the league minimum in 1999. Heading into that season he had over 2000 plate appearances and a .361/.463 line. Nobody else would take a chance on him. Was the going rate for a .361/.463 hitter just 3% higher than major league minimum back then?

I think that might be PF issue with the study. It is really really hard to tease out just OBP costs and values from a contract.

"But again, looking only at BB rate is nonsense, because there were always other factors at play."

Why is looking at BB rate "nonsense?" It's one of the claims made in Moneyball, and it's widely believed. What's wrong with examining it? You seem to want to ask some other question (maybe "was Beane right about anything at all?"), but that's not the question I'm addressing.

Ya I don't get this either. It's a common meme to ridicule looking at the A's BB% and OBP with a retort that "Moneyball was not about OBP, it was about exploiting market inefficiencies with undervalued assets", but the fact is that the 2000s As had an offense that ranged from good to crappy, but consistently ranked high in BB%. There really doesn't seem to be a lot more to their offensive philosophy than OBP. Not to mention Beane was hardly the first human being in the history of mankind to try to find undervalued assets. The first person to pick a berry and eat it did that.

John Jaha signed a minor league contract and ended up playing for the league minimum in 1999. Heading into that season he had over 2000 plate appearances and a .361/.463 line. Nobody else would take a chance on him. Was the going rate for a .361/.463 hitter just 3% higher than major league minimum back then?

That's an OPS+ of 113, for a DH. I don't think teams were looking at that average production, or even his upside (144 OPS+ in 1995). What they saw was a guy who in the two previous seasons had hit 225/361/407, was 33 years old, and injury prone. Beane took a gamble that he had a comeback in him. He won that bet big time. He then took a gamble that Jaha had a bit more left in him, giving the guy 6 million over 2 years. He lost that bet.

I think it's pretty clear that the A's were targeting affordable players who knew how to take ball 4. It's less clear how much value they got from this approach, given that to go after these players you're often sacrificing batting average, defense, or gambling on injury prone players.

Beane took a gamble that he had a comeback in him. He won that bet big time. He then took a gamble that Jaha had a bit more left in him, giving the guy 6 million over 2 years. He lost that bet.

It's an interesting question whether targeting previously injured players is a good strategy. Maybe on the margins it is -- it would make sense that rich teams would be willing to pay a premium for certainty of performance. But to evaluate this you have to look at all these players -- including those who didn't play at all, or put up -1 WAR in 5 weeks before crapping out -- not just the ones we remember because they overperformed.

It's a common meme to ridicule looking at the A's BB% and OBP with a retort that "Moneyball was not about OBP, it was about exploiting market inefficiencies with undervalued assets",

Of course I'm not trying to say that. Like everyone else, I got sick of that comment more than five years ago. Beane didn't invent some new economic principle - he looked for guys with good BB rates, and guys that scouts thought looked funny, and guys with good AAA numbers, and guys that didn't really have a position...

Guy wants to know how much money Beane saved by understand BB rate better than everyone else. I'm saying that it's not possible to come up with a satisfactory answer. Just take a look at the trade for Jeremy Giambi. How do you evaluate it? It's impossible to determine if he was undervalued by other teams, or to what extent, and even if you knew those things, it's impossible to determine to what extent it was because GMs didn't value his BB rate, or because he was an awkward tub of goo or because he was a ########. There are dozens of moves with just as many moving parts. It's not possible to parse out BB rate and come up with a reliable number like -3%.

Even if you can answer all that, how do you evaluate how much better he made the team? Obviously you can't just look at payroll. Do you measure value against replacement level? Maybe Billy's tendency to pluck Ken Phelps types from other organizations meant that his teams had fewer sub-replacement performances, and WAR doesn't capture that. And even if you get this all straight you probably get stuck arguing over the importance of tiny margins ("over 200 ABs Frank Menechino's walking ability only results in an extra 4.6 runs").

Guy's looking for a study that will never exist, and his turn of logic in #12 ("So practically speaking, there was no way to gain any significant edge from this alleged market inefficiency.") doesn't stand up. I agree that Beane's search for market inefficiencies were only a small part of the success of the A's, but mostly what Guy's doing is taking a strawman (that Beane=OBP - yes, a commonly repeated strawman, but still a strawman) and rejecting it because it doesn't meet an impossible standard for evidence.

mostly what Guy's doing is taking a strawman (that Beane=OBP - yes, a commonly repeated strawman, but still a strawman) and rejecting it because it doesn't meet an impossible standard for evidence.

Now why in the world would you say that, when I've already stated explicitly -- to you personally -- that I'm not trying to evaluate Beane and his overall performance or philosophy, but only the claims about whether OBP was undervalued? You're just being willfully obtuse. And the idea that Beane discovered an inefficiency in the pricing of OBP is certainly not a strawman -- it's reported in Moneyball, and many fans definitely still believe it (you even seem to).

But there is good reason to think that Beane could not have benefitted much from OBP being undervalued. The study I refrenced was one that SUPPORTS the Moneyball thesis about OBP. The authors found that MLB was valuing OBP less than it should. My point is simply that if you look at what Beane's players were paid, and compare that to what they should have been paid given their OBP -- again, according to the most important "pro-Moneyball" study -- what you find is that the payroll difference is tiny. (And that's not very surprising, since high BB-rate hitters hit a lot of HRs, and the market rewarded HRs.)

