Throw my game to change playoff teams? (1 Viewer)

[ghostguy123]

They are not "real" odds because they are easy to measure. They are real odds because they are exact proven by math.

The odds of winning 1000 dollars at roulette starting with 500 dollars using the martingale system betting on black is not easy to figure out, but if you know how to do it, it does have en exact percentage chance to happen.

As for the scumbag distiction, if your league specifically says you can not tank under any curcumstsnce, then you can't tank. If it doesn't, then I think its fine in this super rare scenario.

If this situation is even possible in your league, it should be addressed in the rules as being allowed or not.

The odds of winning a fantasy football game can be proven by math if you know all the variables. The physical world is governed by the same math as blackjack (except maybe at a subatomic level that is very unlikely to affect a fantasy football game). It's just that the variables are harder to observe, and thus the odds are tougher to measure.

That's not true. A week of NFL football is a chaotic system; it's no more possible to predict what will happen next week than to predict whether it will rain next July 4. And one of the factors providing input to that chaotic system is the workings of the brains of about 1500 humans, which are governed by quantum-level effects which are known to be unpredictable.

It's more than difficult to know all the variables; it is scientifically impossible.

And yet handicappers are remarkably good at predicting outcomes of NFL games. And they'd be equally good at predicting fantasy games if the incentives were there.

Handicappers don't try and figure out what the outcome of the game will be, they try and figure out the best possible way to get half the public to bet one way, and half the public to bet the other way. But even if they were trying to figure out how a game will turn out, that is still NOT actual odds.

 

[ghostguy123]

And no, predicting the outcome of an NFL game with exact odds is not possible whether you know all the variables or not, because as soon as you make your prediction the variables begin changing.

You are talking about magic, sourcery, psychics at that point.

 

[CalBear]

And yet handicappers are remarkably good at predicting outcomes of NFL games. And they'd be equally good at predicting fantasy games if the incentives were there.

What percentage of games do you think handicappers predict correctly, in terms of win/loss, point spread, and point totals?

Meteorologists can predict the weather sometimes, too.

 

[Bamac]

They are not "real" odds because they are easy to measure. They are real odds because they are exact proven by math.

The odds of winning 1000 dollars at roulette starting with 500 dollars using the martingale system betting on black is not easy to figure out, but if you know how to do it, it does have en exact percentage chance to happen.

As for the scumbag distiction, if your league specifically says you can not tank under any curcumstsnce, then you can't tank. If it doesn't, then I think its fine in this super rare scenario.

If this situation is even possible in your league, it should be addressed in the rules as being allowed or not.

The odds of winning a fantasy football game can be proven by math if you know all the variables. The physical world is governed by the same math as blackjack (except maybe at a subatomic level that is very unlikely to affect a fantasy football game). It's just that the variables are harder to observe, and thus the odds are tougher to measure.

That's not true. A week of NFL football is a chaotic system; it's no more possible to predict what will happen next week than to predict whether it will rain next July 4. And one of the factors providing input to that chaotic system is the workings of the brains of about 1500 humans, which are governed by quantum-level effects which are known to be unpredictable.

It's more than difficult to know all the variables; it is scientifically impossible.

Also, I think most -- and certainly many -- physicists would disagree that large events like a football game (and all its observable components) aren't deterministic. Quantum-level effects may occur in the brain, but it's not clear how that translates to human action. I think we can all agree, though, that these issues have no effect on whether tanking is ethical.

 

[Bamac]

And yet handicappers are remarkably good at predicting outcomes of NFL games. And they'd be equally good at predicting fantasy games if the incentives were there.

What percentage of games do you think handicappers predict correctly, in terms of win/loss, point spread, and point totals?

Meteorologists can predict the weather sometimes, too.

They set lines such that damn near 50% of underdogs and 50% of favorites cover. And they could do the same to create any other resulting percentages.

 

[hamster_13]

Update: As it stands, I will lose/would have lost regardless of which team I played. Lowest scoring week of the year for my team this year (still have Stafford and Megatron going)

 

[ghostguy123]

And yet handicappers are remarkably good at predicting outcomes of NFL games. And they'd be equally good at predicting fantasy games if the incentives were there.

What percentage of games do you think handicappers predict correctly, in terms of win/loss, point spread, and point totals?

Meteorologists can predict the weather sometimes, too.

They set lines such that damn near 50% of underdogs and 50% of favorites cover. And they could do the same to create any other resulting percentages.

They are good, but they are not exact.

Roll dice and there is EXACTLY a 1 in 36 chance you get snake eyes. Every time, no matter what happend on previous rolls. There is a predicted outcome based on true mathematical formulas that are proven to be accurate.

A bookies lines are not based on fact. Whenever trends begin to occur, people betting adapt to it, but then so do the bookies. They actually manipulate the outcomes at times. If there is a long stretch of underdogs covering, and people begin to start betting that way, the bookie will adjust the lines accordingly.

ALthough, even if every underdog covers every week for 5 years in a row, the bookies will not change the lines as long as the public is betting half on one team and half on the other.

Hell, sometimes the lines come out and change before gametime. Say the spread on the game tonight was Detroit -7 and 500 people bet Detroit and 45 peope bet Baltimore. I guarantee you that line is going to go up to like 8 or 9 points. So those bookies are manipulating the outcomes. None of these represent true odds, sorry

I promise you a casino will NEVER alter their payouts on these games with true odds just because people are all betting one way. If 10 people are playing roulette and they all bet black, the casino doesn't care. There is an 18/38 chance they win, and that is in the casino's favor.

If a bookie was so exact and so great at setting their llines, then they would never change them based on how people are betting. The don't because they know they are not setting true odds like the casino games.

 

[ghostguy123]

I think we can all agree, though, that these issues have no effect on whether tanking is ethical.

I only think it is a very, VERY small part of it. But it's just my opinion of how I view the situation when your only chance of making the playoffs is to lose your game. But essentially yeah, you are right. It is either ethical or it isn't. I just happen to think in this scenario, it is not unethical. Odds or no odds who cares, when your only chance of making the playoffs is by losing the week before the playoffs start, I am fine with it.

