Roulette I've seen a new absurd claim that the gambler's fallacy isn't.....

err

 

Gizmotron what is Dr. Sr A A recruiting for ?


A new crop of BA has arrived and looking for a place in the lime light .


Just an observation.

 

Ok Trans, TwoUp and Turbo, let's simplify things..
Calculate the binomial distribution for reds and blacks over 10 spins.
Now suppose the wheel spins 9 blacks.
Would you bet red? Black? Of course it makes no difference.
Because calculating probabilities is useful for FUTURE events, but the past is CERTAIN.
Basically the past does not come into the calculation. That's the simplest way I can put it.
It's an easy mistake to make.
Don't knock Sir Anyone. he's only trying to tell you!

 

"Dorks. They look like dorks."

Dr. Snowman here is trashing the forum as Captain Obvious so that he can contact the fresh flock of flying suckers through private messaging. He's counting on you to be so impressed by his superior logic that you will do battle for him in your own geographical region's casinos. You see. There is only one way to get an advantage. You will search for and collect data on actual spins in search of "broke dick" wheels. Then you get to be his friend. Once you realize that you have been had then you get to be his punching bag. If you are very lucky you get to be honored by the suggestion that you are an idiot for rejecting him by inclusion in his wonderful collection of disparaging artwork. He has not changed this tactic in more than 15 years. Welcome to the new age of boring enterprises.

 

That's ludicrous.

lu·di·crous
/ˈlo͞odəkrəs/
Learn to pronounce
adjective
so foolish, unreasonable, or out of place as to be amusing; ridiculous.

 

Yes it is all fallacy, and you caught me out. I still can't wonder why these methods are used everywhere, including sampling a critical number of voters to predict election polls.

You better go tell everyone that Binomial distribution calculations and confidence intervals are just a fantasy.

You either disagree that the probability of the ball landing on any pocket is constant or your saying the probability varies. The math follows from one of those two positions.

If the probability is not constant then that will be exploitable by observing outcomes and results will be skewed and the house will lose its edge. If however the probability is constant because the house wants it's edge then the binomial distribution calculation is valid, it tells us how many events we can expect in a given number of trials and the degree of certainty, no if's or buts. In the later case we look for those betting opportunities and manage our bet sizing relative to the degree of certainty.

So pick one.

 

@Luckyfella: It's out of the question that this Mr. anyone guy is neither smart nor educated (and probably writing from his recliner in a nursing home). Unfortunately your claim is as unfounded as his. Which makes you sound like - hmm- exactly like him with less words...
Comparing non existing dicks is pointless.

 

You have a 99.87% of seeing at least one red or black (pick a side) in 10 spins on a European wheel.

So after 5 spins you still have a 99.87% of seeing at least one red or black. And after 9 spins you still have a 99.87% confidence of seeing at least one red or black in the series of 10 spins. You ALWAYS have a 99.87% probability of seeing at least one red or black in 10 spins with 99.87% confidence. So you would expect to be right 99.87% of the time over the long haul.

These calucations are used everywhere in science, quality management and medicine. When clinical trials are conducted you measure the occurances of outcomes. Just because you got 9 of the 10 results doesn't effect the tenth. Or 900 out of 1000 or 9000 out of 10,000, more trials means more confidence. The house edge is predicated on this working, the house also knows the maximum number of losing events it will sustain based on the odds, over a series of trials using the same calculation.

The binomial distribution calculation provides the probability OVER a set of outcomes as it's a distribution calculation, it is NOT a probability of a single event which is assumed to be constant 1/37 and fundamental to the binomial distribution calculation.

Need to get your probabilities right.

 

Knowing the probability still doesn't give you the edge on the next spin, or the next series of spins. Again, more gambler's fallacy. The probability of winning on the next spin or next series of spins does NOT change based on the past spins. Math will not help you beat the random game of roulette. It won't even make a dent in the house edge.

 

The binomial distribution of the random game of roulette only proves that you can't win in the long run. It can NOT help you win. Even a child can count the number of hits and discover that there aren't enough of them to win. Why can't you?

 

Echo .. echo ... echo ....

Clearly no bookmaker is silly enough to take action on a parlay bet of Dr. Sir Anyone Anyone mentioning "dent", "house" and "edge".

 

I provided the calculated probability of 99.87% for witnessing at least 1 EC outcome (18/37 probability) within 10 spins using the binomial distribution function.

Do you agree or disagree? When you perform that same calculation, what is your answer?

