Betvoyager No Zero Roulette: worse in short term, better in long term

[Bliss]

Work for betvoyager?

I thought someone might accuse me of that. The answer is no, but even if I did, would it matter? Either the math is correct or it isn't. I haven't posted on the forums for a long time but I used to be a member at VLSroulette and also GG before the forum got screwed up. Some of the guys from GG might remember me - Ken?

Does Snowman post here? he's a math guy and I'm sure will back me up. But really there's no need, you can check the facts for yourself if you have any programming skills. You could even do it on a spreadsheet. If there are any programmers here they can check my code. It's written in simple Microsoft BASIC syntax.

[Bliss]

Here is the proof again done with simple math. I'll use Steve's example of a player who starts with a $1000 bank and stakes $10 for 100 spins.

Assuming a HA of 2.7%, the player can can expect a return of 1 - 0.027 = 0.973.

The average loss over 100 bets is going to be $1000 * 0.973^100 ~ $65.

So he has $1000 - $65 = $935 left from his original bankroll.

Now let's see what he loses with no house edge, but 10% deducted from the winnings.

Since there is no HA, on average he will lose nothing, so will end up with the same amount he started with - $1000. But, again on average, half of these bets will have been losses and half wins, which means he won $500.

10% of $500 = $50, so his net "gain" will be $1000 - $50 = $950.

This is $15 more than if the HA had been present, even after paying the 10% on winnings. And don't forget this is only over 100 bets! Over 1000's of bets the difference will be much greater.

[Priyanka]

Some of the guys from GG might remember me - Ken?

We know you dont work for BV, Bliss. Internet never forgets anything, so you can Chillax!

Now with respect to the fairness bit, Steve is correct. The underlying assumption you have taken around 50% wins is something that I would question. Let me try explaining this.

To understand this we need to get to the basic definition of house edge. There are 37 possibilities or slots in single zero wheel we all know that. Casino pays out 36 as winnings. So the house edge is 36/37 or  1-0.972 or 2.7%.

Now consider the no zero roulette situation. There are 37 possibilities or slots. Casino pays out 36 as winnings but retains 10% of that winnings. So they pay 32.4. So the house edge is 32.4/37 or 1-0.875 or 12.43%.

So you can see what is better. Hope it all makes sense

[RouletteGhost]

Math VS Math

May the best math win

[Priyanka]

Also bliss I would make one correction to the following formula in the code.

It is currently reading.

print "profit = "; 0.9 * (wins - losses)

It should read.

print "profit = "; ( 0.9 * wins )- losses

Hope my reply is of some use.

[psimoes]

We know you dont work for BV, Bliss. Internet never forgets anything, so you can Chillax!

Now with respect to the fairness bit, Steve is correct. The underlying assumption you have taken around 50% wins is something that I would question. Let me try explaining this.

To understand this we need to get to the basic definition of house edge. There are 37 possibilities or slots in single zero wheel we all know that. Casino pays out 36 as winnings. So the house edge is 36/37 or  1-0.972 or 2.7%.

Now consider the no zero roulette situation. There are 37 possibilities or slots. Casino pays out 36 as winnings but retains 10% of that winnings. So they pay 32.4. So the house edge is 32.4/37 or 1-0.875 or 12.43%.

So you can see what is better. Hope it all makes sense

There are 36 slots in the NZ roulette.

But I still do not understand Bliss´calculations.

[Priyanka]

Now consider the no zero roulette situation. There are 37 possibilities or slots. Casino pays out 36 as winnings but retains 10% of that winnings. So they pay 32.4. So the house edge is 32.4/37 or 1-0.875 or 12.43

I should not blame this on my fat fingers. This Should read as follows.

Now consider the no zero roulette situation. There are 36 possibilities or slots. Casino pays out 36 as winnings but retains 10% of that winnings. So they pay 32.4. So the house edge is 32.4/36 or 1-0.9 or 10%.

[psimoes]

They think that if the house edge is say 2.7%, it means they can sit down at the roulette table with $100 and after a few hours leave with a loss of only about $3. What they don't understand is that the house edge doesn't apply to their starting bankroll, but the total amount they wager.

Does "total amount players wage" mean their initial bankroll plus the winnings, i.e. a player arrives with 100, bets 50 on an EC and wins,  uses the now 150 to bet some more for the entire session? Those winnings by means of risk taken were already subject to the house edge, right?. Applying HE again over money taken from the casino sounds like double taxing.

[Priyanka]

Applying HE again over money taken from the casino sounds like double taxing.

