Baseball for the Thinking Fan

You are here > Home > Primate Studies > Discussion
Primate Studies
— Where BTF's Members Investigate the Grand Old Game

## Thursday, October 06, 2005

#### Shootin’ Craps

Voros returns to BTF with a look at probability and the playoffs.

You often hear members of baseball’s underground intelligentsia (otherwise known as ‘stat drunk computer nerds’) describe baseball’s playoffs as a ‘crapshoot.’ Now despite the fact most of us subterranean baseball thinkers would likely soil ourselves if we ever found ourselves in the midst of a real back-alley craps game, in general we seem pretty damn confident we know what we are talking about when we use the term.

Not one likely to take these sorts of things on faith, I decided to take a cursory (please remember that word before someone starts chiming in with stuff you need a PHD just to get the introductory textbook to)…

…um, where was I—oh yes, I decided to take a cursory look at the issue and see if the statistical geekerati know what the hell they are talking about.

The experiment is simple. Estimate the winning chances in a game of two teams with given winning percentages and adjust for home field advantage. Then take those numbers and come up with the chances of winning a 5 or 7 game series.

The following tables list the chances of winning 5 or 7 game series for teams with differing everyday win percentages. The win percentages aligned vertically on the left side are the win percentages of the team with home field advantage, the winning percentages across the top are for the team with fewer home games in the series. The given resulting percentages are the chances of winning the series for the team with home field.

Home field advantage is set so that a .500 team would win at a .600 clip at home and a .400 clip on the road. Or, for our purposes, a series between two evenly matched teams would have the home team win on average 60.0% of the games. Why is that so much different than we see in the regular season?
(around .540) Well since the war, the home team has won roughly (very roughly but there are side issues at work that I don’t want to get into) about 60% of
the games in World Series play. Since there’s no inherent reason why a home team should be the better overall team in the World Series (until recently home field simply alternated leagues), this suggests that home field advantage is stronger in the postseason than in the regular season. I’ve decided to base the above on that assumption, since it’s quite reasonable to come up with reasons why that might be the case. (Insert gratuitous swipe at Eric Gregg here).

To come up with the winning percentage of a single game, the following formula was used:

A = Team A’s Win% (home team)

B = Team B’s Win%

((A*(1-B))/.5)/(((A*(1-B))/.5)+(((1-A)*B)/.5)) = Team A’s chance of winning

Batter-Pitcher Matchup

Then I raised that number by .737 to adjust for home field advantage in that game (an exponent works well here because you can’t go above 1 or below 0 with it). From there it’s easy to come up with full series winning percentages based on those numbers.

Anyway, what the table shows is that the geeks are, to an extent, correct. Any individual series can be a bit of a crapshoot. For example, if we assume the Cardinals face the Padres and their ‘true’ win percentages are .625 and .500 respectively, the Cardinals, even with home field advantage, expect to win the series only about 73.77% of the time. While that may seem like a lot, in actuality the Padres chances would be a little below the chances of rolling a ‘9’or higher in craps. In other words while the Padres are indeed underdogs, they certainly have a shooters chance.

Still, just like in craps, while any outcome can easily happen, some outcomes are more likely than others and the Cardinals do rank as clear favorites to win the series if we assume those win percentages are correct. Of course this is without adjusting for different starting pitchers and so forth, but remember this is just a cursory look to give us an idea of what level of crapshoot things are. Please do not use this info for wagering purposes, unless of course you win, in which case I’ll set up a pay pal site for you to send me my cut.

It’s also worth noting that even a favored team whose chances of winning each series it participates in is 70%, would still only win all three series about a third of the time.

So we geeks are essentially right. (And don’t we just love to announce it to the world when we are). The playoffs are, to a certain extent, a crapshoot. Baseball’s current playoff format ensures that even the very best teams are likely underdogs to win the World Series once they make the playoffs.  There’s just not enough games against too strong of opposition to swing the balance too heavily in their favor. I’ll leave it to others to argue whether this be good or whether this be bad.

And while it’s unlikely the Padres will wind up winning World Series rings (or watches), their chances are not probably not too far from the chances of rolling snake eyes.

Which can happen, something we geeks know from many years of playing cutthroat games of

Risk

“Craps.”

Voros McCracken Posted: October 06, 2005 at 12:54 AM | 314 comment(s) Login to Bookmark
Related News:

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 3 of 4 pages  < 1 2 3 4 >
201. Srul Itza Posted: October 07, 2005 at 08:01 PM (#1669293)

If Tejada’s baserunning gaffe is the proximal cause of the A’s losing the last series, then luck did play a huge role

No, stupidity played a huge role.  As it does in all human endeavors.

202. John Lowenstein Apathy Club Posted: October 07, 2005 at 08:02 PM (#1669294)

Did Lindstrom do everything exactly right? Because those grounders that do find the glove, no one talks about luck.

Well, this was a *really* bad hop - the kind that don’t exist anymore because the fields are all synthetic clay and level as a pool table.  By all accounts he was in position to take the grounder and the last bounce was right over his head.  I wasn’t there, though.  :)

I know that’s extreme, but I think it continues to apply at less extreme levels.  Sure, the fielder has some responsibility when he dives for a ball and it hits a seam in the carpet and goes over his glove - but the luck is playing a significant causative role.  I mean, ask a ballplayer.  It’s not for nothing that they’re all as superstitious as a Basque grandmother.

Ultimately, though, I think it’s fair to say “bad luck? So what. You make your own luck.” That’s not a very intellectually satisfying position, but I think it cuts through a lot of the “luck” talk that people find so frustrating.

203. DCA Posted: October 07, 2005 at 08:03 PM (#1669300)

BL,

All I’m saying is there’s not enough evidence from games played to prove that there is a structural failing to A’s unless you arrive at the structural failing theory independent of knowing the results of the game.

It’s fine to posit structural failings, and it’s fine to look for pontential structural failings, but it’s not fine to ridicule someone for believing that there is nothing structurally wrong based on the results of games played, as if it were self-evident.  You do the first and the last.  I quibble only with the last, not at all with the first.

I wish more people would do the second.  It’s also fine to posit no structural failings, and not fine to riducule those who believe there are structural failings.

204.  Posted: October 07, 2005 at 08:04 PM (#1669301)

The 1990 Reds were, IIRC, just the fifth team to lead their league (or division) from wire to wire. They were definitely not underdogs against the 1990 A’s.

They did lead wire to wire, but the Reds were nevertheless huge betting underdogs in that Series.  In fact, some even termed it (wrongly) the biggest upset in baseball history.  The A’s were then being talked about (again wrongly) as being among the greatest teams of all time, an illusion which seemed to end with that World Series.

I would maintain that at most only a half dozen or so World Series could be termed “upsets” in any real sense of the word.

205. DCA Posted: October 07, 2005 at 08:08 PM (#1669310)

This is not an appeal to authority, fwiw

Really. That sounded exactly like an appeal to authority.

No, because I’m just telling you why *I* don’t believe you.  I couldn’t care less what other people think of my argument.  I think it stands on its own.

206. DCA Posted: October 07, 2005 at 08:09 PM (#1669317)

I mean, what other people think of my authority as a reason to believe my argument.

207. John Lowenstein Apathy Club Posted: October 07, 2005 at 08:11 PM (#1669325)

Voros, if you’re still reading and you have time, could you do me a bigbigbigbigbig favor and run the numbers again for a best-of-nine series?  If there isn’t general interest, you can e-mail me at .(JavaScript must be enabled to view this email address).  Thanks very much in advance.

208. strong silence Posted: October 07, 2005 at 08:12 PM (#1669328)

No. If Tejada’s baserunning gaffe is the proximal cause of the A’s losing the last series, then luck did play a huge role, and could be the major underlying cause. Plays like that, regardless of resolution, are rare events the like of which are highly unlikely to occur in a single game or short series, and I would contend, independent of the team’s abilities. That the event happened in a playoff game instead of july series is largely due to luck. Futhermore, the ump blew the call on Tejada—by the rules, his stopping running does not affect whether or not he is awarded home plate, that decision happens at the time of the foul.

IIRC, if he had kept running he would have been safe. (Replays indicated he almost certainly still could have made it home before the throw.) He stopped and was tagged out. If I’m right the play was in his control and there is a causal factor other than luck.

209.  Posted: October 07, 2005 at 08:16 PM (#1669337)

at most only a half dozen or so World Series could be termed “upsets” in any real sense of the word

1906, 1914, 1985, 1987, 1988, and ...? or not? and are those mid-80s upsets illusions, or did upsets flourish then, or is it Whitey Herzog’s fault, or what?

A lot of AL losses in the 1950s and 60s were seen as upsets at the time, because people were used to the Yankees winning every year, and NL teams with inferior records won in ‘54, ‘59, ‘60, ‘63, and ‘64; but the NL was truly the stronger league at the time ...

210. Srul Itza Posted: October 07, 2005 at 08:19 PM (#1669347)

1969

211. and Posted: October 07, 2005 at 08:22 PM (#1669354)

2001, 2003

1996, 1997

1990

212. strong silence Posted: October 07, 2005 at 08:25 PM (#1669363)

To me 1990 was not an upset. One of five teams to lead wire to wire - that is strong evidence of a special team.

