Revel promotion sounds too good to be true
Quote: randomperson
That is proof, it's proof he cost everyone ev. We don't know how much ev, but my estimation is it's substantial.
The point is that we don't know if the casino would have done this anyway, but there is some chance that they wouldn't have and so we adjust the cost for the nonzero probability that the message board and gawe matter.
I'm not wasting time. We have to show everyone that posts online have consequences. We had a bunch of people that never intended to play the promo costing everyone who did intend to play the promo. We should make sure people know what happens when they do that so they don't do it again.
It's not proof of anything, it's conjecture. If they were going to pull the machines anyway, then this message board did not cost anyone anything. If they were not going to pull the machines anyway, and did it only as a result of this message board, then you have a point.
Quote: Mission146It's not proof of anything, it's conjecture. If they were going to pull the machines anyway, then this message board did not cost anyone anything. If they were not going to pull the machines anyway, and did it only as a result of this message board, then you have a point.
And because we will never know, as rational bayesians, we have to form beliefs about the probabilities of unknown events, and those beliefs should inform our actions.
If you agree the probability is greater than zero, which you pretty much just did, then you agree there was a cost.
"Well, your Honor, he could have killed him, is my point. Of course, there are twenty other people who have a probability greater than zero of killing him, so let's give them all the chair."
Quote: Mission146No, I agree that there exists the potential for a cost. When referring to events that have actually occurred, I prefer the factual to the hypothetical. Imagine if somehow the AP community could sue for the perceived lost EV on this promotion, the case wouldn't have a chance based on your presentment.
"Well, your Honor, he could have killed him, is my point. Of course, there are twenty other people who have a probability greater than zero of killing him, so let's give them all the chair."
I agree with your proposal to move toward a utilitarian court system. However, you are obviously mixing analogies because the court system has a beyond a reasonable doubt or preponderance of the evidence standard. Gambling and correct thinking doesn't have the same standard. We believe things that we aren't sure of and let information update our beliefs when we get new details.
Mission, there is a cost. The cost can't be known, but it can be approximated by reasonable assumptions of the posterior probabilities. For example, "if there is a 10% chance that Revel changed its promotion based on the posts on this website, then ..." I do give credit to the AP who stated he had inside information that may allow him to estimate the posterior probability. There is no reason to doubt that. My experience with high-level APs is that they are very honest and careful about statements such as that.Quote: Mission146No, I agree that there exists the potential for a cost. When referring to events that have actually occurred, I prefer the factual to the hypothetical.
From Wikipedia's entry on posterior probabilities: In Bayesian statistics, the posterior probability of a random event or an uncertain proposition is the conditional probability that is assigned after the relevant evidence is taken into account."
In this case, the conditional probability is the probability that Revel would have removed the VP given that no posts were made on the subject here. Once that posterior is estimated, the approximate cost to the AP of the posts here can be determined. From my point of view, I think it is a good thing to discuss such matters openly. It allows everyone equal access to the information.
Mission, you are approaching this problem from the "Schrodinger's cat" paradigm. Instead, you should read up on Bayesian statistics.
Quote: kvitlekhAnother indication that it wasn't this thread that caused Revel to remove the big VP machines is that the $100 slots are still there. They could have just made all machines $5 max for a month.
Ya the only game there is math for in the thread is one of the few removed and this is evidence against this thread having a causal impact?
I think it's probably evidence the radio show mattered more than the thread but I'm not seeing your logic here.
Quote: randompersonNo fill me in.
Go to DGE, and CCC websites. It is all spelled. Don't be lazy. If you want your information, and strategies to come from boards, be prepared that the information becomes public.
You haven't lost a dime.
Quote: BhappyGo to DGE, and CCC websites. It is all spelled. Don't be lazy. If you want your information, and strategies to come from boards, be prepared that the information becomes public.
You haven't lost a dime.
Ok great, he had some sort of a point in asking the question besides a classic data dump?
While we'll never know what would have happened if I never contributed or busted the whole thread, at the time I started posting I felt the player community had more to gain by an open discussion about it. Yes, there was a possibility the Revel would water down the promotion earlier than they otherwise would have. There is also the possibility that many more players would play it, and know how to play it properly, with no change in behavior from the Revel. My goal was to maximize total player win from this.
It is easy to Monday morning quarterback this and blame me. It think it was John Madden who would say when a team would make an unconventional decision, "You're a genius if it works and an idiot if doesn't." I've always hated that comment. In judging a decision, it doesn't matter whether the play succeeded or not, it matters whether it lead the higher probability of winning the game. All I can say in this case is I thought I was maximizing the total amount players would win by talking about it and I make no apologies for trying to help.