If you stop betting post the money is taken then your statement is true, but if you are going to wager your risk exposure increases.

Psimoes, that statement you quoted from Bliss is completely correct. The house edge should be calculated on the amount wagered and not the bankroll. This is the reason why people get wiped off their bankroll as it compounds based on the amount wagered.

But I still do not understand Bliss´calculations.

The only area what Bliss did not get quite right is the following bit.



"The average loss over 100 bets is going to be $1000 * 0.973^100 ~ $65".


This should have read ~$27 instead of 65.

Every 36 out of 37 times you are going to win. So considering you are playing ECs. You are staking $10 for 100 spins. Out of 100 spins, lets say every 37th spin will be zero. So spin 37 and 74 are 0. That takes out 2 spins from 100. REmaining 98 spins we win 49 or lose 49. So the calculation is

49 spins won = $10*49 winnings = $490
49 spins lost = $10*49 lost = $490
2 spins lost on 0 = $10*2 = $20

Total bankroll at the end is $1000+$490-$490-$20 = $980.

Take Non-zero. Same thing no zeroes. 50% you win and 50% you lose.
50 spins won = 0.9*$10*50 winnings = $450
49 spins lost = $10*50 lost = $500

Total bankroll at the end is $1000+$450-$500 = $950.

So $950 in non-zero with 10% commission as opposed to $980 in a single zero roulette.

[psimoes]

I guess an easy to understand situation would be something like the following: a player has a BR of $37 and bets $1 on a straightup for 37 spins. Wins 1 time, loses 36 times, goes home with $36. The player does this for three sessions. Well, the results are the same, $111 minus 2,7% is $108; the same as $37 minus 2,7% equals $36; times three equals $108.
You just pick the bankroll and subtract 2,7%.

With the NZ Roulette there is no HE so in the longterm the winnings equal the losses. Which means the 10% aftertax over the winnings is equal to half of that subtracted to the bankroll i.e. 5%. Almost doubles 2,7%

[Drazen]

So $950 in non-zero with 10% commission as opposed to $980 in a single zero roulette.

Having explained all the above, may I ask why do you prefer BV non zero wheel in all your showed plays then?

Drazen

[Priyanka]

Having explained all the above, may I ask why do you prefer BV non zero wheel in all your showed plays then?

Drazen

Not the subject of the thread. But to answer that question when you can win every session does it really matter to you whether the house edge is 2.7% or 10%.  As Mr. J would put it, it's my tip

[RouletteGhost]

Not the subject of the thread. But to answer that question when you can win every session does it really matter to you whether the house edge is 2.7% or 10%.  As Mr. J would put it, it's my tip

E. X. A. C. T. L. Y.

[psimoes]

Think the HE on the NZ is actually 5% as in the longterm you´ll win only half of your bets. Are you saying you can actually win all the time? Like, just because there is no zero to mess things up? Just curious.

[Bliss]

Hi Priyanka,

You're right I did make a mistake in my calculation, but actually we are all wrong. 

The deduction on winnings is not made in the way you have calculated it. In fact things are actually better than I thought, because the 10% applies only to net winnings. See the payment rules on Betvoyager-

link:s://:.betvoyager.com/about/payrules/

If a player plays on equal odds casino games, a 10% house fee will be taken from the player's net winnings at the end of the game session or upon a withdrawal of net winnings from the table through the cashier. The duration of the game session cannot exceed 24 hours. The house fee and restrictions are only in regards to equal odds games.

There are two examples given.

My code was basically correct, but I needed to add the following:

Code Select Expand

'payout 90% of net winnings if in profit.
if wins > losses then
  print "profit = "; 0.9 * (wins - losses)
else
  print "profit = "; wins - losses
end if

So it is certainly not accurate to say that the odds are twice as bad as the normal house edge game. They are considerably better. Of course, if you always win every session then you may end up paying more "tax" because your net winnings are always positive, but against that you have to take into account how much the presence of the house edge damages the profit you would have made had it not been present.

Returning to the calculation, suppose you make 1000 bets of $1 on red, your return will be $1000 * 0.973 = 973 (not sure how I messed that up previously, and you were right to point out the error).

So you have lost $27 to the house edge.

For the no zero game, on average, since the odds are completely fair, you will make no net profit, and no net loss. You break even. So in the "long run" you pay no 10% tax at all.

Of course, in a given session profits will fluctuate, you win some and lose some, so of course you end up paying the tax on those occasions when you finish a session with a net profit.

On the other hand, if playing the game with standard house edge, the more you bet, the more you will lose.

So I know which game I'd rather play, and the simulation proves it.