 

[davearm]

I don't even know where to begin, so I'll just reiterate, for all practical purposes, the odds blackjack players use in their decisionmaking are the same as the odds fantasy football players use. They're both estimates.

Wrong. One is an estimate. The other is a discrete value that can be calculated.

Blackjack players aren't calculating discrete values in real time when they've got 10 seconds to decide whether to hit or stand. They're relying on estimates (or a generic table, which is computed assuming one deck and one player). Same as the FF player. Hence the term, "for all practical purposes".

 

[davearm]

This sportsbook discussion is curious. Oddsmakers aren't trying to peg the number that gives each team 50% odds. They're trying to peg the number that maximizes their profits.

If they think tonight's game is a pick em in terms of who will win, but making Baltimore a 7 point favorite will draw equal money on both sides, they'll hang a 7.

They use the term odds, but that's not what they're designed to be.

 

[ghostguy123]

I don't even know where to begin, so I'll just reiterate, for all practical purposes, the odds blackjack players use in their decisionmaking are the same as the odds fantasy football players use. They're both estimates.

Wrong. One is an estimate. The other is a discrete value that can be calculated.

Blackjack players aren't calculating discrete values in real time when they've got 10 seconds to decide whether to hit or stand. They're relying on estimates (or a generic table, which is computed assuming one deck and one player). Same as the FF player. Hence the term, "for all practical purposes".

right, we all know this.

But there are still true mathematical odds for each hand. There is no disputing this. If they had the time and the casino let them do it, statasticians who know how to figure it out would figure out the exact odds of each hand they have, and make their decisions accordingly.

It's not the same, at all. One has true odds, one does not. Just because the casino has rules against letting you count cards and sit there making calculations doesn't mean there are not true odds. If the casino let people do this, they would go bankrupt immediately............un.less they have an automatic shuffle machine that makes every hand new and unique, in which case you can use all the true odds you want, the casino still has the advantage

 

[ghostguy123]

This sportsbook discussion is curious. Oddsmakers aren't trying to peg the number that gives each team 50% odds. They're trying to peg the number that maximizes their profits.

If they think tonight's game is a pick em in terms of who will win, but making Baltimore a 7 point favorite will draw equal money on both sides, they'll hang a 7.

They use the term odds, but that's not what they're designed to be.

Right. They don't care how accurate they are, they care how much money they make. And they make the most money overall if they can get half the people to bet one way and half the other. They make about 5% of all money wagered when this happens and they don't have to adjust any lines.

None of it is true odds at all.

 

[Short Corner]

I don't even know where to begin, so I'll just reiterate, for all practical purposes, the odds blackjack players use in their decisionmaking are the same as the odds fantasy football players use. They're both estimates.

Wrong. One is an estimate. The other is a discrete value that can be calculated.

Blackjack players aren't calculating discrete values in real time when they've got 10 seconds to decide whether to hit or stand. They're relying on estimates (or a generic table, which is computed assuming one deck and one player). Same as the FF player. Hence the term, "for all practical purposes".

I never said they calculated the values, I said the values can be calculated. A player doesn't need to know the exact value, just the decision he needs to make given that value.

There is more than one system and most of the good ones take into account the specific amount of decks in the shoe as well as shoe depth.

For practical purposes, it's not in the same ball park. It's not even the same ####### sport. [/JulesWinfield]

 

[davearm]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

This nonsense about real vs. Perceived is so irrelevant and pedantic. The distinction is completely meaningless to the tanking discussion. So the blackjack player could get exact odds if you gave him a day and access to a phd statistician with a big computer. Who cares? What practical difference does that make?

Let me ask you this. If the FF player could do the same (compute exact odds) would that change your mind about tanking? Ghost guy you've already stated that it wouldn't, by your answer to the "tank to get a bye" hypo.

 

[Short Corner]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

Reading comprehension apparently isn't one of your strong points either.

 

[Hoss Style]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

Reading comprehension apparently isn't one of your strong points either.

I'm not really taking a side here on the over all "tanking" issue, but I have to agree with Short here. These guys are saying that one example has specific, known odds while the other (FF) does not have specific, known odds. It seems that they then go on to say that basing a decision based on known, specific odds is different than basing a decision on unknown, nonspecific odds.

Then you reply with "Both of you guys have just confirmed my point. Both players base their decisions on estimated odds." That seems to me to be the exact opposite of what they are saying - and they've been pretty plain about how they are saying it.

It matters not who I agree with here - I just know that they are definately not proving your point, from what I've read.

 

[CalBear]

And yet handicappers are remarkably good at predicting outcomes of NFL games. And they'd be equally good at predicting fantasy games if the incentives were there.

What percentage of games do you think handicappers predict correctly, in terms of win/loss, point spread, and point totals?

Meteorologists can predict the weather sometimes, too.

They set lines such that damn near 50% of underdogs and 50% of favorites cover. And they could do the same to create any other resulting percentages.

I notice you didn't answer the question.

 

[CalBear]

They are not "real" odds because they are easy to measure. They are real odds because they are exact proven by math.

The odds of winning 1000 dollars at roulette starting with 500 dollars using the martingale system betting on black is not easy to figure out, but if you know how to do it, it does have en exact percentage chance to happen.

As for the scumbag distiction, if your league specifically says you can not tank under any curcumstsnce, then you can't tank. If it doesn't, then I think its fine in this super rare scenario.

If this situation is even possible in your league, it should be addressed in the rules as being allowed or not.

The odds of winning a fantasy football game can be proven by math if you know all the variables. The physical world is governed by the same math as blackjack (except maybe at a subatomic level that is very unlikely to affect a fantasy football game). It's just that the variables are harder to observe, and thus the odds are tougher to measure.

That's not true. A week of NFL football is a chaotic system; it's no more possible to predict what will happen next week than to predict whether it will rain next July 4. And one of the factors providing input to that chaotic system is the workings of the brains of about 1500 humans, which are governed by quantum-level effects which are known to be unpredictable.

It's more than difficult to know all the variables; it is scientifically impossible.

