### Author Topic: Are Some Bets Worse Than Others?  (Read 7687 times)

0 Members and 1 Guest are viewing this topic.

#### Scarface

##### Are Some Bets Worse Than Others?
« on: November 16, 2017, 11:26:38 PM »
Are some bets worse than others?  The house edge is the same for all bets, so in theory no bets can be worse on a fair wheel.  It doesn't matter what system you play, all will have the same fate toward the house edge.

But is this true.  Let's look at even bets.  There are probably hundreds of different even bets that can be played...not just red/black, high/low, etc.  You can play the last 18 numbers hit, or last 6 streets on the table, last 6 streets on the wheel.  There are even progressions of adding numbers that qualify as even bets.

Has anyone tested anything like this before?  Are there bets that are "better or worse" when it comes to variance?

The following users thanked this post: kav

#### heatmap

##### Re: Are Some Bets Worse Than Others?
« Reply #1 on: November 16, 2017, 11:59:07 PM »
The worst one I can think of... you arrive at the casino and then go up to the table with all of your money and put it on one bet. Win or lose its stupid in my opinion.

The following users thanked this post: kav

#### kav

##### Re: Are Some Bets Worse Than Others?
« Reply #2 on: November 17, 2017, 12:01:16 AM »
heatmap,
Mathematically it is not stupid.
Practically though it's a whole different discussion ;-)

Scarface,
I asked a similar question here

The following users thanked this post: Jesper, heatmap

#### McCoy

##### Re: Are Some Bets Worse Than Others?
« Reply #3 on: November 17, 2017, 09:38:28 AM »
Are there bets that are "better or worse" when it comes to variance?

The variance changes according to how many numbers you bet.

Var = n * p * q  (p = probability of win, q = probability of loss, n = number of trials).

You can find the turning point by calculus. Because q = 1 - p the variance is Var = n * p * (1 - p) = n * p  - n * p^2
The variance curve for a binomial distribution is a quadratic with the minimum at the bottom.

To find the min value of p set the derivative of p to zero. n is a constant so is taken outside the differential operator-

n * dVar/dp = n * d (p - p^2)/dp = 1-2*p

set equal to 0 and solve for p

1-2*p = 0

p = 1/2 , so the bet which has the minimum variance is the even chance.
« Last Edit: November 17, 2017, 10:07:58 AM by McCoy »

The following users thanked this post: kav, Reyth

#### dobbelsteen

##### Re: Are Some Bets Worse Than Others?
« Reply #4 on: November 17, 2017, 10:36:49 AM »
HE is a myth .HE calculates  the loss of 2,7% only in the long run. Short run events have no HE. The result is unpredictable. There is also a Expected Value. Most players do not know the difference between EV and HE.

The following users thanked this post: Reyth

#### Reyth

##### Re: Are Some Bets Worse Than Others?
« Reply #5 on: November 17, 2017, 01:44:44 PM »
Great post Mccoy!

Now, how do we add profit generated into the equation because that is part of the picture too?
« Last Edit: November 17, 2017, 03:29:24 PM by Reyth »

#### Scarface

##### Re: Are Some Bets Worse Than Others?
« Reply #6 on: November 18, 2017, 03:20:40 PM »
Let's look at 2 different strategies.  Player 1 plays 4 numbers based on the last 4 numbers that have repeated.  Player 2 will always play 4 numbers that haven't hit the longest.  Both players 4 number bet will constantly change throughout the game.  But player 1 will always be playing the most recent hot numbers, while player 2 will constantly be chasing the 4 coldest numbers.  Wouldn't player 1 have a more superior bet selection?

4 cold numbers can go 100+ spins without hitting once.  Can playing the 4 most recent repeats go that many spins with a no hit as often?

Player 1 will also see more streaks of wins than player 2.
« Last Edit: November 18, 2017, 03:24:49 PM by Scarface »

#### Reyth

##### Re: Are Some Bets Worse Than Others?
« Reply #7 on: November 18, 2017, 03:28:47 PM »
How many times would 4 numbers have to wink out to ruin Player 1?
« Last Edit: November 18, 2017, 03:32:02 PM by Reyth »

#### Scarface

##### Re: Are Some Bets Worse Than Others?
« Reply #8 on: November 18, 2017, 03:33:34 PM »
How many times would 4 numbers have to wink out to ruin you?