I'd also like to add I got a lot of criticism for writing about the Mohegan Sun triple down promotion, and that went the full 24 hours without any watering down of the rules. Thanks to me additional players made money from that and many personally thanked me.
In the end, my philosophy is that the world would be a better place with unfettered access to truthful public information.
"Knowledge is good." -- Emil Faber
Quote: WizardI make no apologies for trying to help my readers and radio listeners make as much money as possible on this promotion. Let me remind everybody that I didn't start this thread and was late coming into it. I finally read up on the promotion and analyzed it after several people Emailed and PM'd me asking me to contribute to the discussion.
While we'll never know what would have happened if I never contributed or busted the whole thread, at the time I started posting I felt the player community had more to gain by an open discussion about it. Yes, there was a possibility the Revel would water down the promotion earlier than they otherwise would have. There is also the possibility that many more players would play it, and know how to play it properly, with no change in behavior from the Revel. My goal was to maximize total player win from this.
It is easy to Monday morning quarterback this and blame me. It think it was John Madden who would say when a team would make an unconventional decision, "You're a genius if it works and an idiot if doesn't." I've always hated that comment. In judging a decision, it doesn't matter whether the play succeeded or not, it matters whether it lead the higher probability of winning the game. All I can say in this case is I thought I was maximizing the total amount players would win by talking about it and I make no apologies for trying to help.
I'd also like to add I got a lot of criticism for writing about the Mohegan Sun triple down promotion, and that went the full 24 hours without any watering down of the rules. Thanks to me additional players made money from that and many personally thanked me.
In the end, my philosophy is that the world would be a better place with unfettered access to truthful public information.
"Knowledge is good." -- Emil Faber
Would it have made players more money to just pm the details to those with a certain number of posts that were interested? If so, I don't think you accomplished your stated goal. Not to mention the goal of completely open information eliminates all ap.
Quote: randompersonWould it have made players more money to just pm the details to those with a certain number of posts that were interested?
We'll never know.
Quote:Not to mention the goal of completely open information eliminates all ap.
How so? Blackjack and poker have been written about to death and AP's still practice both of those games.
Quote: WizardWe'll never know.
How so? Blackjack and poker have been written about to death and AP's still practice both of those games.
Perfect information in poker means I know my opponents whole cards. It means I know the ev of all players in the game including myself.
Perfect information in blackjack means the casino knows the ev of all players, which ones are losers and which ones are aps.
Quote: randompersonPerfect information in poker means I know my opponents whole cards. It means I know the ev of all players in the game including myself.
Perfect information in blackjack means the casino knows the ev of all players, which ones are losers and which ones are aps.
What I said was, "the world would be a better place with unfettered access to truthful public information." Hole cards in poker and the dealer's hole card in blackjack are intrinsic to each game. That is not "public" information.
Quote: WizardI believe the world would be a better place if there were unfettered access to truthful information, and I aim to provide that.
If that is a slip up why stop at public information? Why is the world better with one type of information but worse with another?
Personally, I think the statement that the world is better off with all information known is true. That doesn't mean aps win more money in that world. There is an inherent conflict, you have to pick the idolized world or the existence of ap. The ap uses hidden information to make money.
The casino ap game is a zero sum game, it's not really value adding.
My feeling is that you care more about morality than advantage play, and the talk otherwise is self contradictory.
Quote: Wizard
I'd also like to add I got a lot of criticism for writing about the Mohegan Sun triple down promotion, and that went the full 24 hours without any watering down of the rules. Thanks to me additional players made money from that and many personally thanked me.
In the end, my philosophy is that the world would be a better place with unfettered access to truthful public information.
I blame you for the MS fiasco, WIZ. I blame you for not talking more about it and getting it cancelled! lol From reading the message boards in the immediate aftermath I think I am the only AP that lost money that day. I took the train and because of train schedule limitations only played 19 hours, over bet, lost big early and spent most of the day digging out getting back towards even. I didn't mind losing, but I did mind reading about how much money everyone else made, while I seemed to be that 1 in a hundred person who didn't make money. lol
+2. I live by that.Quote: Wizard"Knowledge is good." -- Emil Faber
I think you are overdue for tenure with an attitude like that. +2.Quote: Wizardthe world would be a better place with unfettered access to truthful public information
The count in blackjack is not hidden in any shape, manner or form whatsoever. It is freely available for one and all to see and work with.Quote: randompersonThe ap uses hidden information to make money.
Quote: SanchoPanzaThe count in blackjack is not hidden in any shape, manner or form whatsoever. It is freely available for one and all to see and work with.