Also, I think most -- and certainly many -- physicists would disagree that large events like a football game (and all its observable components) aren't deterministic. Quantum-level effects may occur in the brain, but it's not clear how that translates to human action.

Quantum-level effects are what cause human decisions to be unpredictable. If a QB is presented with the same split-second decision to throw or not to throw a pass multiple times, he will not make the same decision every time, nor will his muscles behave the same way every time. The result of that decision and that throw (and every other decision and action a human makes during the game) has a chaotic effect; the difference between a touchdown and an interception alters every play for the rest of the game.

 

[davearm]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

Reading comprehension apparently isn't one of your strong points either.

Elaborate if you will.

The blackjack player is sitting on a 14. The dealer shows 10. There are 5 other players at the table, so 10 other cards have been revealed. It's the second hand out of this chute, so another, say, 18 cards were dealt last hand.

You have 10 seconds to decide whether to stand or hit. Are you basing your decision on an estimate of your odds, or are you factoring all of the above information into an algorithm that will spit out your precise odds to 8 digits?

 

[davearm]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

Reading comprehension apparently isn't one of your strong points either.

I'm not really taking a side here on the over all "tanking" issue, but I have to agree with Short here. These guys are saying that one example has specific, known odds while the other (FF) does not have specific, known odds. It seems that they then go on to say that basing a decision based on known, specific odds is different than basing a decision on unknown, nonspecific odds.

Then you reply with "Both of you guys have just confirmed my point. Both players base their decisions on estimated odds." That seems to me to be the exact opposite of what they are saying - and they've been pretty plain about how they are saying it.

It matters not who I agree with here - I just know that they are definately not proving your point, from what I've read.

The odds are not known in either instance. In the blackjack case, we could compute them, given the time and resources, but they are not known at the time the player is deciding.

 

[ghostguy123]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

This nonsense about real vs. Perceived is so irrelevant and pedantic. The distinction is completely meaningless to the tanking discussion. So the blackjack player could get exact odds if you gave him a day and access to a phd statistician with a big computer. Who cares? What practical difference does that make?

Let me ask you this. If the FF player could do the same (compute exact odds) would that change your mind about tanking? Ghost guy you've already stated that it wouldn't, by your answer to the "tank to get a bye" hypo.

If a FF could compute exact odds?? You may as well ask me if I would change my mind about what I asked for for Christmas is Santa actually existed.

And I have no idea what you think "we confirmed" for you. A fantasy football player can get more info to make better decisions, and make their chances better, sure. But the guy playing blackjack is not going to make more precise odds, he is goign to find the EXACT odds.

I mean, that is night and day difference in terms of explaining real and perceived odds.

 

[ghostguy123]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

Reading comprehension apparently isn't one of your strong points either.

I'm not really taking a side here on the over all "tanking" issue, but I have to agree with Short here. These guys are saying that one example has specific, known odds while the other (FF) does not have specific, known odds. It seems that they then go on to say that basing a decision based on known, specific odds is different than basing a decision on unknown, nonspecific odds.

Then you reply with "Both of you guys have just confirmed my point. Both players base their decisions on estimated odds." That seems to me to be the exact opposite of what they are saying - and they've been pretty plain about how they are saying it.

It matters not who I agree with here - I just know that they are definately not proving your point, from what I've read.

I actually think he knows this, but might be arguing something else. At least I hope so, for his sake.

 

[ghostguy123]

Quantum-level effects are what cause human decisions to be unpredictable. If a QB is presented with the same split-second decision to throw or not to throw a pass multiple times, he will not make the same decision every time, nor will his muscles behave the same way every time. The result of that decision and that throw (and every other decision and action a human makes during the game) has a chaotic effect; the difference between a touchdown and an interception alters every play for the rest of the game.

Not to mention this fact. Lets assume it is possible (even though it isnt) to know every single variable that goes into an NFL game. Lets say some guy out there knows all 10 zillion variables to predict the real odds. The exact second he sets those odds, the variables begin to change. A guy could get hurt before the game, eat 3 eggs for breakfast instead of his usual 4..............or a billion other things that would change.

Not to mention even just one issue would throw things off, the weather. This guy who knows all the variables can only know what the weather man predicts, whichpretty much is at least a little off pretty much most days.

 

[ghostguy123]

Both of you guys have just confirmed my point. Both players base their decisions on estimated odds. And both players could compute more precise odds, if given more time and resources. And the FF player has way more time... Several days between decisions rather than several seconds.

Reading comprehension apparently isn't one of your strong points either.

Elaborate if you will.

The blackjack player is sitting on a 14. The dealer shows 10. There are 5 other players at the table, so 10 other cards have been revealed. It's the second hand out of this chute, so another, say, 18 cards were dealt last hand.

You have 10 seconds to decide whether to stand or hit. Are you basing your decision on an estimate of your odds, or are you factoring all of the above information into an algorithm that will spit out your precise odds to 8 digits?

You are just simply not understanding, so I guess he was right that you are not reading it right.

The casino's rules against counting cards and allowing you to take time to do these math calculations is not exactly a strong argument for you in saying that those odds are not real odds. The casino is prohibitting you from using those odds because THEY KNOW the odds are real odds. It's also why they use either 6 decks and only go through about half of it, or an automatic shuffle machine.

You are stuck on what players are doing rather than the facts that the odds are there if you are given the time to calculate them.

 

[davearm]

My point is that "the odds are there if you are given the time to calculate them" is irrelevant, since you aren't given the time to calculate them.

In the absence of having the time to calculate them, you have to make an estimate. Just like FF players do.

 

[davearm]

And I have no idea what you think "we confirmed" for you. A fantasy football player can get more info to make better decisions, and make their chances better, sure. But the guy playing blackjack is not going to make more precise odds, he is goign to find the EXACT odds.

I mean, that is night and day difference in terms of explaining real and perceived odds.

You have said more (WAY WAY more) than needs to be said about "real" and "perceived" odds. I understand just fine the hair-splitting you're engaging in here.

Now if you are teling me that the difference between the real and perceived odds is irrelevant, that is fine.

We both agree the distinction is irrelevant to the tanking discussion, so why do you keep harping on it?