That dynasty claim was without merit, IMO. 1988 - they lost the WS. In 1989 A’s got the break of the earthquake (it gave their two dominant pitchers, Stewart and the other guy a chance to rest) I knew in 1990 that they were not the dynasty people claimed.

213. DCA Posted: October 07, 2005 at 08:25 PM (#1669367)

If I’m right the play was in his control and there is a causal factor other than luck.

But it shouldn’t have been in his control ... and it was highly improbable that it or something like it would occur in the first place.

214. Chris Dial Posted: October 07, 2005 at 08:29 PM (#1669384)

1954

215. and Posted: October 07, 2005 at 08:29 PM (#1669385)

If I’m right the play was in his control and there is a causal factor other than luck.

But it shouldn’t have been in his control ... and it was highly improbable that it or something like it would occur in the first place.

This may show the gap is unbrideable.  Yes, the play was unlikely and, therefore, that it happened could be seen as bad luck.  But, just as the crapshoot advocates point out, while the odds that a single unlikely thing happen are low, the odds that some unlikely thing happens isn’t.  Thus, in a close matchup, the team that is physically and mentally prepared for oddness and reacts to the unlikely situation first will often win.

Either you think that it isn’t the duty of a player, manager or team to be prepared for unlikely things or you don’t.  But I don’t see how you can’t hold players responsible for this.

216.  Posted: October 07, 2005 at 08:32 PM (#1669397)

But I would discourage someone from insisting

Ok, what am I insisting on?

That, of course, doesn’t tell us which of those “reasons” correlated to a judgment reached or a decision made by Beane, Howe, or one or more of the A’s players. Nor does it tell us—as to those reasons that were tied to such a judgment or decision—which of those judgments and decisions can also said either to have been wrong at the time, or even with the benefit of hindsight.

Some events have predessors, some can occur in parallel.

By the time I make this statement, we are close to post 200.

Apparently in this thread, where I have made 0 assessments on Billy Beane, I have a burden to produce evidence that THE A’S LOST FOUR PLAYOFF SERIES AND THE PROBABILITY THAT THEY LOST FOUR PLAYOFF SERIES IS 1.

I have to get that out of the way, before I can even fend off “luck this, and luck that”  I have to prove this thing, in which I have done like I was in an undergraduate math class.  Its absolutely necessary for me to get where I’m going, but its NOT RELEVANT compared to the rebuttal that deals with subatomic particles ontology.

And this starts while I am minding my own business and multiple people are articulating what MY argument is.

Is it so impossible to see that I am trying to get where you are going.  I was trying to four years ago, but admidst the insults and the sophism, the faux expertise and the just general lack of ability for people to understand, I can’t get there even when being super sugary nice.

Can you not read DCA’s post and see why it gets so windy.

And this isn’t any “bar, bar, bar” situation.  I’m on point, being misquoted, misattributed, and moving in a linear fashion.

Daly wants to aver its because I can’t write well.  Ok, point of you Jon.  But nobody writes well enough to trudge through this, and all the irrelevant collateral attacks.

And then your rebuttal is that I haven’t gotten to the finish line fast enough.

I may be guilty of at times presuming that everybody already knows what you are articulating.  But let me ask you, do you really think I don’t know these things.  I know you think I’m an idiot, but do you not think that I don’t see a start, finish and progression.

And do you want to hold me accountable in this forum, because I can’t keep control of what 1000 posters are going to do with the arguments, and what directions they are going to take them in.

I mean, you do see that there are even some very serious posters that legitimately believe that I think all events are within the control of the GM.

I am seperating forecasting, causality, externality, randomization, planning, responsibility and experimentation pretty damn expressly, and 80% of the rebuttals I hear are people telling me I’m confounding the issues.  And then they are using their basis as a single term called “luck” that they extend over all these disciplines and applying it at the most convenient point in the time continuum.

217. Srul Itza Posted: October 07, 2005 at 08:36 PM (#1669416)

and it was highly improbable that it or something like it would occur in the first place.

Why?  Baserunning screw ups happen on a regular basis.  They are not that damn unusual.  Was Tejada a demonstrably heady baserunner for whom such a gaffe was unthinkable?  Or was he just a young player under the pressure of the post season who suffered a total brain fart on the bases?

He screwed up.  If there is one thing that is NOT unexpected, it is that people are going to screw up, even people who are highly trained or experienced in their chosen profession, even without the pressure of a play off.

I would probably be out of a job, were it otherwise.

218.  Posted: October 07, 2005 at 08:38 PM (#1669420)

then luck did play a huge role, and could be the major underlying cause

So luck caused the event?

Again, a perfect example.  I hope your skill and experience realizes that Tejeda not running is a cause and not luck.  If not, you are losing your expert credentials pretty fast.

My point is that it is extremely difficult to know if it is luck that placed the pebble in front of the ball

Luck doesn’t place pebbles in front of balls.  Pebbles are in front of balls.  Whether you can control whether the pebble is in front of the ball is another question.  Whether you should exercise control on the pebble is another question.  Whether you are responsible for the pebble is yet another question.  Whether, independent of the pebble, you are still responsible for the final outcome is a fourth seperate question.

219. Chris Dial Posted: October 07, 2005 at 08:38 PM (#1669425)

luck is defined as a combination of circumstances, events, etc. operating *by chance* to bring good or ill to a person (team).

Luck is a word and it has meaning that allows the usage as it is being used.

If the events that are causal are of sufficiently small enough chance, then I think it is rightly called “decided by luck”.

Now the goal, as I’ve agreed, is to reduce the issues that allow “luck”.

Unfortunately, in playoff baseball, both teams are good teams (say the ChiSox and Bosox).  Both hit well, both field well.

Both teams have eliminated as much luck as *YOU OR I* can identify right now.  Both *seem* to have enough pitching and hitting.

What both teams have assembled already have the “luck” factor down to tenths of a percent.  Very little, at this point is left to chance.

What happens now - Iguchi gets hot (baseball is a game of streaks) - Ortiz goes cold (for no apparent or controllable reason) -

Both teams are presently even - who wins now is *luck*.

Which team has the player that gets hot (the 2002 Adam Kennedy/Troy Glaus) is hte team that wins (or the most players that get hot/least players that go cold).

Teams simply don’t “limp into the playoffs” in an identifiable manner (Padres excepted).

These teams already have the game managed down to tenths of a percent for “luck”.  So the scale tips on luck.

220. Chris Dial Posted: October 07, 2005 at 08:41 PM (#1669434)

THE PROBABILITY THAT THEY LOST FOUR PLAYOFF SERIES IS 1.

What is the probability they will lose the next four?

221. Srul Itza Posted: October 07, 2005 at 08:42 PM (#1669438)

where I have made 0 assessments on Billy Beane

Except to disparage him as an author.  Or a landlord.  I am not sure which.

222. 185/456(GGC) Posted: October 07, 2005 at 08:42 PM (#1669440)

Daly wants to aver its because I can’t write well. Ok, point of you Jon. But nobody writes well enough to trudge through this, and all the irrelevant collateral attacks.

Anyways, I’m getting out of here for a few, but maybe I’ll catch up on this thread later.

223. strong silence Posted: October 07, 2005 at 08:49 PM (#1669466)

But I don’t see how you can’t hold players responsible for this.

Bunyon,
Players are taught to play until the ref calls the whistle. Maybe Beane couldn’t afford a running coach and so Tejada didn’t know this. 99% of all High School players are taught this.

224. strong silence Posted: October 07, 2005 at 08:53 PM (#1669481)

<i>My point is that it is extremely difficult to know if it is luck that placed the pebble in front of the ball

Luck doesn’t place pebbles in front of balls. Pebbles are in front of balls. i>

Nit.
Of course I know that.
I have to type fast because my boss might see me.

And I agree. The players who removes the pebble is removing luck and therefore has control.

225. John Lowenstein Apathy Club Posted: October 07, 2005 at 08:53 PM (#1669482)

Ooh, this is fun.  Now I’m writing an article on World Series upsets.  I get all my good ideas from Primer these days.

226. DCA Posted: October 07, 2005 at 08:55 PM (#1669487)

Players are taught to play until the ref calls the whistle. Maybe Beane couldn’t afford a running coach and so Tejada didn’t know this. 99% of all High School players are taught this.

That’s a good point.  Tejada was not a American high school player, he may not have learned this lesson the same way that we have.  Certainly, with foreign players, I don’t think we can assume the same things are obvious to them as are obvious to us.  This should be, but still ...

227. Chris Dial Posted: October 07, 2005 at 08:55 PM (#1669492)

Where do you write these articles?

BTW, I didn’t know who JLAC was somewhere else, and you said something where I went “Okay, now you are too smart for me not to know who this is.”

Oh, it’s just dumb old Craig…

228. Mefisto Posted: October 07, 2005 at 08:56 PM (#1669503)

Ok, what am I insisting on?

As I interpret your posts—and I could be wrong again—you are “insisting” that teams are responsible for their losses. Period. Full stop.

I’m saying they are not, at least not necessarily. I say that because I think teams (and people generally) are only responsible for something to the extent they have the ability to change it. Thus, in terms of the playoffs, GMs or managers are responsible for the losses ONLY IF they could have predicted the cause of the loss and prepared for that eventuality.