I think Beane was right about closers being overvalued. Maybe he was right about HS and college players. Hell, maybe he was right to go after lousy defensive players. I'm not commenting on any of that. Why you can't get that through your head, I can't imagine.....

I think it's pretty clear that the A's were targeting affordable players who knew how to take ball 4. It's less clear how much value they got from this approach, given that to go after these players you're often sacrificing batting average, defense, or gambling on injury prone players.

The other thing they seemed to be doing was to try to teach minor leaguers to take Ball 4. I'm not sure how well that worked either. Even though I'm an A's fan, I find Moneyball and Beane really irritating.

Hakes-Sauer in 2007

. The “frontier” that describes the most costeffective combinations of payroll and wins contracted sharply in 2004 as other teams
began to compete for inputs, ending a period of years when Oakland was able to push out the frontier by purchasing wins cheaply in the form of walks.

The relative increase in returns to patience can more easily be seen by calculating
ratios of the coefficients in Panel B of Table 3. While the relative returns to Bat and
Power were roughly constant throughout the sample, the ratio of the coefficients of Eye
over Power more than tripled, from 1.67 in 1986-1993 to 5.36 in 2004-2006. Similarly,
the ratio of the coefficients of Eye over Bat increased 177% between the earliest and
latest period, from 0.30 in 1986-1993 to 0.83 in 2004-2006. This supports the Moneyball
hypothesis that it was walks in particular that the Oakland A’s could purchase relatively
cheaply during the earlier periods in the player market.


Although it is clear that the returns to plate discipline are far greater now than in
past years, the salary coefficient ratios in Table 3 in all instances remain smaller than the
corresponding productivity ratios in Table 2. However, this is not necessarily evidence of 10
persistently inadequate compensation for players who draw more walks. The salary
regression coefficients are measured as returns to a one point increase in a statistic. Yet
one point of on-base percentage may represent a larger skill improvement than a point of
slugging percentage.

Looking across the 21 sets of annual labor market regressions in Table 5, it is
clear that hitting for power has consistently been rewarded in salaries, and at more or less
the same “price.” Even hitting for average (Batting) has traditionally influenced salaries,
allowing for a couple of tiny hiccups. Plate discipline (Eye), however, and on-base
percentage as a whole were typically ignored prior to 1995. Then, after a hesitant
flirtation in the late 1990s, returns to on-base and Eye dropped sharply in 2001 and 12
remained statistically insignificant until the sudden two-year burst in 2004-2005. The
magnitudes of the coefficients are still reasonably high in 2006, but they are not
statistically significant. The spike in the salary returns to Eye relative to Power can be
seen more clearly in Figure 1. The plot point for the ratio in 2006 corresponds to the
regression results in that it is still among the higher historical values, but a marked
decline from its 2004 maximum. It is too soon to tell if this is itself a hiccup, or if the
market reflects a determination that teams were overpaying for plate discipline.

Hakes and Saur in 2004

Our work is essentially an assessment of
Lewis' argument, and as such is merely an after-the-fact replication of work done by the
innovators at the heart of his book. But we do find that they (and Lewis) were right, and
further, that the process that they set in motion had in large part been completed by the
time the book was published. That the addition of only a few individuals was able to
correct a long-standing anomaly illustrates both the inefficiency which can result when
markets are isolated from competition, and the swiftness of market corrections once entry
occurs.

Everything I have found from Hakes and Sauer has them saying that Billy Beane purchased OBP cheaply.

Everything I have found from Hakes and Sauer has them saying that Billy Beane purchased OBP cheaply

Umm, yes, that was their conclusion (as I indicated earlier). But they neglected to examine whether the underpricing of OBP made any practical difference when assembling a team. If you take what H & S say is the correct pricing of OBP and SLG, and compare that model to what A's players were actually paid, you find the difference is too small to matter. There are many other problems with the papers -- their estimates of what teams were allegedly paying for OBP pre- and post-Moneyball aren't worth much. And their conclusion that the salary market suddenly corrected in 2004 is frankly absurd. But the key point is that even if the authors were correct about the underpricing they found, it just didn't matter in baseball terms.

So let me get this straight, you've been using Hakes and Sauer as proof that Beane wasn't getting value and yet you note that their papers have many problems? Sounds like cherry picking to me. Like PF said, why should we take anything in these papers seriously? Garbage in-garbage out.

Now why in the world would you say that, when I've already stated explicitly -- to you personally -- that I'm not trying to evaluate Beane and his overall performance or philosophy, but only the claims about whether OBP was undervalued?

My apologies. I assure you that any obtuseness on my part is natural, not put on.

The reason I keep bringing the other Beane-attributes (lack of speed, good MLEs, etc) back into the conversation is that they cannot really be separated from the BB rate part of it. You're trying to talk about BB rate alone, and I understand why, but I don't think it's possible.

There are many other problems with the papers ... And their conclusion that the salary market suddenly corrected in 2004 is frankly absurd. But the key point is that even if the authors were correct about the underpricing they found, it just didn't matter in baseball terms.

It seems to me that the key point ought to be that this study can be safely ignored.