 

[Adam Harstad]

Let me ask you this.

Say I'm playing blackjack. I've got 14. The dealer is showing a 10.

The odds of me winning are X% if I stand.

The odds of me winning are Y% if I hit.

I have no idea what X and Y are, but my best guess is that Y > X, so I hit.

Are X and Y "real" odds, or "perceived" odds?

Umm, they are real/actual odds based on mathematical certainty. Why are you asking? You have me curious.

But again, as stated, this is only a small part of my reasoning. Even though I disagree with the way a couple of you are using that article to try and back up your argument, it doesn't matter anyway.

I don't think this situation calls for any more explanation other than "it is your only chance to win".

A nice example of perceived odds would be this. I ask you what your chances are of winning your league this year (assume I asked right before the playoffs started). Any answer you give me is perceived odds.

Sort of like tanking to change your playoff seed because you perceive your odds to be better that way. No, it is true your odds may very well be better if you manipulate it so that the next team you face is a horrible team, but the increase in your odds to win the title by doing that is completely perceived.

I guess another example of perceived odds would be anything in the Vegas Sportsbook.

I ask because the situation is no different than what we face in fantasy football.

Say I'm rostering Peyton Manning and Joe Flacco.

The odds of me winning are X% if I start Flacco.

The odds of me winning are Y% if I start Manning.

I have no idea what X and Y are, but my best guess is that Y > X, so I start Manning.

You told me X and Y were real in the blackjack scenario (and I agree). They're no less real in the FF scenario. More difficult to compute, possibly, but no less real.

Perhaps more to the point, in both situations the player is making a choice based upon a best guess of what those odds are... what you characterize as "perceived" odds. Now an MIT guy with a big brain and a big computer could compute the Xs and the Ys for us in both situations, but that degree of information is not available in real time to either the blackjack player or the fantasy football player.

Bottom line, to your way of thinking, they must both be actual, or both be perceived. There's no meaningful difference in how they're guiding players' decisions.

wow, you are very wrong here.

Real/actual are odds are mathematical certainties based on exact formulas.

Nothing in fantasy football is a mathematical certainty.

And the MAIN basis of me saying tanking is ok, which is the most important thing pertaining to this discussion, is that it the ONLY way your team can win. As in, zero chance to win the league if you win that game.

MOre diffucult to computer??? There is no actual way to compute the odds. There are people that are better at guessing them, sure, but they are anything but true odds.

The following phrases need to be retired for the remainder of this thread for being oxymorons: "real odds", "actual odds", "exact odds". While we're at it, let's retire "perceived odds", "estimated odds", and "hypothetical odds" for being redundant. All odds, by definition, are not real, actual, or exact. All odds, by definition, are perceived, estimated, and hypothetical.

There is no such thing as "real" odds. Probability is a human creation applied to an event rather than an intrinsic property of the event itself. When you flip a coin, that coin is either going to come up heads, or it is going to come up tails. You can say ahead of time that a coin has a 50% chance of coming up heads based on your ex post knowledge that over a large enough number of flips, the proportion of heads to total flips will approach 50%. That doesn't change the fact that the "odds" are made up- the coin doesn't have a percent chance of coming up heads, either it's going to come up heads or it won't. That percentage is an ex ante estimate we make based on the information we have to make up for the fact that we don't have complete enough information to know for sure which way the coin will come up.

Likewise, either Joe Flacco is going to outscore Peyton Manning this week, or Peyton Manning is going to outscore Joe Flacco. We don't know ahead of time which is true, so we create ex ante estimates based on the limited information we have available. If we're good with our estimates, the sum of our ex ante estimates will, over a long enough timeline, closely mirror the ex post results. Even if our estimates line up 100% with the results (like our 50% estimate will tend to line up with the coin flip results), that doesn't magically make those odds "real" or "actual". In both cases, the odds are a figment, a human construct, and not an intrinsic property of the object whose behavior we are predicting.

 

[Bamac]

And yet handicappers are remarkably good at predicting outcomes of NFL games. And they'd be equally good at predicting fantasy games if the incentives were there.

What percentage of games do you think handicappers predict correctly, in terms of win/loss, point spread, and point totals?

Meteorologists can predict the weather sometimes, too.

They set lines such that damn near 50% of underdogs and 50% of favorites cover. And they could do the same to create any other resulting percentages.

I notice you didn't answer the question.

Your question was off base. They -- by the power of the betting market -- can turn an NFL game into a coin flip. If they could predict games with 100% certainty, there would be no odds (or the odds would always be 1 for the predicted outcome and 0 for everything else).

 

[Adam Harstad]

Maybe you should get a hold of the guy who wrote the article and send him this thread, see what he says.

I suspect Daniel Kahneman has better things to do.

I suspect you shouldn't assume what he would think of this fantasy football scenario. I think there is an excellent chance he would not view my thoughts on this as being illogical.

I do not know Daniel Kahneman, and I wouldn't presume to speak for him, but I would point out that he has pretty much spent the last 40 years demonstrating how people behave irrationally in every manner of situation imaginable, and I find it exceedingly unlikely (using my estimated, perceived, hypothetical odds ) that he would somehow find his research to be compatible with every aspect of life... except for fantasy football.

Kahneman wrote in the introduction to his book that his purpose for writing the book was to inspire people to apply his research to the way they think about and approach ordinary everyday situations in their life. I do not think he would oppose us applying his research to an ordinary, everyday fantasy football matchup.

 

[davearm]

The following phrases need to be retired for the remainder of this thread for being oxymorons: "real odds", "actual odds", "exact odds". While we're at it, let's retire "perceived odds", "estimated odds", and "hypothetical odds" for being redundant. All odds, by definition, are not real, actual, or exact. All odds, by definition, are perceived, estimated, and hypothetical.

There is no such thing as "real" odds. Probability is a human creation applied to an event rather than an intrinsic property of the event itself. When you flip a coin, that coin is either going to come up heads, or it is going to come up tails. You can say ahead of time that a coin has a 50% chance of coming up heads based on your ex post knowledge that over a large enough number of flips, the proportion of heads to total flips will approach 50%. That doesn't change the fact that the "odds" are made up- the coin doesn't have a percent chance of coming up heads, either it's going to come up heads or it won't. That percentage is an ex ante estimate we make based on the information we have to make up for the fact that we don't have complete enough information to know for sure which way the coin will come up.