Just to be clear, I don’t believe you have to prove the predictable/prepared prong (love the alliteration). I do believe you have to provide actual evidence of it, though. It’s not enough to generate hypotheticals, e.g. (post 200) that “indifference, laziness or carelessness” could have been involved in the pebble that cost the Giants the ‘24 Series.

Pretty much anything “could have been”. There has to be a factual showing that a reasonable person could have foreseen the bad hop and prepared for it. If a reasonable person could not do so—and I strongly believe he/she could not—then it makes no sense to me to say that that person is “responsible” for it.

I have a burden to produce evidence that THE A’S LOST FOUR PLAYOFF SERIES

That’s not my position, and I don’t believe that anyone else has taken it either. Speak up boys.

We all agree that they lost. The issue, which I address above, is whether they are responsible for that fact.

Can you tell from all this which side of the Palsgraf case I’d be on?

229. John Lowenstein Apathy Club Posted: October 07, 2005 at 08:57 PM (#1669506)

So is it wrong of ballplayers to be superstitious?  What I mean to say is, are they wrong to think that luck plays a major role in what happens on the field?  (Not whether superstitious behaviour can control luck, which is another question entirely).

230. strong silence Posted: October 07, 2005 at 08:57 PM (#1669511)

What is the probability they will lose the next four?

I will go out on a limb and say the White Sox will be least affected by luck this postseason. IOW, luck will least likely be a factor in their victories and losses.

231. strong silence Posted: October 07, 2005 at 08:59 PM (#1669518)

What is the probability they will lose the next four?

I will go out on a limb and say the White Sox will be least affected by luck this postseason. IOW, luck will least likely be a factor in their victories and losses.

232. John Lowenstein Apathy Club Posted: October 07, 2005 at 08:59 PM (#1669520)

Where do you write these articles?

In my office at home.

233. Chris Dial Posted: October 07, 2005 at 09:00 PM (#1669526)

Where do you publish them?

234. John Lowenstein Apathy Club Posted: October 07, 2005 at 09:01 PM (#1669529)

The issue, which I address above, is whether they are responsible for that fact.

Well, I’d say that they are responsible by definition.  This is the Parliament of baseball, and when the department screws up, the Minister is responsible.

235. Greg Maddux School of Reflexive Profanity Posted: October 07, 2005 at 09:01 PM (#1669531)

To me 1990 was not an upset. One of five teams to lead wire to wire - that is strong evidence of a special team.

Who gives a fuck that they went wire-to-wire?  The only difference between the Reds and any other 90-win team is that they happened to win their first few games.  That makes them special?

236.  Posted: October 07, 2005 at 09:02 PM (#1669534)

BL = Joe Morgan? Who knew.

I didn’t figure you would be the one that took the bait, Srul.

I think it stands on its own.

Which one, that you can’t use results to determine if there are failings?  That ain’t standing to well.

Again in my meager, petty little background, I was taught on my first day in this little out of the wall place that at least one poster will be familiar with that :  models < simulations < real world results.

So I’d like to apply this theory.  Let’s say there was some looney profession like a forensic engineer.  He goes and he sees a crack in the buildings foundation that will cost three million dollars to repair.  It also lead the building unusable by a tenant who lost another two million dollars for lack of use.

By your rules, he can’t test the buildings foundation or look at the crack.  He must go back and look at the plans, then simulate the results, and find out the probability of the crack based only on that information.  He can’t look at the actual construction, he can’t examine the crack and the underlying foundation.

Then, he has to use those results only.  If he finds a 51% likelihood that a crack would result, he can then conclude the building does have a crack.

Now that is the loony part.  But lets get to a practical part and wet Srul’s appetite.

He MUST find from that information alone that there was a 51% chance of a crack to conclude that there was a design defect, and responsible behavior on the part of the building designer.  He MUST find from that information that there was a 51% chance of a crack to conclude the building owner has any responsibility to any party that was injured resulting from that crack.

If it is 50% or less, than the causation of the crack is LUCK and no person has any responsibility to anyone else.  It wouldn’t matter if the industry standard was reducing the crack potential under 1% due to the use of a concrete that cost only 5% more than the concrete that gave you a 49% chance.  As long as we are under 50%, its a crapshoot.

And what if the industry knows that you have to protect against sheer stress and comperssion stress.  The concrete the designer called for was 5% cheaper and gave you better sheer protection, but just sucked monkey nuts for compression stress.  Was the designer exploiting a market inefficiency in sheer stress?

What if on his next building, he didn’t want to make the same mistake the compression stress.  So he goes out and buys some concrete that costs the same as the industry standard concrete.  It offers better compression protection but sucks at sheer stress.  It still puts the likelihood of cracks at 39%.  Wow, was that the market recognizing the deficiency in sheer stress, and now has the designer made a genius move by exploiting the market for sheer stress.  Because its still more likely than not he want get a crack.  If he gets one, is it just another crapshoot where he can avoid all liability.

I’m interested.  I’ve know seen a definition that says if anything unlikely to happen does happen its the result of the luck fairy.

So this whole little Voros escapade has taught us something.  Standards in the industry, or what you saberists call rote following of idiotic conventional wisdom due to risk averse industry, has no bearing on responsibility or liability or whether something is a crapshoot.

And that is true even if that standard of the industry is right, and your spreadsheet is wrong, like pitchers influencing hits on balls in play.

So its not the Red Sox fault they didn’t have a closer two years ago, that was luck, we should have all known that bullpen-by-committee should work, and we can’t argue with it because it still works in that simulation based on everything THOSE ANALYSTS knew at the time, (even if not what the industry knew).  And the fact that the Sox grabbed DIPS darlings like Rupe.  No fault there, that was just luck too.  DIPS says Rupe should have dominated.

And those little probability tables tell me the A’s should have won those series.

237. John Lowenstein Apathy Club Posted: October 07, 2005 at 09:02 PM (#1669536)

Publish?  That would imply that I ever finish anything.

Seriously, I’m still writing (sporadically) for The Hardball Times.

238. Chris Dial Posted: October 07, 2005 at 09:06 PM (#1669547)

JLAC,
I sincerely regret that we live so far apart.

I wish you lived in my neighborhood.

239. John Lowenstein Apathy Club Posted: October 07, 2005 at 09:08 PM (#1669555)

I wish you lived in mine too, my man.  I’m just happy to say - following GGC’s lead - that my Dial number is 1.

240. DCA Posted: October 07, 2005 at 09:09 PM (#1669557)

What is the probability they will lose the next four?

What interests me more is the question, if they played again, given the same situation, what is the probablility that would lose all four series?

It’s not 1.  I believe it’s probably on the order of 1/16, adjusted slightly by quality of opposition, details of matchups, and home field.  I don’t think there is any structural failing that would make it much higher than that, but I am open to the possibility that this is the case, if someone would hypothesize such a factor.  I think it is highly improbable that there is a structural reason that it is much lower.

241.  Posted: October 07, 2005 at 09:12 PM (#1669568)

Real WS upsets:

1906, 1987, 1988

Mild upsets:

1914, 1969, 1979, 1985, 1990

Illusionary upsets:

1931, 1942, 1946, 1954, 1960, 1963, 1972, 2001, 2003

A lot of AL losses in the 1950s and 60s were seen as upsets at the time, because people were used to the Yankees winning every year, and NL teams with inferior records won in ‘54, ‘59, ‘60, ‘63, and ‘64; but the NL was truly the stronger league at the time ...

Absolutely.  Those “upsets” only showed that the oddsmakers couldn’t see the forest for the trees.

242. Chris Dial Posted: October 07, 2005 at 09:15 PM (#1669576)

but the NL was truly the stronger league at the time ...

I still don’t buy this.  The AL won too many WS for that to be the case.

243. DCA Posted: October 07, 2005 at 09:16 PM (#1669577)

BL,

Your example isn’t “probable” to use your word.  The main reason is that it’s possible for my building to not crack and for your building to not crack.  This is not true in baseball.

If this weren’t the case, then there might not be anything wrong if my building cracked, and there might be better ways to determine who’s the better engineer that who’s building happened to crack.

244. DCA Posted: October 07, 2005 at 09:16 PM (#1669582)

d’oh, I meant “probative”.  “probable” is one of my words.

245.  Posted: October 07, 2005 at 09:22 PM (#1669611)

you are “insisting” that teams are responsible for their losses.

As I’ve said in a series of posts, I can’t even get close to really doling out responsibility.

What is expressly clear is that I have said no such thing IN THIS THREAD, to where you have expressly asked me for evidence.

But, FWIW, and this will cause things to get tangential again.

A GM is responsible for the performance of his team.  PERIOD.  How that responsibility translates to an evaluation of his performance is something that its difficult to get close too.

Now if you want to read between the lines on that gander at 237, because you are going to look for my endgame regardless of whether I state it, and you are going to fill in the gaps anyway.  I’ve disclosed enough in 237 you might be able to get there without filling the gaps by attributing error prone arguments to me.  I know you are bright enough to see it; a few others might too.  I just hope everyone doesn’t take the same invitation.

We all agree that they lost. The issue, which I address above, is whether they are responsible for that fact.

Oh goody goody.  So what is the test Mark.  Because now your resonating on my frequency, and we can have this discussion.

Upthread someone said any decision that does not result in an adverse result more often than not.  Do you agree with that.