The ap has hidden information like someone selling a car who knows if its damaged when the seller doesn't. They know their own skills better than the pit boss and try to hide that information from being revealed, as evidenced by the extreme amount of effort put into cover plays. What exactly are they covering?
CarFax has radically diminished that ploy.Quote: randompersonThe ap has hidden information like someone selling a car who knows if its damaged when the seller doesn't.
It's all one big show, and the dupes eat it up. We even had seen quite a bit of that evident right here.Quote: randompersonThey know their own skills better than the pit boss and try to hide that information from being revealed, as evidenced by the extreme amount of effort put into cover plays. What exactly are they covering?
Quote: whatmeSo the real Q is what is this promo worth on a $5DDB w/98.98% pay back?
World could use a 5 minute course in excel. Teach a man to fish.
Quote: Mission146For example, there is still a ton of value in those $100/denom slots.
Great, another one who hasn't learned a thing.
Quote: kvitlekhWhat about the other forums which openly discuss which BJ games are beatable (in great detail)....
People shouldn't be discussing particularly good specific blackjack games in public, either. At many sites, anyone who puts up something along the lines of, "Casino Tangiers is cutting off only a quarter deck on their six deck shoes!" will rightly get slammed.
Quote: SanchoPanzaCarFax has radically diminished that ploy.
It's all one big show, and the dupes eat it up. We even had seen quite a bit of that evident right here.
You do understand that trying to say asymmetric information is not a problem in no way refutes the existence of said asymmetric information in blackjack right? It does the opposite, if you need to develop countermeasures against it, then it exists.
Quote: WizardIn the end, my philosophy is that the world would be a better place with unfettered access to truthful public information.
In this case, what endangered the survival of the play wasn't the fact that its existence was publicized. Revel was already doing a great job of that. It wasn't the fact that the possibility of it being highly valuable was communicated to a lot of people. That would already have been immediately obvious to those with the learning, money, and guts to go after it. It wasn't the fact that detailed information on the best play was disseminated. That same rather small group with the resources and experience to do the casino serious damage would have found a way to develop the information anyway. What endangered the play was the high visibility of information explaining exactly how to beat it best, the stark numbers, and the mobs cheering about how much money there was to be made. All of this was calculated to scare the crap out of casino management. We can't know exactly what conversations went on between management staff, and what was in their minds. But we know that casinos do read message boards, that they do respond to what they see, and that there is a risk that what is written publicly will influence them. Oh, you could say that it doesn't matter, that Eliot Jacobson was probably already knocking on the Revel general manager's door with dire warnings. But casinos don't always listen to advice from experts. Just ask Don Johnson.
What has happened here shouldn't be laid at the feet of any one individual, or any one Web site. It was a group effort. A worthy goal for APs would be to avoid scaring the crap out of casinos.
Quote: Bhappycome on guys wizard did not do anything wrong. Randy Fine is a very intelligent person. Only a moron would not think of the potential liability of high denomination VP machines.
100% correct. Casinos hire people like the Wizard to show them potential liabilities and probably had this planned before announcing the promotion.
Quote: kewljFrom reading the message boards in the immediate aftermath I think I am the only AP that lost money that day. I took the train and because of train schedule limitations only played 19 hours, over bet, lost big early and spent most of the day digging out getting back towards even. I didn't mind losing, but I did mind reading about how much money everyone else made, while I seemed to be that 1 in a hundred person who didn't make money. lol
Kewl, if it makes you feel any better, I know several individuals and groups who lost money at the MS promo.
Quote: teliotMission, there is a cost. The cost can't be known, but it can be approximated by reasonable assumptions of the posterior probabilities. For example, "if there is a 10% chance that Revel changed its promotion based on the posts on this website, then ..." I do give credit to the AP who stated he had inside information that may allow him to estimate the posterior probability. There is no reason to doubt that. My experience with high-level APs is that they are very honest and careful about statements such as that.
From Wikipedia's entry on posterior probabilities: In Bayesian statistics, the posterior probability of a random event or an uncertain proposition is the conditional probability that is assigned after the relevant evidence is taken into account."
In this case, the conditional probability is the probability that Revel would have removed the VP given that no posts were made on the subject here. Once that posterior is estimated, the approximate cost to the AP of the posts here can be determined. From my point of view, I think it is a good thing to discuss such matters openly. It allows everyone equal access to the information.
Mission, you are approaching this problem from the "Schrodinger's cat" paradigm. Instead, you should read up on Bayesian statistics.
I understand your position, but respectfully find it inapplicable, in this case.