Likewise, either Joe Flacco is going to outscore Peyton Manning this week, or Peyton Manning is going to outscore Joe Flacco. We don't know ahead of time which is true, so we create ex ante estimates based on the limited information we have available. If we're good with our estimates, the sum of our ex ante estimates will, over a long enough timeline, closely mirror the ex post results. Even if our estimates line up 100% with the results (like our 50% estimate will tend to line up with the coin flip results), that doesn't magically make those odds "real" or "actual". In both cases, the odds are a figment, a human construct, and not an intrinsic property of the object whose behavior we are predicting.

Thank God for that. I hope we can get back to "odds are odds" here in this discussion. I don't have much hope for that, though.

 

[Adam Harstad]

I don't even know where to begin, so I'll just reiterate, for all practical purposes, the odds blackjack players use in their decisionmaking are the same as the odds fantasy football players use. They're both estimates.

Wrong. One is an estimate. The other is a discrete value that can be calculated.

It's not a discrete value, though.

Let's say I deal you a card face down from a freshly shuffled deck. What are the odds that that card is a King? Now let's say I flip the card over and it's a 3. What are the odds that that card is a King?

The card hasn't changed. It's the same now as it was before I flipped it over. The card was never a King, even when you estimated that there was a 1/13th chance there was. All the odds tell us is that if we played that game a billion times, you'd guess each time that you'd be right 1/13th of the time, and in the end you'd be right 1/13th of the time. The odds give us useful information about YOU- your level of knowledge and ignorance vis-a-vis the card- but they give us absolutely no useful information about the card itself.

 

[Short Corner]

Let me ask you this.

Say I'm playing blackjack. I've got 14. The dealer is showing a 10.

The odds of me winning are X% if I stand.

The odds of me winning are Y% if I hit.

I have no idea what X and Y are, but my best guess is that Y > X, so I hit.

Are X and Y "real" odds, or "perceived" odds?

Umm, they are real/actual odds based on mathematical certainty. Why are you asking? You have me curious.

But again, as stated, this is only a small part of my reasoning. Even though I disagree with the way a couple of you are using that article to try and back up your argument, it doesn't matter anyway.

I don't think this situation calls for any more explanation other than "it is your only chance to win".

A nice example of perceived odds would be this. I ask you what your chances are of winning your league this year (assume I asked right before the playoffs started). Any answer you give me is perceived odds.

Sort of like tanking to change your playoff seed because you perceive your odds to be better that way. No, it is true your odds may very well be better if you manipulate it so that the next team you face is a horrible team, but the increase in your odds to win the title by doing that is completely perceived.

I guess another example of perceived odds would be anything in the Vegas Sportsbook.

I ask because the situation is no different than what we face in fantasy football.

Say I'm rostering Peyton Manning and Joe Flacco.

The odds of me winning are X% if I start Flacco.

The odds of me winning are Y% if I start Manning.

I have no idea what X and Y are, but my best guess is that Y > X, so I start Manning.

You told me X and Y were real in the blackjack scenario (and I agree). They're no less real in the FF scenario. More difficult to compute, possibly, but no less real.

Perhaps more to the point, in both situations the player is making a choice based upon a best guess of what those odds are... what you characterize as "perceived" odds. Now an MIT guy with a big brain and a big computer could compute the Xs and the Ys for us in both situations, but that degree of information is not available in real time to either the blackjack player or the fantasy football player.

Bottom line, to your way of thinking, they must both be actual, or both be perceived. There's no meaningful difference in how they're guiding players' decisions.

wow, you are very wrong here.

Real/actual are odds are mathematical certainties based on exact formulas.

Nothing in fantasy football is a mathematical certainty.

And the MAIN basis of me saying tanking is ok, which is the most important thing pertaining to this discussion, is that it the ONLY way your team can win. As in, zero chance to win the league if you win that game.

MOre diffucult to computer??? There is no actual way to compute the odds. There are people that are better at guessing them, sure, but they are anything but true odds.

The following phrases need to be retired for the remainder of this thread for being oxymorons: "real odds", "actual odds", "exact odds". While we're at it, let's retire "perceived odds", "estimated odds", and "hypothetical odds" for being redundant. All odds, by definition, are not real, actual, or exact. All odds, by definition, are perceived, estimated, and hypothetical.

There is no such thing as "real" odds. Probability is a human creation applied to an event rather than an intrinsic property of the event itself. When you flip a coin, that coin is either going to come up heads, or it is going to come up tails. You can say ahead of time that a coin has a 50% chance of coming up heads based on your ex post knowledge that over a large enough number of flips, the proportion of heads to total flips will approach 50%. That doesn't change the fact that the "odds" are made up- the coin doesn't have a percent chance of coming up heads, either it's going to come up heads or it won't. That percentage is an ex ante estimate we make based on the information we have to make up for the fact that we don't have complete enough information to know for sure which way the coin will come up.

Likewise, either Joe Flacco is going to outscore Peyton Manning this week, or Peyton Manning is going to outscore Joe Flacco. We don't know ahead of time which is true, so we create ex ante estimates based on the limited information we have available. If we're good with our estimates, the sum of our ex ante estimates will, over a long enough timeline, closely mirror the ex post results. Even if our estimates line up 100% with the results (like our 50% estimate will tend to line up with the coin flip results), that doesn't magically make those odds "real" or "actual". In both cases, the odds are a figment, a human construct, and not an intrinsic property of the object whose behavior we are predicting.

The probabilities are not made up nor a figment of your imagination when it comes to coin flips, rolls of the ice, or predicting cards. To infer otherwise is idiotic to say the least.

 

[Bamac]

As to the thread subject, another way to think about it is this: imagine you're going to join one of two fantasy leagues. Neither league has an express rule against tanking, but you know through inside info that one has a strong anti-tanking culture, while the other has an "anything goes" culture. Which would you prefer to join and why?