How about industry standards.  I’m sure some people think ol’ Backlasher is pulling things out of his asss on that one.  Do you Mark.

Remember, the claim to fame is that he’s bucking the system.  So how do you hold a Maverick accountable.  Do they get a free pass on everything they do different as long as THEIR MODEL says its less than 50% likelihood.  That sounds more like a mens rea test to me.  Is that what we use to define “negligence”  And do I need, negligence, I mean simple malfeasance would suffice wouldn’t it.

And then, here’s the kicker.  If you expressly state that someone is ABOVE AVERAGE.  Do you even need malfeasance to make your point.

So what is my evidentiary burden to show a flaw, a flaw that moves.

I’ve expressly stated my hypothesis, and I’ve expressly shown evidence.

YOU CAN"T COMPETE LONG TERM ON A LOW PAYROLL, YOU MUST INVEST, AND NO LITTLE SPREADSHEET TRICKS WILL CHANGE THIS FACT.

246.  Posted: October 07, 2005 at 09:26 PM (#1669619)

Your example isn’t “probable” to use your word. The main reason is that it’s possible for my building to not crack and for your building to not crack. This is not true in baseball.

LOL.  So let me get this straight.  You analysis of when you cannot look at output only deals with zero sum events.  Is that your new position.

Mr. Itza, is this how things work in securities regualation for futures?  (I’m going to ask an expert rather than bore you with my uninformed analytical background)

And my analogy is very probative.  I’m sorry if you don’t see why.

247. DCA Posted: October 07, 2005 at 09:28 PM (#1669621)

Who gives a #### that they went wire-to-wire? The only difference between the Reds and any other 90-win team is that they happened to win their first few games. That makes them special?

Thank you.  That needed to be said.

248. DCA Posted: October 07, 2005 at 09:36 PM (#1669633)

LOL. So let me get this straight. You analysis of when you cannot look at output only deals with zero sum events. Is that your new position.

LOL too.  Buildings are not supposed to crack, so we have codes and theory and standard practices that are designed to ensure the vast majority of buildings—ideally all—do not crack, or get damaged in such a way that they are unsafe or unusable.  Actually, it’s a lot more complicated than that, and cracked concrete isn’t necessarily bad, but that’s a good quick summary.

If you think that it’s possible to assign responsibility for failure and draw conclusions from failure in an arena where failure can be made rare, and theoretically prevented entirely in the same way as in an arena where failure is guaranteed for all but one partipant, instant failure for 50% of participants, and repeated failure for a smaller but still significant group, then I don’t know what to say, except that I’m glad that you’re not running a baseball team or a municipal building department (though I’m not sure which, and you might be fine at the other).

249. DCA Posted: October 07, 2005 at 09:44 PM (#1669657)

Also, BL, please don’t project every disagreement you’ve had with Treder, Voros, etc onto me.  I’m not them, I don’t agree with them all the time, I’ve never met them, and I wouldn’t defend half the arguments you’ve assigned to me.

250. Chris Dial Posted: October 07, 2005 at 09:46 PM (#1669661)

YOU CAN"T COMPETE LONG TERM ON A LOW PAYROLL, YOU MUST INVEST, AND NO LITTLE SPREADSHEET TRICKS WILL CHANGE THIS FACT.

251.  Posted: October 07, 2005 at 09:46 PM (#1669663)

... odds of losing the next four in a row

Based on available information the odds would be at about 6.25%.  Because I can’t calculate years that they will actually make the playoffs, what the organizational strategy will be, and who they will be playing against.

Now what do you want to add to the equation:

That the exact same team that played this year without any improvement or decline in the exact state they finished this year.  I doubt that team is going to make many playoffs.  But if you do want that:

Somewhere north of 6.25%.  If you want an exact number, here is the best way you can get it.  Take the roster that was at the end of the year, take away all the injuries, and sim 4 playoff series runs against 100000000 past champions and see how often it comes up with 4 losses in a row.  The best engine would be diamond mind, and that want even count the chill of players like Oakland Vice Dan Johnson.  If you really want to see the final number, we can hire a clerk to run the SIMs because that in no way seems fun to me.

Now how about freezing their payroll.  Do the same experiment as before, except start lopping off your best performers with someone that gives you a mean average of 67% of their output.  Again, I think its going to be tough to even make the playoffs.

And if you hire Danny or something to run those numbers, please god tell me what good they do you.  That seems to be more interesting to you and DCA than examining what the actual problems were with the past teams.  Is there anything at all you can do with that, other than looking for a sucker to make a bet.

And remember, you can’t say you could learn X, because DCA won’t let me do a casaul analysis after I have data.

Forecasting is a droll dick measuring exercise by most of its practitioners, and it only has value if you are updating your engine based on new information.

If not, anybody can just look up Voros probability table.  There is nothing to be gained by keeping this at a sophmoric level.

252. 185/456(GGC) Posted: October 07, 2005 at 09:51 PM (#1669683)

Absolutely. Those “upsets” only showed that the oddsmakers couldn’t see the forest for the trees.

So the betting market in baseball was inefficient back then (I’m not suggesting that markets are ever 100% efficient)? I’m not that familiar with baseball gambling from that era or any era, really (I don’t share Gene Carney’s interest in the Black Sox for whatever reason), but I am curious.

253. DCA Posted: October 07, 2005 at 09:56 PM (#1669694)

And remember, you can’t say you could learn X, because DCA won’t let me do a casaul analysis after I have data.

WTF?  I won’t let you use the fact that a team lost as </i>proof that particular team has a structural failing that causes them to lose unless you arrive at the hypothesis of structural failing without prior knowledge of the results.

I would not, ever, say that you couldn’t use the fact that they lost to motivate<i> a causal analysis, and I welcome you to come up with structural reasons why they might have lost.  And I am absolutely open to believing that there is such a structural failure, I just am not convinced that there is one, because the observed outcome is consistent with there being none.  That doesn’t mean there is none, and I have never stated conclusively that there is none.

254.  Posted: October 07, 2005 at 10:05 PM (#1669713)

If you think that it’s possible to assign responsibility for failure and draw conclusions from failure in an arena where failure can be made rare, and theoretically prevented entirely in the same way as in an arena where failure is guaranteed for all but one partipant, instant failure for 50% of participants, and repeated failure for a smaller but still significant group, then I don’t know what to say, except that I’m glad that you’re not running a baseball team or a municipal building department (though I’m not sure which, and you might be fine at the other).

You would be thoroughly amazed at what I do and what I have done.

But, again, the reason that you are lost, even with your vast experience, is that you don’t even recognize the event space.

If you ignore a standard in the industry, lets say LIKE DEFENSE or PITCHERS ABILITY TO CONTROL HITS ON BALLS IN PLAY, and you make decisions based on these failures, you have already shown shortcomings in your duties.

What happens in the game is just the measure of your damage.  If you win anyway, there is no damage.  If you lose, you have created damage.

Its games with probability.  Its what you all do, you define spaces where you will always have the excuse of luck.

Don’t you realize that it is impossible to get a space on the preformance of the 2001 As in playoff series based on real world results.

And I don’t care what Voros, Steve, etc. have done, you are doing the unpardonable here.  Anytime that anyone does try to give you a space or a method to analyze the results YOU MOVE EVERYTHING RIGHT BACK TO A SPACE to where you can’t make an affirmative argument, you just say WELL IT COULD HAVE BEEN LUCK.

Right back to the same place every single fukking time like a homing pigeon.

And along the way, you are talking a lot of smack about your credentials and making facially absurd statements about how and when you can analyze things.

And your basis is supposedly this depth of experience in analysis and your presumptions about my training and CV.

And you have a few people on your train.  Dial keeps saying over and over again that the only important thing is forecasting.

And it appears to me you can’t determine an appropriate space selection or seperate the building of a plan algorithm from the projection algorithm.

And what is even more absurd is that when someone challenges a value component of your algorithm, you want to prove it correct by using that algorithm to forecast a result, and then use that result as the basis for experimentation and proof rather than a real world result.

And no matter how many times I explain it, you can’t grasp that this does not mean or is not related to what your binomial distribution with no INFORMATION may forecast.

I mean you have this vast training.  I have seen you name drop Bayes before.  I presume your experience tells you about the value of information, and you can do a few conditional probabilities.

255.  Posted: October 07, 2005 at 10:08 PM (#1669716)

Buildings are not supposed to crack

Players are suppose to slide into home on close plays.

Players are not suppose to stop running on live balls.

We have rules that are designed to call them out to prevent these occurences.

256.  Posted: October 07, 2005 at 10:15 PM (#1669737)

proof that particular team has a structural failing that causes them to lose unless you arrive at the hypothesis of structural failing without prior knowledge of the results.

What is your key word PROOF.  Who has offered that as PROOF.

Did you not see the 4 times that I expressly stated that I laugh when people use the excuse IT WAS BOUND TO HAPPEN.

Because they lost, they failed to reach an objective.  Now, try just for once, not to read anything into that statement—there are no words about responsibility or causation.  Its just a true statement; its pretty close to FOL.  The big problem is that people will jump all over the place for a very simple statement.

Because they failed to reach an objective we should search for why.  Again look very closely, no words of responsibility in this statement.

Now go back to my earlier post where I very expressly tell you how to identify a systematic failing.