Depends on bankroll and progression.  Just more curious on which strategy is more stable.  Hard to believe that both strategies are equally bad.  I would think the repeat strategy would have less variance.  Doesn't necessarily have to have an edge...just less variance.

I'm more interested in the idea of is there really better bet selections than others.  Can some selections be shown to be worse in the long run.  If not, then there seems to be no point in having a strategy at all
« Last Edit: November 18, 2017, 03:36:28 PM by Scarface »

#### MrPerfect.

##### Re: Are Some Bets Worse Than Others?
« Reply #9 on: November 18, 2017, 04:21:58 PM »
It's brings up a simple qwestion : " how in this world do select your bets? "
If it's not obvious,  then probably you do something wrong....

#### Reyth

##### Re: Are Some Bets Worse Than Others?
« Reply #10 on: November 18, 2017, 04:35:36 PM »
I would think the repeat strategy would have less variance.  Doesn't necessarily have to have an edge...just less variance.

Rinad points out that if we can identify the hottest selections, we know that we aren't betting the coldest.

Quote
I'm more interested in the idea of is there really better bet selections than others.  Can some selections be shown to be worse in the long run.

I truly believe, risking 100's of dollars, that roulette functions in this way; hot selections persist for a longer term than the short-term fluctuations that every selection goes through.

The important thing to realize though, is that these hot selections do not always remain the hottest; roulette will EVENTUALLY correct the disparity.  But how does it do it?  With ANOTHER hottest selection!

#### Scarface

##### Re: Are Some Bets Worse Than Others?
« Reply #11 on: November 18, 2017, 11:30:28 PM »

The important thing to realize though, is that these hot selections do not always remain the hottest; roulette will EVENTUALLY correct the disparity.  But how does it do it?  With ANOTHER hottest selection!

Exactly!  Always stay with the most recent hottest.  Hot numbers can change, so you got to change with them.  Sometimes, if playing 4 or 5 numbers, at least 1 of these will stay hot for 200 spins or more.
« Last Edit: November 18, 2017, 11:58:01 PM by Reyth »

#### Reyth

##### Re: Are Some Bets Worse Than Others?
« Reply #12 on: November 18, 2017, 11:58:27 PM »
I wish it were that easy for me!

I know there are many people that do this easily, but I am unable.

I have found that every selection will become hot at one point or another as part of its natural cycle and I have not found a way to look at these cycles with a short-term view and be able to tell the difference between a short-term cycle and one that will extend for 100's or even 1000's of spins; in other words, I don't know how to track the most recent spins reliably.

Furthermore, I have not found a way (yet?) to determine when a hot selection is going to gap (it WILL!).

I have a certain "weakness" in roulette where I don't do well with existing losses on the books on a regular basis and I am not good at thinking-acting creatively.
« Last Edit: November 19, 2017, 05:09:45 PM by Reyth »

#### McCoy

##### Re: Are Some Bets Worse Than Others?
« Reply #13 on: November 19, 2017, 08:54:00 AM »
I have found that variance can be reduced, whatever the bet, by using techniques from machine learning.

https://stats.stackexchange.com/questions/18891/bagging-boosting-and-stacking-in-machine-learning [nofollow]

Combining different models (bet selections) in these ways results in less 'noise' and better predictions. I have already outlined a very simple way to do this in another thread, but it can be improved considerably. It does involve quite a lot of work though and can't be used in bricks and mortar casinos for obvious reasons.

The following users thanked this post: Reyth

#### dobbelsteen

##### Re: Are Some Bets Worse Than Others?
« Reply #14 on: November 19, 2017, 09:30:29 AM »
From my point of view there is no difference between 4 random numbers or particular 4 numbers. A 4 number bet has a very large DTOP. That means that every result is possible. Sometimes is player 1 in profit and the other time player 2.
The only way to proof it ,is long term simulation. The short term result can be influenced by the kind of progression positive or negative.

The examples are not strategies but systems. Feelings are danger to draw conclusions. Research gives the answer!!!!

The following users thanked this post: Reyth