Calling All Lotto Players

[snarzberry]

re the martingale betting discussion (that's what these doubling up systems are called in the casino world), it's not a guaranteed way to make money which is thwarted by the casino having limits. It wouldn't matter if there were no limits, it has a negative expected return either way.

You can arrive at that conclusion very quickly when you consider that, taking the red/black roulette scenario on a non-American wheel (the kind with only 1 zero), the outcome you are betting on pays 1 - 1 and has 18/37 chances to win and the outcome the casino is betting on has 19/37. Obviously this is a negative expected return no matter how much you bet. It's similar to you and I tossing coins where if I win you pay me $1 and if you win I pay you $0.97.

By increasing bets in later rounds all you are doing is placing further negative expectation bets. negative + negative + negative... generally never adds up to a positive.

From the casino's perspective they're not betting against 'you' individually, it's just a bunch of bets on the table that all have negative returns from the players perspective. They couldn't give a toss if you're doubling up amounts from previous bets trying to win back your losses and make a small profit. All they care is that you're willing to place a bet on this particular round and you're willing to accept a payout which is somewhat less than the true odds.

I worked as a croupier for 6 years and ran into this malarkey all the time, it is nonsense. Let's imagine that the casino has no high limit on red/black and a low limit of ten bucks. After ten losses you're betting ten grand a spin, after twenty losses it's ten million per spin. Even if bill gates played this game for long enough he'd go broke.

The lotto scenario is even darker news for the player as the true odds of a number coming up compared to payout is a worse ratio than any casino game.

 

[Tarquelne]

The key to the martingale system is having infinite amounts of both time and money.

Betting in this way seems an odd activity if you've already got an infinite amount of money. Maybe it's a side-effect of having an infinite amount of time.

 

[amadeus]

Lotteries are a tax on stupidity.

Then what is voting?

 

[Mise]

The martingdale system "works" (assuming infinite capital and no limits) even if the expectation value is less than 1. There's no reason why you should merely double your bet each time. Imagine if you quadroupled your bet each time in a game that has a 50/50 payout but a 33% chance of winning:

Bet 1: $1 = loss
Bet 2: $4 = loss
Bet 3: $16 = loss
Bet 4: $64 = loss
Bet 5: $256 = win

Total loss = 1 + 4 + 16 + 64 = $85
Total win = $256

It's easier to see when you use unusual numbers like quadroupling your bet and 33% chance of win vs 50/50 payout, but actually, it works with any bet that covers your losses, and any ratio of winnings to prob of winning (i.e. expectation value). Even if the expectation value was 0.0001, the point of the martingdale strategy is that, eventually, you're going to win, and your winnings will more than compensate for your losses. At that point of course, you should stop playing...

 

[snarzberry]

The martingdale system "works" (assuming infinite capital and no limits) even if the expectation value is less than 1.

Firstly, it's possible that every roulette ball spun from this moment will come down red.

Assuming infinite capital speaks to the observation that when you double a value over and over it doesn't take many generations before you have some very large numbers.

For instance, if you were betting on red/black and started with $10 bets on red then you'd need about $20 trillion to survive a run of 40 blacks!

 

[warpus]

But the odds of 25 straight losses are slimmer than 24 straight or 2 straight. Still, if you play long enough, you will hit that streak that hits the betting limits or breaks you.

This is something we tackled in one of the many stat classes I had to take in University as part of my CS degree. If each event is independent then none of that matters, even if it seems like it should. Each individual event will have the same odds, no matter what happened during previous flips, even if the odds of you getting head 25 times in a row are incredibly low. You do not consider what happened in the past at all. If you've got 24 heads in a row, that 25th flip is going to have 50/50 odds. Tails won't be any more likely than heads.

 

[rugbyLEAGUEfan]

This is something we tackled in one of the many stat classes I had to take in University as part of my CS degree. If each event is independent then none of that matters, even if it seems like it should. Each individual event will have the same odds, no matter what happened during previous flips, even if the odds of you getting head 25 times in a row are incredibly low. You do not consider what happened in the past at all. If you've got 24 heads in a row, that 25th flip is going to have 50/50 odds. Tails won't be any more likely than heads.