So what do you do, you post this same argument five seperate times.  You make up a position for me. Ignore it when you are expressly told I have no such opinion.  Ignore it when I expressly tell you what is being done.

And keep posting the same argument like you have found some logical hole that decimates all arguments regarding failures.

257. Srul Itza Posted: October 07, 2005 at 10:29 PM (#1669779)

LOL. So let me get this straight. You analysis of when you cannot look at output only deals with zero sum events. Is that your new position.

Mr. Itza, is this how things work in securities regualation for futures? (I’m going to ask an expert rather than bore you with my uninformed analytical background)

Actually, we don’t get many cases involving the futures markets.  We represent people who have been ill used by their fincial advisors and brokerages.  They tend to be people who are not sophisticated.

Yes, we have cases involving trading in stock options, but not the futures markets, which is very different.  Those are far more sophisticated markets, for people who really know what they are doing, and who have money they can afford to lose.

If any of the kind of people I represent had been placed in the futures markets by the kind of people I arbitrate against, I would expect to receive a hefty settlement offer, because they would have no plausible excuse for involving them in that kind of risk.

258. DCA Posted: October 07, 2005 at 10:34 PM (#1669796)

BL—

What is your key word PROOF. Who has offered that as PROOF.

If that is indeed the case, that you have not been offering the as proof of structural failing, then I am sorry that I inferred that meaning from your tone, here and in previous threads.

Because they failed to reach an objective we should search for why. Again look very closely, no words of responsibility in this statement.

That’s true.  However, in the past you have stated, and I paraphrase, that if you were Beane’s employer, you would demand a better explanation that “it’s a crapshoot”—there seems to me an undercurrent of “Beane is responsible” or at least “some structural failing is at work.”  If here and in other statements you have not meant to imply that Beane is responsible, then I am sorry that I have misunderstood you.

And if you could point out the post # where you very expressly tell me how to identify a systematic failing, I’ll go back and read it more carefully.  Hopefully it’s something like notice, analyze, theorize, test, repeat.

259. DCA Posted: October 07, 2005 at 10:42 PM (#1669822)

I mean you have this vast training. I have seen you name drop Bayes before. I presume your experience tells you about the value of information, and you can do a few conditional probabilities.

Do I need to prove to you that I understand?  Let’s say I have two teams that I think are evenly matched, the A’s and the B’s.  The B’s then beat the A’s.  Depending on the amount of prior information I have, I now believe it’s more likely than not that the B’s are better than the A’s.  However, if my hypothesis was that the event is a crapshoot and each team is equally likely to win, that B one and A lost is not evidence against that hypothesis.  These two things are not mutually exclusive.

With more complexity, and “a crapshoot” to “basically a crapshoot” since the teams are not quite equally matched, that’s what I think the playoffs have shown.  My assessment of the A’s based on their 2001-04 regular seasons would be one of the best 2-3 teams in baseball for sure.  Because they lost those playoff series, I think they’re not quite as good, clearly deserving of a playoff spot but a notch below the top teams.  Yes, I can incorporate additional information.  However, as a whole, the playoffs appear to be consistent with a crapshoot, and so I don’t think that the results of the playoffs, including but not limited to the four series the A’s played, does not suggest that the playoffs are not basically a crapshoot.

260. DCA Posted: October 07, 2005 at 10:45 PM (#1669838)

In that last sentence, it’s supposed to be a double negative, not a triple negative.  Drop the don’t.

261.  Posted: October 07, 2005 at 10:45 PM (#1669843)

Absolutely. Those “upsets” only showed that the oddsmakers couldn’t see the forest for the trees.

So the betting market in baseball was inefficient back then (I’m not suggesting that markets are ever 100% efficient)? I’m not that familiar with baseball gambling from that era or any era, really (I don’t share Gene Carney’s interest in the Black Sox for whatever reason), but I am curious.

More likely the explanation in the 1950’s was that (a) the Yankees were good; (b) they’d won every year from 49 to 53; (c) it took awhile for the public to catch onto the NL’s growing superiority; and (b) a disproportionate percentage of the betting action then was centered in New York, which would influence the odds in the Yanks’ favor.  The only efficiency of the betting market is to provide odds which produce roughly equal amounts of money bet on each possible outcome.  I’m sure that the Yankees being the public choice influence the odds in their favor to a greater extent than their actual talent would have warranted.  This was particularly true in the late 50’s and early 60’s.  It’s not that they didn’t deserve to be favored, only not by as much as they were.

The 1954 Indians were big favorites based on their 111 wins and the recent AL Series dominance, which was in fact Yankee dominance.  IMO that Series was the biggest Phantom Upset in history, given that the Indians had racked up that record by going 89-21 against 5 teams which were barely “Major League” in terms of talent; against the Yanks and White Sox they’d gone 22-22, and against the Giants in Spring Training they’d been whomped consistently over a long series of Cactus League games—no small sample size there.

Oh, and P.S.—-How bout El Duque!

262. 185/456(GGC) Posted: October 07, 2005 at 10:55 PM (#1669877)

Oh, and P.S.—-How bout El Duque!

We’ll see this inning.

263. Mefisto Posted: October 07, 2005 at 11:06 PM (#1669918)

The AL won too many WS for that to be the case.

You misspelled “Yankees”.

A GM is responsible for the performance of his team. PERIOD.

Ok, I think that’s what I said too, so I believe I’ve understood you.

I disagree. To me, the correct way to state it is: “a GM is responsible for the performance of his team” only to the extent that he can affect it.

I’m not going to get into the details of 237 despite your invitation because I think it takes us too far into arcane legal arguments (especially because CA law on construction defects is intricate. To be polite.). I’ll just say this: I’m arguing that the proper standard is negligence, and I’m disagreeing on anything beyond that.

Upthread someone said any decision that does not result in an adverse result more often than not. Do you agree with that.

Not really. That’s too simple a test for analyzing strategic decisions. I understand it for tactical moves (hit and run, SB, sacrifice, etc.), but I don’t know what it means in, say, choosing between Dave Roberts and Kevin Millar.

How about industry standards. I’m sure some people think ol’ Backlasher is pulling things out of his asss on that one. Do you Mark.

No, I don’t think you’re pulling that out of your ass. However, in a debate about what the standard should be, the existing standard can’t serve as the touchstone.

And since I think negligence is the standard, I continue to believe that you have the burden of providing some evidence that the alleged misfeasance proximately caused the damage.

264. Mefisto Posted: October 07, 2005 at 11:22 PM (#1669968)

I need to modify the last sentence of 264. It should read:

...I continue to believe that you have the burden of providing some evidence that (a) there was misfeasance, and (b) the alleged misfeasance proximately caused the damage.

265.  Posted: October 07, 2005 at 11:38 PM (#1670029)

Beane’s employer, you would demand a better explanation that “it’s a crapshoot”—there seems to me an undercurrent of “Beane is responsible” or at least “some structural failing is at work.”

Yes, yes, and possibly.

I would demand a better explanation.

Beane is responsible.  Whether or not that means he has poor perfomance is another question.  And when I repeatedly state that based on the Information that he is an above average GM, I don’t see how so much negativity is involved.

And yes, I have posited there is one systematic failing - the failure to invest.  And I have stated the solution to that problem does not lie in the techniques that are expressly shown in “Moneyball”  There is no real evidence that Beane has exploited market inefficiencies when looking at the entire player.

And frequently, I am met with the rebuttal, “he only has so and so payroll” which doesn’t read on any of those points.

And if you could point out the post # where you very expressly tell me how to identify a systematic failing

198(4)

Do I need to prove to you that I understand?

No, I know you are brighter than the mass of fanboys, and I have not put you in there group.  But there have been multiple times that I have had to tell you I don’t believe something.

And that is common.  Contrary to Starbucks Boys delusions of grandeur, I could care less about his pithy little articles.  I really didn’t pay attention to him until every college freshman who wanted to be part of the next big thing started polluting every conversation with “What’s his DIPS.”  I just had somebody out here representing what was identified as MY argument.  I’m actually a little flattered because it was a smart guy and he was giving me credit where he didn’t need to give credit.

He could have been like Treder and just co-opted it and claimed it for his own.  And then I get the standard, “I agree with Backlasher on that, but not his point X.”  And then I get mystified.  When has Backlasher said X.  I am wondering about these radical views I have:

(1) Don’t coerce others to use drugs
(2) Don’t hit people with chairs
(3) A pitchers influence on a batted ball is a skill
(4) Reliever usage is more optimal than the past
(5) Luck is not a catch all excuse

Let’s say I have two teams that I think are evenly matched, the A’s and the B’s. The B’s then beat the A’s. Depending on the amount of prior information I have, I now believe it’s more likely than not that the B’s are better than the A’s. However, if my hypothesis was that the event is a crapshoot and each team is equally likely to win, that B one and A lost is not evidence against that hypothesis. These two things are not mutually exclusive.

And where and who do you think disagrees with this assessment.  Its where this assessment gets taken, sometimes in article posts that is the problem.

A and B aren’t evenly matched.  Considering that B’s objective is not only to beat A, but to beat C and D, its a good idea to figure out if they have a shot.

And the playoffs aren’t the regular season.  Thank God and Mark Field for taking this to sub-atomic particles, because if it didn’t get that complex from the gate, we would have spent 300 some posts explaining why those two environments are different.