You have, "The Revel was going to pull them anyway," you have, "Gambling with an Edge," you have, "WoV," and then you have everything else. Even if you wanted to determine probabilities for the first three, you cannot possibly determine probabilities for, "Everything else," because it would be impossible to even conjecture what, "Everything else," would consist of. You can't even attempt to assign probabilities to events without first knowing what every individual possibility that caused something is, in my opinion.
You are more learned than I am, certainly educationally and mathematically...and you may or may not be a better Philosopher...we haven't really had occasion to discuss Philosophy. However, I am somewhat learned in Economics, and if someone gave me a hard number that Revel pulling the machines cost the AP's and asked me to assign costs to each individual event that could have been a factor, I'd simply say, "To request such a breakdown is nonsense."
Also look at the probability that 'YOU' will get to play on those two machines. I seem to remember from one of the posts that you would have to grind for about 90 hours. Assuming 100% non stop utilization, 2*24*31/90 = only 16.5 people would be able play on those machines. Now with that known number start assigning Baysian or any other probabilities to number of people who would eventually make any money.
Quote: BhappyI don't understand some of the people here. They come on a free site, they ask questions, they learn and then they think others should not know the open secret. Half of the people do not have the analytic mathematical abilities to fully understand the probabilities of success. Wizard does them a favor by providing them a 'free' educational service. Perhaps those grumblers should retain Wizard for a fee, run personal simulations for them, and then ask him to shut up. Again, Randy Fine is not a moron. He did not create this promotion on a whim. He certainly has resources to perform similar analysis, and eliminate weak points. Of course it was Revel's money, but he had to put his reputation (and future income) on line for this promotion to be either revenue neutral or make money off it. For those who believe this site or other sites contributed for those two lucrative machine to be pulled off, I have land to sell in Louisiana swamps.
Also look at the probability that 'YOU' will get to play on those two machines. I seem to remember from one of the posts that you would have to grind for about 90 hours. Assuming 100% non stop utilization, 2*24*31/90 = only 16.5 people would be able play on those machines. Now with that known number start assigning Baysian or any other probabilities to number of people who would eventually make any money.
I'm pretty sure the people asking questions and the people wanting to stop the thread are not the same.
And no one has addressed the idea that we can communicate with the questioners behind a wall rather than in public view.
Quote: Mission146Quote: teliotMission, there is a cost. The cost can't be known, but it can be approximated by reasonable assumptions of the posterior probabilities. For example, "if there is a 10% chance that Revel changed its promotion based on the posts on this website, then ..." I do give credit to the AP who stated he had inside information that may allow him to estimate the posterior probability. There is no reason to doubt that. My experience with high-level APs is that they are very honest and careful about statements such as that.
From Wikipedia's entry on posterior probabilities: In Bayesian statistics, the posterior probability of a random event or an uncertain proposition is the conditional probability that is assigned after the relevant evidence is taken into account."
In this case, the conditional probability is the probability that Revel would have removed the VP given that no posts were made on the subject here. Once that posterior is estimated, the approximate cost to the AP of the posts here can be determined. From my point of view, I think it is a good thing to discuss such matters openly. It allows everyone equal access to the information.
Mission, you are approaching this problem from the "Schrodinger's cat" paradigm. Instead, you should read up on Bayesian statistics.
I understand your position, but respectfully find it inapplicable, in this case.
You have, "The Revel was going to pull them anyway," you have, "Gambling with an Edge," you have, "WoV," and then you have everything else. Even if you wanted to determine probabilities for the first three, you cannot possibly determine probabilities for, "Everything else," because it would be impossible to even conjecture what, "Everything else," would consist of. You can't even attempt to assign probabilities to events without first knowing what every individual possibility that caused something is, in my opinion.
You are more learned than I am, certainly educationally and mathematically...and you may or may not be a better Philosopher...we haven't really had occasion to discuss Philosophy. However, I am somewhat learned in Economics, and if someone gave me a hard number that Revel pulling the machines cost the AP's and asked me to assign costs to each individual event that could have been a factor, I'd simply say, "To request such a breakdown is nonsense."
So you never act in situations with imperfect information? You never form beliefs about things that you can never know for sure? My bet is you do it every day.
Quote: randompersonI'm pretty sure the people asking questions and the people wanting to stop the thread are not the same.
And no one has addressed the idea that we can communicate with the questioners behind a wall rather than in public view.
why are you here in an open forum? You do seem take advantage of open forums. I have seen your screen name (& I am assuming it was you) in one of the other forums where you did ask for information. So if you are in a information gathering mode (trivial or non trivial) from open forums, you can not dictate what information others should ask, and what others should provide, and what the site owner should do.