 

Fallacy or not, almost every player uses 'past spin records' in one form or another to choose their bets. It's how they use their 'past bet records' to lay their bets that makes them winners or losers.

 

TwoUp, I know binomial probability is not about a singular event. It's about future events.
Unfortunately, I can't say to the croupier "If I see 12 repeats in the next 37 spins, give me some money!"

 

True you can't ask the croupier to give you money and expect them to comply.

However the probability of the minimum number of events in a series of trials with a defined certainty is still a very useful tool for understanding worst case when evaluating and designing staking plans.

Also remember it is not just about future events, it's about any set of independent trials will have a distribution of outcomes based on the inherent probability (18/37) and a probability/confidence value for each distribution count.

Hence why winning 10 out 10 EC events is possible but low probability whilst winning 1 out of 10 is a higher probability. However "high" is relative, given 10 events is not a large sample set we expect to see it fail and not see 1 win once every 784 times and hold true the other times.

So can you design a bet selection, staking plan and money management strategy that when the failure event happens where the 1 in 10 event does not hold, the loss is less than the total gains up to that point?

So think about:

• how many success events does it take to double a bankroll?
• What do you need to do with unit size to double a bankroll before the nemises event?
• What do you need to do with bankroll management before the nemises event so it's just a set back and you have reserves locked in?

 

Yes you can analyse past trials/spins. But they have no bearing on the future, at least not in roulette. Unless you have a wonky wheel.

I would like to pose a question I hope you guys could put your heads together and answer. I have asked before but no-one could come up with an answer..
Suppose we are hoping for at least x wins in y spins (EC.)
For example 40 wins in 100 spins. How do we bet with the least risk?
I know the labouchere wins if we have a third+1 wins, but what about a more general x wins in y spins? Is it possible to write a formula for this?
I have asked this on a maths forum, and even there no-one could crack it.
Would really appreciate it if any of you guys could help, as I'm curious whether it can be done.

 

You guys crack me up. One says you can't win and others say you can. And it all comes down to what you must get in the long run. Just about every book written regarding gambling addiction puppets the same notions too. They all say that you can get lucky in the short term but that you will lose at the expected rate eventually. They mean you will eventually give it all back, to that 5% house's edge expectation.

But there is a problem with that. You have been lied to and the whole world has accepted that lie. But something happened that wipes out that lie and exposes it for what it is. The gangsters lost control of the casinos several decades ago. The casinos where sold off and became incorporated and began to sell stocks in their ownership. And an inconvenient problem came up in them going legit. They must file a year end public statement of profit or loss from their gambling activities. You can look them up on line for those reports.

So I did, and guess what. The casinos make around 14% to 22% off of their table games where the math says that they should only make around 5% on average. Busted. Player ignorance accounts for more than your Binomial distribution calculations predict it should. And a savvy player wrecks that because they don't obey your magical math either. So it's time to figure this out. Math does not have magical powers. The odds might say that this or that is what your odds are on the next spin. But those same odds do not predict if you will win or lose on the next spin. Funny how you are supposed to use confirmation bias to make sense when it is convenient. Not so funny when it contradicts itself.

 

When you write the math formula for this you will get a prize of some kind.

"Bet big when you are doing good, and bet small any other time." This requires the skill of situational awareness. Math will never tell you when you will be doing good or when it will be any other condition. Now this almost always brings up more magical thinking by math oriented magical thinkers. They try to insist by some kind of unwritten law of confirmation bias that nobody can know when they are doing good, like a win streak. But guess what sports fans? I have never seen a pit boss miss one. They always come over to observe the play. You see. If they do miss one then it will come up in their employee evaluation and / or possibly a corporate meeting of executives. Funny how casinos take luck seriously and mathZombies don't.

 

Stop with the future crap again, this is the go to argument by mathZombies. What about them having bearing on the present? Why do you guys always put on the blinders and ignore the situational awareness? You can confirm that you are in a win streak in the present tense. You can also see evidence once it ends, likely your first loss.

Now let's see you pull the wool over your own eyes or run and hide like always happens. What magical conclusion will dig you out of this dilemma?

 

The binomial calculation is exactly that kind of worst case X wins in y spins and tells you the probability of how often it will hold (i.e. how confident we can be).

So I've already given figures for roulette EC of 60/210 with a failure rate of 1 in a billion. If we relax that to 70 in 210 (exactly 1/3) then we have a 1 in 387,000 failure rate. The 1/3 win rate fits the win rate of a basic labby.

Other approaches are possible as well.