And A is not a static entity, and probability is not transitive.  Its possible to have P(A over B) > P(B over A) > P(B over C) and have P(A over C) < P(C Over A).  Moreover P(A over B in June) <> P (A over B in October.

Somebody back in the first 50 posts posited actual evidence of a 20 percent swing in those probabilities.  Dial thinks its a tenth of a point and offers no evidence.  (And Field jumps on me for just throwing out hypothesis without evidence, and I have stayed strictly mathematical in this entire thread.)  Considering that its pretty unbounded on the probability swing, I think the working hypothesis should be enough swing that we better investigate.

And that is the major problem, you get to a reasonable point of investigation or suspicion and every body curls up in a little snake posture ready to defend their icons.

And its all posture.  It starts “I believe…” with no support, then it goes right to Burden is on you to prove x and a massive mischaracterization of arguments.

But more important, do you think P(A) = P(A|B) is a true (not supportable, but true, so stay with me) statement.  So if I got B, why the he11 do I want to calculate P(A), unless iff B was unknowable in all circumstances at time t, the decision point, and I’m trying to evaluate performance based on the failure to receive the desired result.

Again, I don’t know what your expertise is allegedly in, but this is a simple fact.  There are real world systems that if you don’t produce result A, its a failure.  It doesn’t matter if you are the best, worst or somewhere in the middle.  Its a failure.  And yes, there are plans you evaluate based on accumulations or percentage of failures.

If that offends your since of equity, I’m sorry.  The world sucks. If you have had no such experience with such systems, then the question may need to be broached as to what systems you are analyzing.

And if you want to identify baseball, its not a low functionary system to measure effort.  Its not a drug dipensing machine either.  The GM role is somewhere in the middle.

And the Beane supporters argue both ends of the spectrum.  “BAR BAR BAR, that is too complex, I’ll just go with the eay little tables Voros uses and call it a crapshoot.”  and then out of the other side of their mouth, “BAR BAR BAR Well you have to evaluate what another GM would have done.”  Instead of focusing on things that should be accurate and quantifyable, they just jump all around for excuse generation.

And its not like there is any body to frame the issue under discussion.  Its not like you can get any direct answers.  You just cycle and repeat to have another well reasoned evaluation to which the unwashed masses will revolt, ITS JUST LUCK.  Or you might have a little kindergarten report from Voros or Davis, to which the saberists will revel as THE NEXT BIG THING.

I mean I’m pretty curious.  Field and VOU turned me to an interesting topic, that is why this is still worthwhile.  But do you think there was anything in that story.  No news, no analysis, nothing.  Its the classic Davis/McCracken fun with numbers.  Let’s see if I can devise a system to make something look good.  Its not even close to being probative or educational.

266.  Posted: October 07, 2005 at 11:43 PM (#1670053)

“a GM is responsible for the performance of his team” only to the extent that he can affect it.

A GM is evaluate on his performance only the extent of things (s)he can affect.

(S)He is very much responsible.

And that statement is very, very different than the one you attribute to me.

267.  Posted: October 07, 2005 at 11:45 PM (#1670059)

continue to believe that you have the burden of providing some evidence that (a) there was misfeasance, and (b) the alleged misfeasance proximately caused the damage.

Really, you can only negatively evaluate an employee if there is misfeasance.  So as long as they don’t miss a SOL, your clerks and associates get perfect performance evaluations?

268.  Posted: October 07, 2005 at 11:45 PM (#1670062)

Oh, and P.S.—-How bout El Duque!

We’ll see this inning.

Pedigree shows.

269. Mefisto Posted: October 07, 2005 at 11:45 PM (#1670063)

Well, at least we all can agree that Theo is not sabermetric this year.

He’s still dreamy though.

270.  Posted: October 07, 2005 at 11:46 PM (#1670068)

Finally, let’s try that misfeasance standard on

Chuck LaMar and Allard Baird

271. Mefisto Posted: October 07, 2005 at 11:54 PM (#1670083)

A GM is evaluate on his performance only the extent of things (s)he can affect.

(S)He is very much responsible.

Well, if you’re making a distinction between “things for which s/he has responsibility” (everything) and “the things for which s/he’s evaluated” (only that subset of “everything” which the GM can affect), I understand you but I wouldn’t phrase it that way. To me it makes more sense to equate the two: one gets evaluated only for the things for which one is responsible.

Really, you can only negatively evaluate an employee if there is misfeasance.

As long as “misfeasance” includes omissions, then yes.

272. DCA Posted: October 08, 2005 at 12:00 AM (#1670096)

I think there’s enough evidence to convinct LaMar and Baird of misfeasance.  LaMar seems to have already been convicted.  And when Baird’s biggest coups have been a single good year of Angel Berroa for future star Ellis (seriously, his season is like Brian Roberts’ only backward and only 90% as good) and current star Damon, and one good year of Zach Greinke, while squandering Dye, Damon, and Beltre, he’s looking at long odds of acquittal.  And no solid seasons between them.

273.  Posted: October 08, 2005 at 12:04 AM (#1670099)

Well, if you’re making a distinction between “things for which s/he has responsibility” (everything) and “the things for which s/he’s evaluated” (only that subset of “everything” which the GM can affect

Didn’t I expressly state that many, many posts ago.

As long as “misfeasance” includes omissions, then yes.

No efficiency, no book, no determination of ability.  Wow.  Let me know how to get in touch with you, I’ve got students that I can send to the Perfect Firm.

274. Sam M. Posted: October 08, 2005 at 12:24 AM (#1670149)

(5) Luck is not a catch all excuse

There are very, very good reasons not to lean on luck as a “catch all excuse.”  It can prevent you from looking twice, three, four times to figure out what your organization might have done better.  It can delude you into thinking you have fewer changes to make than you really do.

But I just want to add that there are also very good reasons not to ignore the role that luck does play in producing outcomes.  Giving it the proper weight can prevent you from making knee-jerk changes that actually end up setting you back, changes that react too heavily to outcomes that really were NOT attributable to your design, your decisions, or your personnel.  Sometimes, you have to trust your judgment that you are on the right track, and you’ve got the right pieces in place, even if the outcome of the one game or the one series wasn’t favorable.

275.  Posted: October 08, 2005 at 01:04 AM (#1670245)

Sometimes, you have to trust your judgment

And how does one do that naturally in a real time environment.

Is it by setting goals and milestones? or is that too radical an idea.

And when you don’t reach those goals and milestones, what does one do?  I would think one would evaluate his plan.

And then it depends on who we are talking about for evaluating this meta-plan.

If its ol Billy Barue, we have to give him some credit.  He at least realized that fatfukk softball players weren’t going to work.  He adapted and he moved on.  Now there were plenty of people that could and did try to tell him that, but we were too busy being called idiots for anyone to notice.  Then he learned that he needed relief pitching.  Now he’s hording relief pitching and defense without improving his offense.  I am happy to tell him that is a mistake, but that is just a cue for Danny to google up runs allowed and talk about how prospect 279 will hit 40 homers next year.  If I’m lucky he’s on Columbus Day vacation.

Now if its the fanboys, no credit at all.  They defended fatfukk softball players to their dying breath, and then when Beane abandoned the strategy, they talked about some super secret metric.  That’s not evaluation, that is spin and rhetoric.

276. Sam M. Posted: October 08, 2005 at 01:36 AM (#1670322)

And how does one do that naturally in a real time environment.

I was thinking in terms of the post-mortem that every team does in the off-season.  For a team that had enough success to make the post-season, the evaluation should neither overreact to, nor ignore, the results of its play-off series.  It may be a nice mental exercise—and even a useful evaluative device—for the team to adopt the posture of assuming/believing everything was within its control.  That posture is a nice safeguard against the temptation to attribute failure JUST to luck, and hence failing to see the flaws that may be holding you back (or, put another way, leaving you too susceptible to the swings of good or bad fortune in a short series).

My point is simply that, overdone, that posture can also cause misjudgments in the opposite direction, because it can risk overattributing the outcome to factors that never were to blame in the first place, and hence making changes that either (a) didn’t need to be made, or (b) worse till, actually prove to be harmful going forward.

As for “goals and milestones,” sure, that’s a big part of it.  Failure to reach those should be an important—but hardly determinative—indicator of the need for changes in your approach or your execution of the plan.  If and when you fall short of goals and milestones, you STILL need to do the hard work of evaluating the reasons, and that process still needs to properly balance the factors that were attributable to you, your staff, or your players, against those that really were out of your control.

277. RoyalsReview Posted: October 08, 2005 at 03:46 AM (#1670614)

Its interesting that people generally just accept these formats as a given. (Other than the endless prattling for a playoff system in NCAA football.) At some level people understand that a single baseball game is more variable than a single football game, but I dont think I’ve ever heard it fully stated.

Is a single playoff game really that valuable for determining whats a better football team?

Of course, the football community isn’t a questioning kind, the ‘02 Patriots were always truly better than the ‘02 Rams, they just needed a larger stage to truly expose their hidden greatness, intelligence, toughness, etc.

Bleh.