I can imagine three answers to this question:

1. Purists -- these folks would prefer the anti-tanking culture because they think such a culture is most likely to reward the "best" fantasy owners. This group tends to advocate victory point standings, or at least all-play or total points as a tiebreaker. I'm in this group.

2. Lombardis -- these folks prefer the "anything goes" culture because "winning isn't everything. It's the only thing." This group tends to favor H2H as a tiebreaker. I admit I don't understand this group's reasoning, and I prefer not to be in leagues with them, but I respect their worldview and don't view my own as objectively superior.

3. Sleazeballs. These folks would join the anti-tanking league in order to take advantage of the culture. They will win at all costs, even in direct defiance of the league culture, because "it's not in the rules." I have no use for these people.

I don't usually bump my own posts, but doing so here to try to get the discussion back on track. What do you all think of this?

It seems that owners have a spectrum of beliefs about what fantasy leagues should value. At one pole, the Purists think the league should, first and foremost, reward owners who are good at predicting player performance. At the other, the Lombardis think the league should reward owners who can win, regardless of tactics, so long as they stay within the written rules. In between are the people who generally adhere to one view or the other, but with discrete reservations (either principled or unprincipled).

I don't know that either side can make much progress converting the other, and the best we can do is try to play in leagues with others who have similar beliefs on this issue.

 

[davearm]

Let me ask you this.

Say I'm playing blackjack. I've got 14. The dealer is showing a 10.

The odds of me winning are X% if I stand.

The odds of me winning are Y% if I hit.

I have no idea what X and Y are, but my best guess is that Y > X, so I hit.

Are X and Y "real" odds, or "perceived" odds?

Umm, they are real/actual odds based on mathematical certainty. Why are you asking? You have me curious.

But again, as stated, this is only a small part of my reasoning. Even though I disagree with the way a couple of you are using that article to try and back up your argument, it doesn't matter anyway.

I don't think this situation calls for any more explanation other than "it is your only chance to win".

A nice example of perceived odds would be this. I ask you what your chances are of winning your league this year (assume I asked right before the playoffs started). Any answer you give me is perceived odds.

Sort of like tanking to change your playoff seed because you perceive your odds to be better that way. No, it is true your odds may very well be better if you manipulate it so that the next team you face is a horrible team, but the increase in your odds to win the title by doing that is completely perceived.

I guess another example of perceived odds would be anything in the Vegas Sportsbook.

I ask because the situation is no different than what we face in fantasy football.

Say I'm rostering Peyton Manning and Joe Flacco.

The odds of me winning are X% if I start Flacco.

The odds of me winning are Y% if I start Manning.

I have no idea what X and Y are, but my best guess is that Y > X, so I start Manning.

You told me X and Y were real in the blackjack scenario (and I agree). They're no less real in the FF scenario. More difficult to compute, possibly, but no less real.

Perhaps more to the point, in both situations the player is making a choice based upon a best guess of what those odds are... what you characterize as "perceived" odds. Now an MIT guy with a big brain and a big computer could compute the Xs and the Ys for us in both situations, but that degree of information is not available in real time to either the blackjack player or the fantasy football player.

Bottom line, to your way of thinking, they must both be actual, or both be perceived. There's no meaningful difference in how they're guiding players' decisions.

wow, you are very wrong here.

Real/actual are odds are mathematical certainties based on exact formulas.

Nothing in fantasy football is a mathematical certainty.

And the MAIN basis of me saying tanking is ok, which is the most important thing pertaining to this discussion, is that it the ONLY way your team can win. As in, zero chance to win the league if you win that game.

MOre diffucult to computer??? There is no actual way to compute the odds. There are people that are better at guessing them, sure, but they are anything but true odds.

The following phrases need to be retired for the remainder of this thread for being oxymorons: "real odds", "actual odds", "exact odds". While we're at it, let's retire "perceived odds", "estimated odds", and "hypothetical odds" for being redundant. All odds, by definition, are not real, actual, or exact. All odds, by definition, are perceived, estimated, and hypothetical.

There is no such thing as "real" odds. Probability is a human creation applied to an event rather than an intrinsic property of the event itself. When you flip a coin, that coin is either going to come up heads, or it is going to come up tails. You can say ahead of time that a coin has a 50% chance of coming up heads based on your ex post knowledge that over a large enough number of flips, the proportion of heads to total flips will approach 50%. That doesn't change the fact that the "odds" are made up- the coin doesn't have a percent chance of coming up heads, either it's going to come up heads or it won't. That percentage is an ex ante estimate we make based on the information we have to make up for the fact that we don't have complete enough information to know for sure which way the coin will come up.

Likewise, either Joe Flacco is going to outscore Peyton Manning this week, or Peyton Manning is going to outscore Joe Flacco. We don't know ahead of time which is true, so we create ex ante estimates based on the limited information we have available. If we're good with our estimates, the sum of our ex ante estimates will, over a long enough timeline, closely mirror the ex post results. Even if our estimates line up 100% with the results (like our 50% estimate will tend to line up with the coin flip results), that doesn't magically make those odds "real" or "actual". In both cases, the odds are a figment, a human construct, and not an intrinsic property of the object whose behavior we are predicting.

The probabilities are not made up nor a figment of your imagination when it comes to coin flips, rolls of the ice, or predicting cards. To infer otherwise is idiotic to say the least.

Adam didn't infer anything. You did.

 

[Short Corner]

I don't even know where to begin, so I'll just reiterate, for all practical purposes, the odds blackjack players use in their decisionmaking are the same as the odds fantasy football players use. They're both estimates.

Wrong. One is an estimate. The other is a discrete value that can be calculated.

It's not a discrete value, though.

Let's say I deal you a card face down from a freshly shuffled deck. What are the odds that that card is a King? Now let's say I flip the card over and it's a 3. What are the odds that that card is a King?