Quote: randomperson
So you never act in situations with imperfect information? You never form beliefs about things that you can never know for sure? My bet is you do it every day.
I don't know about Mission, but you are asking two different things. Yes I do act in situations with imperfect information. That is called educated guess, or hunch or whatever - that is logical. No I never form beliefs about things I am not sure of (except religion, and God). It is not logical.
Quote: 21formeKewl, if it makes you feel any better, I know several individuals and groups who lost money at the MS promo.
A team at my table was down $50K before noon. Then it really got ugly.
You weren't alone. I specifically flew home from a business trip in order to take advantage of the triple down at Mohegan, had about 10 hours spreading $50 to three hands of $150 (if I remember correctly, there was a $500 cap on tripling down) and lost just under $10k. Had the opportunity to meet several WOVers, including Ben who did not get backed off, but wonged around my pit.
I seem to remember there was consternation about not discussing the details the +EV opportunity, but there were plenty of spots at plenty of tables.
Wish they would bring it back.
Quote: BhappyI don't know about Mission, but you are asking two different things. Yes I do act in situations with imperfect information. That is called educated guess, or hunch or whatever - that is logical. No I never form beliefs about things I am not sure of (except religion, and God). It is not logical.
Belief in the Bayesian sense, like for example: there is a probability x this person is lying.
Quote: Bhappywhy are you here in an open forum? You do seem take advantage of open forums. I have seen your screen name (& I am assuming it was you) in one of the other forums where you did ask for information. So if you are in a information gathering mode (trivial or non trivial) from open forums, you can not dictate what information others should ask, and what others should provide, and what the site owner should do.
I'm here to try to convince some people they shouldn't cost other people money. No morality, just selfishness. There are ways to communicate details in private and get the information out there without ruining the play.
Using the formula below on $25 ($125 per spin) DDB, with a loss limit of $100k (800 units), then the optimal stopping point is a win of 1343 units, which gives an expected win of 349.5 units ($43687).
This was very quick and easy. It's not as accurate as simulation because the distribution is defined only by its first two moments. But, it was certainly a lot easier than the massive simulations I ran.
The probability of hitting the upper limit b before the bottom limit 0 starting at x, 0 < x < b:
p = [exp(-2xμ/ σ^2) - 1] / [exp(-2bμ/ σ^2) - 1]
x = 800 (units)
μ = -0.010192
σ^2 =42.366
b = unknown upper stopping point
Expected win = p*(b - 800)
The average completion time can also be computed using a very easy formula. For the situation above, the formula gives 23730 spins.
[Edit. 07/01/2013 -- it looks like I got the variance wrong in this -- it should be 41.984981. In this case I show a win exit point of $165750 (1326 units), an expected win of $43,531 (348 units), and 23709 spins].
Optimal win goal = $49,000 (1960 units).
Probability of success = 35.27%.
Expected win = $17,280 (691 units).
Expected time = 186241 spins.
At a rate of 500 spins per hour, this will take 372 hours. This amounts to a theoretical loss of $46.40 per hour per $25 machine for Revel. The per day/per machine cost to Revel is $1,113. Multiply this by the number of days and number of machines (run a similar analysis for the other VP), and an upper bound on Revel's promotional cost can be approximated for all their $25 video poker.
For those who don't have 372 extra hours:
With a lesser win goal of $7,500 (300 units).
Probability of success = 84.62%.
Expected win = $6,346 (257 units).
Expected time = 35471 spins.
LoLQuote: teliotFor those who may still be interested (including Revel), using the formula above, playing $5 DDB 9/6 ($25 bet per hand), the formula above gives:
Optimal win goal = $49,000 (1960 units).
Probability of success = 35.27%.
Expected win = $17,280 (691 units).
Expected time = 186241 spins.
At a rate of 500 spins per hour, this will take 372 hours. This amounts to a theoretical loss of $46.40 per hour per $25 machine for Revel. The per day/per machine cost to Revel is $1,113. Multiply this by the number of days and number of machines (run a similar analysis for the other VP), and an upper bound on Revel's promotional cost can be approximated for all their $25 video poker.
For those who don't have 372 extra hours:
With a lesser win goal of $7,500 (300 units).
Probability of success = 84.62%.
Expected win = $6,346 (257 units).
Expected time = 35471 spins.
500 spins is easier then you think, wen it comes to job you will hear 700-800 spins p/hr. The more you think the fewer hands p/hrQuote: Bhappyis 500 spins/hr a realistic spin rate? That is about 7 sec/spin. You may get that in first few hrs, but then fatigue sets in.
so 500 for ddb is realistic.