278. Jasper109 Posted: October 08, 2005 at 05:56 PM (#1671117)

Well, baseball (despite it’s inherant beauty) is the most imperfect of all major sports.
A very slight difference in bat to ball contact will produce very different results.  It is the sport where playoff results are most meaningless.
In England the regular season determines the “champion” of a football (soccer) league.  The FA cup is a secondary competition that determines the best “playoff” team.  It is more or less a novelty competition that is dwarfed by the competition to be the regular season champion.  Despite the fact that soccer is a boring game, we have a lot to learn from our European brethren.
The Marlins have been “playoff” winners twice in the last decade, while the Braves have won 14 consecutive NL East championships.  Which has been the more successful franchise?  To me, it isn’t even close.
I won’t get into the “luck” discussion, but I’ll just close my first post by saying that “Backlasher” you are a complete and utter moron.
You can write paragraph after paragraph of your crap, but the bottom line is that the A’s playing 0.400 ball against good competition over 4 AL divisional series means very little.  They could have just as easily played 0.600 ball and won 2 world championships in the process.  When you “decide” which of two baseball teams is “better” in a best of 5 series IT IS A CRAPSHOOT no matter what crap you continue to spew.

279.  Posted: October 09, 2005 at 12:37 PM (#1672023)

This is a comment test.

280.  Posted: October 09, 2005 at 12:41 PM (#1672027)

This is another test.

281.  Posted: October 09, 2005 at 07:23 PM (#1672305)

The Marlins have been “playoff” winners twice in the last decade, while the Braves have won 14 consecutive NL East championships. Which has been the more successful franchise? To me, it isn’t even close.

Who here has argued the contrary?  Pure straw horse.

I won’t get into the “luck” discussion, but I’ll just close my first post by saying that “Backlasher” you are a complete and utter moron.
You can write paragraph after paragraph of your crap, but the bottom line is that the A’s playing 0.400 ball against good competition over 4 AL divisional series means very little. They could have just as easily played 0.600 ball and won 2 world championships in the process.

And if I had some ham, I could make a ham sandwich, if I had some bread.

The A’s still would have had to win 4 more series to win those two world championships, let alone the 2 division series they didn’t win. That’s a lot of ham, and that’s a lot of bread.

And of course “luck” never plays a part in determining a regular season division winner, where one team beats out another by a game or three, or where one team is crippled by injuries.  Why is “luck” something which apparently determines only postseason results?  Is this part of the “unwritten code,” where a key injury or a misplaced pebble in October is “bad luck” but the same injury or pebble from April through September is “part of the game?”

Atlanta deserves all the respect and admiration in the world for winning those division championships, but this is only an NL East dynasty, and nothing else.  It’s a great run, but to call it a baseball dynasty is just to cheapen the term.  The closest thing we’ve had to a real dynasty in recent years was the 1996-2001 Yankees, who were one throwing error and two bad pitches removed from the good chance of winning at least four, and possibly six straight World Series, and who in any case won four in five years.

But even that pales to the only sustained dynasty that baseball has ever seen, which was between 1921 and 1964.  To use Jasper’s term, the rest of them aren’t “even close.”

282. Mefisto Posted: October 09, 2005 at 11:06 PM (#1673058)

Why is “luck” something which apparently determines only postseason results?

It’s not. Bill James once estimated that teams would have to play 10,000 games in a season in order eliminate the influence of luck.

283. Mister High Standards Posted: October 09, 2005 at 11:39 PM (#1673127)

I think a pretty good case can be made that Baird tries to be “sabrementric” and fails.

After the 2003 season their was a lot of talk on his site that the Royals were going to compete in 2004.

This is a team that has had a few Sabrmetric darlings employeed over the last couple of few years.  Matt Stairs, Rontrez Johnson, Morgan Burkhart, Cal Pickering, Mark Teehan, Juan Gonzalez (worth the risk).

He has also tried to do the Indians thing of locking up young talent to long term deals.  He has been active in the rule v draft. A quick look shows more College than Highschool among high draft picks… though somewhat balanced.

I know when listening to Baird he talks a lot about onbase percentage…

284. Chris Dial Posted: October 10, 2005 at 03:37 AM (#1673724)

And of course “luck” never plays a part in determining a regular season division winner, where one team beats out another by a game or three, or where one team is crippled by injuries. Why is “luck” something which apparently determines only postseason results?

“Who here has argued the contrary? Pure straw horse.”

Goodness, that was easy.

285. DCA Posted: October 10, 2005 at 01:29 PM (#1673955)

Sometimes, you have to trust your judgment

And how does one do that naturally in a real time environment.

An example of this is in targeting (projectile weapons, machine processes, etc).  Any targeting procedure will have a systematic and a random component to the error—the first due to inexact or improper alignment, the second due to random effects due to the limits of the machinery, fluctuation of environmental conditions, etc.  If you take only one test shot and use that to calibrate, if your initial aim is “pretty good”, it’s very likely that your adjustment is going to make the subsequent calibrated shot worse, by turning random error in the first shot into bias error for subsequent shots.  In real time, you can make such adjustments by (a) taking more shots to average out the random error—this is best, but may be expensive in time, money, etc (b) use fewer shots and undercorrect, since if you think you have pretty good aim to begin with, the random error will probably be in the direction of the absolute error.

Obviously, this is different from running a baseball team, but it’s an example of where acknowledging randomness—and correcting for it in real time—will give you a better response than attributing results entirely to nonrandom factors.

286. and Posted: October 10, 2005 at 01:43 PM (#1673973)

Just popping by to note that my team, the Braves, have just lost their fourth consecutive Division Series.

I don’t think I’ll be smarting off to A’s fans any time soon as I can now say with certainty that it hurts.  Ouch.

287. DCA Posted: October 10, 2005 at 02:10 PM (#1674016)

Now if its the fanboys, no credit at all. They defended fatfukk softball players to their dying breath, and then when Beane abandoned the strategy, they talked about some super secret metric. That’s not evaluation, that is spin and rhetoric.

I can buy this ... however, I think this is a result of Beane’s admirers being behind Beane in their evaluation—if you prefer the dramatic phrasing, the disciples have not caught up to the master.

Also, from many accounts, there actually is some super secret metric (for defense, and for offense that uses primary results—type and location of batted ball—rather than secondary results—hits, outs, etc) that Beane and DePo have referred to multiple times.

Back to the first point, I think many of us came in sabremetrics at the time when the real revelation was that OBP was king—the Indians and then the Yankees were going on an OBP-driven rampage on the rest of the league, and Oakland was competing with them on a limited budget by buying up OBP on the cheap.  Just as we reflexively think of baseball as aesthetically best when we were kids and first got into it, we reflexively think of sabremetrics as it was we found it, when in fact it has moved on, both in terms of the state of the art (MGL et al) and the put-it-in-practice decision-makers (Beane et al).

This reminds me of once when someone wrote into a BP chat asking why BP kept knocking the Royals for getting raped by the A’s in the Neifi-Dye-Ortiz trade.  By BP’s own advanced metrics, Neifi was more valuable than Dye at the time of the trade. The BP chatter basically disavowed the metric and said, well we know Dye was better than Neifi and OBP, SLG, etc.  Which is not to say that Neifi was better than Dye—BP’s ever changing stats now have them even pre-trade, and edge to Dye the prior season, just that the chatter’s method of response was notable.

It was the same kind of thing that causes many here to (rightfully IMO) call bullsh1t on Joe Morgan and his ilk when they go into their “players were better back in my day” because “I played the game and I say so” when careful analysis suggests the opposite, or even that its not clear-cut.  The standard bearers of the saber movement too often engage in that kind of thinking wrt their own discipline, and instead of looking first at the evidence, use their now-out-of-date knee-jerk response and rules of thumb.

People on the board tend to echo that mentality.  For all the griping of the Cubs fans about Neifi being the biggest suck ever, he was actually a league-average starting SS this year—I agree the Cubs should have been trying to find out if Cedeno could swim, but Neifi wasn’t bad (although he seems to have lost a win of WARP since the season ended, what’s with that).

288. Slinger Francisco Barrios (Dr. Memory) Posted: October 10, 2005 at 02:22 PM (#1674038)

Posts #280 and #281 are the best arguments I’ve seen here.

I’m almost not joking.  It is astounding, the ability of intelligent people (one in particular) to twist words and engage in rank sophistry.  It’s like I imagine a Mensa meeting to be like:  full of jousters, each (one in particular) trying to prove he’s smarter than the next zhlub.

My 2¢:  “luck” exists, and does not have an amorphous definition.  Luck is merely by definition amorphous.  It comprises the factors we haven’t the wit or ability to measure.

If you would still insist that luck (however you define it, and even whether or not you believe it exists) doesn’t play a significant part in, for instance, the A’s lack of success in the postseason, you should show all your inarguable measurements showing why those failures were completely predictable.  That is, quite simply, the only way you can prove your thesis.

289. DCA Posted: October 10, 2005 at 02:45 PM (#1674073)

I’m almost not joking. It is astounding, the ability of intelligent people (one in particular) to twist words and engage in rank sophistry. It’s like I imagine a Mensa meeting to be like: full of jousters, each (one in particular) trying to prove he’s smarter than the next zhlub.

Replace “smarter” and “better” and I’m reminded of that old idiom with the pot and the kettle.

290. Jasper109 Posted: October 10, 2005 at 04:49 PM (#1674348)

The Marlins have been “playoff” winners twice in the last decade, while the Braves have won 14 consecutive NL East championships. Which has been the more successful franchise? To me, it isn’t even close.