The card hasn't changed. It's the same now as it was before I flipped it over. The card was never a King, even when you estimated that there was a 1/13th chance there was. All the odds tell us is that if we played that game a billion times, you'd guess each time that you'd be right 1/13th of the time, and in the end you'd be right 1/13th of the time. The odds give us useful information about YOU- your level of knowledge and ignorance vis-a-vis the card- but they give us absolutely no useful information about the card itself.

It sounds like you don't understand the LMAD problem either.

It doesn't matter what the card is.

 

[davearm]

If we all got together with a deck of cards, a die, a coin, and a fantasy football roster:

We'd all agree that the probability of drawing a king is 1/13.

We'd all agree that the probability of rolling a 6 is 1/6.

We'd all agree that the probability of fiipping tails is 1/2.

We wouldn't all agree on the probability of Manning outscoring Flacco.

Does this have any bearing on the tanking question? Would tanking be OK if we all agreed on the probability of Manning outscoring Flacco, but since we don't all agree, tanking isn't OK? Surely nobody is forwarding that argument, right?

 

[Ignoratio Elenchi]

It's not a discrete value, though.

Let's say I deal you a card face down from a freshly shuffled deck. What are the odds that that card is a King? Now let's say I flip the card over and it's a 3. What are the odds that that card is a King?

It sounds like you don't understand the LMAD problem either.It doesn't matter what the card is.

Assuming "LMAD" stands for Let's Make A Deal, I'd wager a lot of money that Adam understands it just fine. And I don't see why you'd conclude he doesn't from what he wrote. In fact, it seems entirely consistent with the lesson of the LMAD problem, which deals with how probabilities change when new information is introduced. A face-down playing card has a 1/13 probability of being a King. A face-up 3 has a 0 probability of being a King.

 

[Bamac]

As much as I've enjoyed this thread, I think we've given the OP plenty of advice. Good work, everybody.

http://www.socwall.com/images/wallpapers/2653-1400x931.jpg

 

[Adam Harstad]

The probabilities are not made up nor a figment of your imagination when it comes to coin flips, rolls of the ice, or predicting cards. To infer otherwise is idiotic to say the least.

Odds are educated guesses about what will happen based on our knowledge of the makeup and history of the event we are predicting. They're made up, though. Again using the card example, if I deal you a card face down, you'll say there's a 1/13th chance it's a king. That's a made-up number based on your incomplete knowledge of the event. If I told you that the deck in my hands was a Pinochle deck, your estimate would change, despite the fact that it's the same card. If I flipped the card over, your estimate would change again (to 0% or 100%), but again, it's the same card. Odds are not a description of the card itself, they are a description of our own ignorance of the card itself. As that ignorance changes- as we get more or less information- the "odds" will change, despite the card remaining exactly the same. In that sense, "odds" tell us a whole lot about ourselves, and very little about the card.

 

[Adam Harstad]

I don't even know where to begin, so I'll just reiterate, for all practical purposes, the odds blackjack players use in their decisionmaking are the same as the odds fantasy football players use. They're both estimates.

Wrong. One is an estimate. The other is a discrete value that can be calculated.

It's not a discrete value, though.

Let's say I deal you a card face down from a freshly shuffled deck. What are the odds that that card is a King? Now let's say I flip the card over and it's a 3. What are the odds that that card is a King?

The card hasn't changed. It's the same now as it was before I flipped it over. The card was never a King, even when you estimated that there was a 1/13th chance there was. All the odds tell us is that if we played that game a billion times, you'd guess each time that you'd be right 1/13th of the time, and in the end you'd be right 1/13th of the time. The odds give us useful information about YOU- your level of knowledge and ignorance vis-a-vis the card- but they give us absolutely no useful information about the card itself.

It sounds like you don't understand the LMAD problem either.

It doesn't matter what the card is.

I'm assuming you're referring to the Monty Hall problem? I understand it very well. If Monty has no foreknowledge of what is behind the doors, and he happens to reveal a goat, then you should switch. If Monty has perfect foreknowledge of what is behind the doors and deliberately reveals a goat every time, then whether you switch or stay will have no impact on your odds. If Monty is a profit-maximizing entity with perfect foreknowledge, then the contestant should assume that he only reveals a goat when the contestant has already selected the car in an effort to trick them into switching, and the contestant should stay put.

The fact that Monty's foreknowledge of what's behind the doors actually has an impact on your odds of selecting the car by switching should provide the best illustration imaginable that odds are not an actual thing intrinsic to the event itself, but rather a human creation based on incomplete information.

 

[Adam Harstad]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

 

[Short Corner]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

How about we take a fair deck and I bet $5 against ur $15 that the next card is a spade. You happen to catch a glimpse of the card on the bottom which happens to be the ace of spades. Do you take the bet? What if the card was a non-spade?

 

[Ignoratio Elenchi]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

How about we take a fair deck and I bet $5 against ur $15 that the next card is a spade. You happen to catch a glimpse of the card on the bottom which happens to be the ace of spades. Do you take the bet? What if the card was a non-spade?

I won't answer the question for him, but I'm curious about this line of questioning. What has Adam said that would indicate to you that he doesn't understand how basic conditional probabilities work?

 

[Short Corner]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

How about we take a fair deck and I bet $5 against ur $15 that the next card is a spade. You happen to catch a glimpse of the card on the bottom which happens to be the ace of spades. Do you take the bet? What if the card was a non-spade?

I won't answer the question for him, but I'm curious about this line of questioning. What has Adam said that would indicate to you that he doesn't understand how basic conditional probabilities work?

His response to the LMAD problem for one:

If Monty has perfect foreknowledge of what is behind the doors and deliberately reveals a goat every time, then whether you switch or stay will have no impact on your odds.

This is exactly what does happen every time and changing every time changes your odds from 33% to 67%.

 

[Ignoratio Elenchi]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

How about we take a fair deck and I bet $5 against ur $15 that the next card is a spade. You happen to catch a glimpse of the card on the bottom which happens to be the ace of spades. Do you take the bet? What if the card was a non-spade?

I won't answer the question for him, but I'm curious about this line of questioning. What has Adam said that would indicate to you that he doesn't understand how basic conditional probabilities work?

His response to the LMAD problem for one:

If Monty has perfect foreknowledge of what is behind the doors and deliberately reveals a goat every time, then whether you switch or stay will have no impact on your odds.