JR's post isn't at odds (excuse the pun) with this though. He's saying 25 straight is less likely than 24 or 2 straight at the time the first toss is made.

 

[Defiant47]

This is something we tackled in one of the many stat classes I had to take in University as part of my CS degree. If each event is independent then none of that matters, even if it seems like it should. Each individual event will have the same odds, no matter what happened during previous flips, even if the odds of you getting head 25 times in a row are incredibly low. You do not consider what happened in the past at all. If you've got 24 heads in a row, that 25th flip is going to have 50/50 odds. Tails won't be any more likely than heads.

I have a strategy that allows me to recoup previous losses by doubling my bet. I only have money enough to keep doubling the bet 25 times. My strategy is that if I get a win at least one time in the 24 coin flips, then I'm back at 0 and can try again. But if I lose all 25 times, then my strategy failed. The chances of my strategy failing is based on the likelihood of getting 25 failures in a row, regardless of whether they're independent from each other.

 

[warpus]

Doesn't matter, the strategy will work whether you go with heads or tails, each time you play. All that needs to remain constant is the probability of you winning, which is 1/2 no matter whether you go with heads or tails.

JR's post isn't at odds (excuse the pun) with this though. He's saying 25 straight is less likely than 24 or 2 straight at the time the first toss is made.

Yeah, that's true! It doesn't affect the strategy we're talking about though, unless I've missed too many posts and the conversation has turned elsewhere

 

[Mise]

It does affect it, because it's highly unlikely that, after another 25 tosses, they will all be bad, irrespective of whether you start at toss 1 or 25 or 634553.

 

[Terxpahseyton]

Then what is voting?

necessary?

 

[warpus]

It does affect it, because it's highly unlikely that, after another 25 tosses, they will all be bad, irrespective of whether you start at toss 1 or 25 or 634553.

I'm not following, so I'm not sure if we're disagreeing or not.

Say you flip 50 coins and you get 50 heads in a row. The probability of the next flip being heads is 1/2 or 50%. The probability of the next flip does not take any previous flips into consideration.

 

[Traitorfish]

necessary?

Necessary in what sense?

 

[Terxpahseyton]

To enable the bourgeois to exploit the working class, of course.

Kidding, actually in the sense that some kind of voting is necessary to organize huge masses of people in a way other than to have them simply be told what to do by some kind of authority. So voting, while absolutely useless to the individual other than in an idealistic sense, still is necessary to have the people rather than some dude decide (weather we like what they vote for or under what conditions they vote in a given system is a mythinks entirely different matter).
In contrast to that, playing the lottery serves no bigger collective gain.

 

[Traitorfish]

Who are "the people"?

 

[Terxpahseyton]

People who happen to live in a place and who out of some historic circumstances happened to organize themselves as one entity of some kind? Or in other words - if we assume the people as the sovereign, the people decide who the people are

 

[Traitorfish]

People who happen to live in a place and who out of some historic circumstances happened to organize themselves as one entity of some kind?

You mean the House of Lords?

Or in other words - if we assume the people as the sovereign, the people decide who the people are

That seems a bit circular.

 

[Formaldehyde]

I don't think you understand my point.

You are betting on 25 bets, each with 50:50 odds. If you lose you double your next bet and so on. If at anytime during this streak you win, you are reset to 0 and you awak away. It is extremely unlikely that you are going to lose all 25 bets and you only need 1 bet to pay out to walk away even on the night.

At that stage, you are $1 up if that was the size of your original bet. But if you go bankrupt before winning, you have lost everything.

The other obvious problem is that the odds are never even in a casino. They aren't even close to being even in any lottery.

 

[Terxpahseyton]

You mean the House of Lords?

Well no, but I suppose I could.

That seems a bit circular.

Yea well without a god who then gave you the queen or some kind of superior "anchor" of the sort democracy is a circular process, isn't it?

 

[Traitorfish]

Yea well without a god who then gave you the queen or some kind of superior "anchor" of the sort democracy is a circular process, isn't it?

I don't see why it has to be. Whether or not "the people" exist, it's self-evident that people do; why get metaphysical about it?