Who here has argued the contrary? Pure straw horse.

I wasn’t saying that anyone here was making that argument.  I was saying that the general perception among the majority of baseball fans is that the Braves have in some way underachieved or choked.  I don’t believe that for a second.  To me, the Cardinals were the best team in the NL this year, and if they make it the world series or don’t won’t change my opinion.  If Bill James is even remotely close to being right that it would take a 10000 game season to eliminate the “luck factor”, what exactly are we deciding in a best of 5 or 7 game series.  If we are deciding which team has the best 3 starters, closer, and is fortunate enough to have more balls drop in than any other team, that’s fine.  Just don’t say that it determines which is the best team, because it doesn’t.

I play poker fairly seriously, and I see more and more parallels between the two.  If I make all of the best decisions over many thousands of hands I will be a winner because the cards will eventually be equal for everyone.  If I enter a tournament where only a few hundred hands are played, I might be the best player, but if I get a particularly bad run of cards I have no chance to win.  The Florida Marlins were a mediocre team getting a good run of cards, and the Oakland A’s were an overall better team getting a crappy run of cards.

I don’t know how anyone could have watched the game yesterday between the Braves and the Astros and not come to the conclusion that it’s all one big coin flip.  If Ausmus hits the ball 1/32 of an inch lower the Braves win and who knows what comes next.
There were 20 different instances in that game that could have tipped the scales one way or another.

Still, having said all of that I thought Bobby Cox didn’t have his best day yesterday.

291. DCA Posted: October 10, 2005 at 05:08 PM (#1674412)

Still, having said all of that I thought Bobby Cox didn’t have his best day yesterday.

I’ll say.  Going to Devine when he still had McBride and Boyer available, I knew the game was over.  Was Boyer even on the postseason roster?  IIRC, he didn’t pitch at all, and he’s way better than Devine right now.

292.  Posted: October 10, 2005 at 05:08 PM (#1674413)

The Marlins have been “playoff” winners twice in the last decade, while the Braves have won 14 consecutive NL East championships. Which has been the more successful franchise? To me, it isn’t even close.

Who here has argued the contrary? Pure straw horse.

I wasn’t saying that anyone here was making that argument. I was saying that the general perception among the majority of baseball fans is that the Braves have in some way underachieved or choked.

“Choked” is a loaded term, but given the talent level of the Braves, at least as evidenced by those 14 division titles, I don’t think that it’s exactly unfair to say that they’ve “underachieved” in the postseason.  Balls go over or under lines by a foot this way or that way—-that’s luck.  Bullpens throw fat pitches in key situations—-that’s not luck.

You can also say that “well, Houston was better this year,” or “Houston was better last year,” or “the Yankees were better in 1999,” and keep going down the list so that every year there’s a perfectly logical explanation for not winning the whole thing.  But after 1 for 14, you have to except a certain amount of curiosity as to why it just keeps happening—-or that those 14 division championships aren’t all that Braves fans make them out to be.

IOW they’re either great postseason underachievers, or they’re performing about like they should, in which case they weren’t all that exceptional to begin with.  They may just be lucky to be in the East instead of the Central.

293. Mister High Standards Posted: October 10, 2005 at 05:17 PM (#1674438)

I play poker fairly seriously, and I see more and more parallels between the two. If I make all of the best decisions over many thousands of hands I will be a winner because the cards will eventually be equal for everyone.

If you play poker even reasonably well, then you know that you play a Satellite different than you play a cash ring game.

In a cash game, you are properly bankrolled you should be making any plays that are PLUS ev.

In a satelite, that is a poor strategy, as risk management takes a much larger role rather than just max EV plays.  Because you don’t have 1,000 of hands to determine a winner.

It doesn’t mean that winning a satellite requires less skill than winning in a cash ring game, it just requires a different type of skill.

Much like winning in the regular season and winning in the playoffs.

The difference is in poker it doesn’t matter where your skill is, all that matters is the bottom line of your bankaccount.

In baseball it does matter, as winning the short series has more value.

294. Mister High Standards Posted: October 10, 2005 at 05:24 PM (#1674463)

I play poker fairly seriously, and I see more and more parallels between the two. If I make all of the best decisions over many thousands of hands I will be a winner because the cards will eventually be equal for everyone.

If you play poker even reasonably well, then you know that you play a Satellite different than you play a cash ring game.

In a cash game, you are properly bankrolled you should be making any plays that are PLUS ev.

In a satelite, that is a poor strategy, as risk management takes a much larger role rather than just max EV plays.  Because you don’t have 1,000 of hands to determine a winner.

It doesn’t mean that winning a satellite requires less skill than winning in a cash ring game, it just requires a different type of skill.

Much like winning in the regular season and winning in the playoffs.

The difference is in poker it doesn’t matter where your skill is, all that matters is the bottom line of your bankaccount.

In baseball it does matter, as winning the short series has more value.

295. Jasper109 Posted: October 10, 2005 at 06:15 PM (#1674576)

You are right that there are different skills and strategies in a satellite or multi player tournament compared to a ring game.  (especially if one is limit and one is no limit)

However, if you are a better ring game player than your competition and you play over a few hundred hours you will be an overall winner.  You might lose during some of your sessions, even over a week or two, but given a sufficiently large number of hands you will be a winner if you are a better player.

In any one, two, or even 7 satellites, SNGs, or multi table tournaments you can be a better player than your opponents and still not win even once in 7 tries.

I play SNGs (single table tournaments) with reasonably good success.  I finish in the money (top 3) 45% of the time and win more than my share of them and have a net profit after playing close to 2000 such tournaments.

Having said all of that I have had streaks of 12-15 of these tournaments out of the money, and have gone over 20 without winning one.  Other successful players have stories of MUCH higher such streaks.

What does it mean?  The Braves being 1 for 14 given the nature of the baseball playoffs is not all that outstanding and might not mean anything at all about the nature of the team or manager.  It goes even more so for the A’s being 0 for 4 against good competition.  Assuming they were 50/50 each time there is a 1/16 chance that you flip a coin 4 times and it comes up heads each time.

296. Mister High Standards Posted: October 10, 2005 at 06:54 PM (#1674676)

Having said all of that I have had streaks of 12-15 of these tournaments out of the money, and have gone over 20 without winning one.

Then your not maximizing your return.  Either playing at a level too high for max returns or playing suboptimal strategy.

Baseball is not flipping a coin.  I coin flip doesn’t even reasonably simulate a playoff series or series of games.  Its far more complex than that.

sufficiently large number of hands you will be a winner if you are a better player.

Who cares?  These situations aren’t reality in baseball.  There are no 10,000 game sample sizes.

297. Mister High Standards Posted: October 10, 2005 at 06:56 PM (#1674682)

Having said all of that I have had streaks of 12-15 of these tournaments out of the money, and have gone over 20 without winning one.

Then your not maximizing your return.  Either playing at a level too high for max returns or playing suboptimal strategy.

Baseball is not flipping a coin.  I coin flip doesn’t even reasonably simulate a playoff series or series of games.  Its far more complex than that.

sufficiently large number of hands you will be a winner if you are a better player.

Who cares?  These situations aren’t reality in baseball.  There are no 10,000 game sample sizes.

298. Mister High Standards Posted: October 10, 2005 at 06:59 PM (#1674690)

Having said all of that I have had streaks of 12-15 of these tournaments out of the money, and have gone over 20 without winning one.

Then your not maximizing your return.  Either playing at a level too high for max returns or playing suboptimal strategy.

Baseball is not flipping a coin.  I coin flip doesn’t even reasonably simulate a playoff series or series of games.  Its far more complex than that.

sufficiently large number of hands you will be a winner if you are a better player.

Who cares?  These situations aren’t reality in baseball.  There are no 10,000 game sample sizes.

299. Jasper109 Posted: October 10, 2005 at 07:13 PM (#1674724)

I heard you the first time, and you are just plain WRONG.  Even playing optimal strategy against competition you are better than can lead to such streaks.  Check out a poker forum and learn a few things before posting such nonsense.

Daniel Negreannu is one of the best tournament poker players in the world.  In this years WSOP he didn’t even come close to making the money (never mind winning) any of the numerous tournament he entered.  It’s the nature of the game.  Maybe you haven’t played long enough to experience a 300BB downswing in ring games or what I have described in SNGs.  Trust me, if you play long enough, you will, no matter how good a player you are.

300. Mister High Standards Posted: October 10, 2005 at 07:24 PM (#1674750)

Having said all of that I have had streaks of 12-15 of these tournaments out of the money, and have gone over 20 without winning one.

Then your not maximizing your return.  Either playing at a level too high for max returns or playing suboptimal strategy.

Baseball is not flipping a coin.  I coin flip doesn’t even reasonably simulate a playoff series or series of games.  Its far more complex than that.

sufficiently large number of hands you will be a winner if you are a better player.

Who cares?  These situations aren’t reality in baseball.  There are no 10,000 game sample sizes.

Page 3 of 4 pages  < 1 2 3 4 >

You must be Registered and Logged In to post comments.

<< Back to main

### Support BBTF

Thanks to
Ray (CTL)
for his generous support.

### Bookmarks

You must be logged in to view your Bookmarks.

### Syndicate

Page rendered in 1.3697 seconds
58 querie(s) executed