This is exactly what does happen every time and changing every time changes your odds from 33% to 67%.

I didn't even read his response to that very closely since I assumed he got it right. That's what I get for assuming things, I guess.

Having said that, I'm going to double down and assume that he does actually understand how it works and just misspoke. Adam's a bright guy, you guys can disagree on the tanking or whatever but I don't think testing his probability skills will wind up being very fruitful. But I'll stop sticking my nose in.

 

[Short Corner]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

How about we take a fair deck and I bet $5 against ur $15 that the next card is a spade. You happen to catch a glimpse of the card on the bottom which happens to be the ace of spades. Do you take the bet? What if the card was a non-spade?

I won't answer the question for him, but I'm curious about this line of questioning. What has Adam said that would indicate to you that he doesn't understand how basic conditional probabilities work?

His response to the LMAD problem for one:

If Monty has perfect foreknowledge of what is behind the doors and deliberately reveals a goat every time, then whether you switch or stay will have no impact on your odds.

This is exactly what does happen every time and changing every time changes your odds from 33% to 67%.

I didn't even read his response to that very closely since I assumed he got it right. That's what I get for assuming things, I guess.

Having said that, I'm going to double down and assume that he does actually understand how it works and just misspoke. Adam's a bright guy, you guys can disagree on the tanking or whatever but I don't think testing his probability skills will wind up being very fruitful. But I'll stop sticking my nose in.

I jumped in when a comparison was made between FF and BJ with regard to probability and expectation.

I am against tanking FWIW.

 

[Adam Harstad]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

How about we take a fair deck and I bet $5 against ur $15 that the next card is a spade. You happen to catch a glimpse of the card on the bottom which happens to be the ace of spades. Do you take the bet? What if the card was a non-spade?

I won't answer the question for him, but I'm curious about this line of questioning. What has Adam said that would indicate to you that he doesn't understand how basic conditional probabilities work?

His response to the LMAD problem for one:

If Monty has perfect foreknowledge of what is behind the doors and deliberately reveals a goat every time, then whether you switch or stay will have no impact on your odds.

This is exactly what does happen every time and changing every time changes your odds from 33% to 67%.

IE was right that I misspoke; I had the scenarios reversed. If Monty Hall has foreknowledge, you should switch. If he is opening doors at random, switching does you no good.

Scenario #1:

You pick a door. Monty Hall is then compelled to open a second door at random (say, the farthest unselected door to the left).

Scenario #2:

You pick a door. Monty Hall is then compelled to open a second door, but the door is not selected at random (say, he deliberately selects a goat).

Scenario #3:

You pick a door. Monty Hall then decides whether or not he wants to open another door.

In scenario #1, assuming the opened door reveals a goat, switching does nothing for you. In scenario #2, assuming that Monty Hall deliberately reveals a goat every time, switching improves your odds. In scenario #3, whether switching helps you or hurts you depends on Monty Hall's motivations.

If I were a diabolical game show host and I did not want my contestants to win, then I would bring in nobody but mathematicians as contestants, and I would only open another door if they selected the car with their first guess. Every time they picked a goat initially, I would hold them to it and stick them with the goat. Every time they picked the car initially, I would trick them into switching to a goat.

Edit to add: You select a door. Monty reveals a goat. You decide to switch. I ask Monty "did you know what was behind that door before you opened it?" If Monty says yes, your odds of winning a car are 67%. If Monty says no, your odds of winning a car are 33%. If this doesn't illustrate my point that there's no such thing as "real odds" or "actual odds", then I don't know what does- what Monty did or did not know changes the odds that a car is behind the door you wound up with. Odds are not a measurement of reality, they are a measurement of our ignorance of reality.

 

[Short Corner]

Another example. I deal a card face down into the middle of the table from a standard deck. I then show Bamac the next three cards in the deck. I then show Elenchi the five cards after that. I then show ghostguy the next 8 cards, and I show Short Corner the 10 cards after that. After that, I look at the rest of the deck, and davearm takes a peek at the card in the middle of the table. What are the odds that the card in the middle of the table is a king? Well, that's going to depend on who you ask, isn't it?

Exactly.

However, in BJ, everyone has access to the same information so the odds are known by those willing to learn the system and have the attention span to count. What you described is more of what you get from FF prognosticators, except the information they are using is still speculatory at best. Some speculate better than others, but it is not the concrete knowledge of what cards are already out of a BJ shoe.

 

[Short Corner]

Edit to add: You select a door. Monty reveals a goat. You decide to switch. I ask Monty "did you know what was behind that door before you opened it?" If Monty says yes, your odds of winning a car are 67%. If Monty says no, your odds of winning a car are 33%. If this doesn't illustrate my point that there's no such thing as "real odds" or "actual odds", then I don't know what does- what Monty did or did not know changes the odds that a car is behind the door you wound up with. Odds are not a measurement of reality, they are a measurement of our ignorance of reality.

They were never intended to be a measure of reality. They were intended to quantify expectation.

You still don't have it right, btw, the odds in the bolded scenario are 50%

 

[Adam Harstad]

Edit to add: You select a door. Monty reveals a goat. You decide to switch. I ask Monty "did you know what was behind that door before you opened it?" If Monty says yes, your odds of winning a car are 67%. If Monty says no, your odds of winning a car are 33%. If this doesn't illustrate my point that there's no such thing as "real odds" or "actual odds", then I don't know what does- what Monty did or did not know changes the odds that a car is behind the door you wound up with. Odds are not a measurement of reality, they are a measurement of our ignorance of reality.

They were never intended to be a measure of reality. They were intended to quantify expectation.

You still don't have it right, btw, the odds in the bolded scenario are 50%

Cool. Because I believe that's what I've spent the last two pages saying. Odds are not a measure of reality, and there's no such thing as "real" odds vs. "perceived" odds. All odds are perceived odds, even the real ones.

Apologies on getting the specific calculations in the Monty Hall problem wrong, and I'll gladly blame it on faulty heuristics. I think my broader understanding of the problem and its implications are just fine, though.