Craps Math is not important - lets have at it

Quote: AlanMendelson

Make no mistake... I always bet full odds on my passline bets... but I can afford the losses and I am only playing 3, 4, 5 X odds.

Now, if I were playing over at the Riviera and wanted to bet their max of 1,000X odds I would not automatically bet full odds on every point.

There I would have to judge my "chance of winning" and I would be more likely to bet 1,000X odds on 6 and 8 than on 4 and 10.

When I gamble sometimes I have to put chance of losing ahead of payoffs on winning.

And as I said before, am I the only one who thinks this way???

No, and it's a perfectly sensible way to think, if you have a tight bankroll. But if your bankroll is that tight, you should probably be looking for a lower limit, or not playing at all until you've saved up some more. If you have a solid bankroll, larger wins will over time make up for more frequent losses, and vice versa.

Quote: AlanMendelson

And as I said before, am I the only one who thinks this way???

No, I see many Craps players play this way.
Some do not even take the odds on the 4/10 thinking now since the point is a 4/10 there is just not enough of a chance to win, too great of a chance to lose, to justify the bet and the 5&9 gets less odds than the 6&8 for the same reason.
Comparing wrongfully one bet to another solely based on a winning probability.

Just making a $5/$5odds bet on any number a player can lose then same $$$ at different rates,
but the overall amount on the odds, on average, is 0, with just one more win needed in most sets of 9,10 and 11 total bets made to show a profit, and a larger one than taking no odds.

The bet does not pay even money, it does not pay 9 to 5 or even 2 to 1 and 5% vig on a win. (4/10)

It is the same as saying I want "no action" on my don't/don't come 6 and 8. They lose too many times.
That would be correct if you only making just a few bets, say 5, over one's lifetime of play.

One is not looking at the complete picture, just a picture for one session.
If this is simulated over 10 million times, as 10 million different players play this way,
one will see it just does not matter even in a short session since there are times you bust out of a session because of the winning probs but that is exactly compensated for by the times you do win, and win big on the odds.

If one calculates all the possible outcomes, and not just a few from what you have remembered seeing, one can see the complete range of where one could end up at.
Too many non math gals and guys just make assumptions based on the 2^e^729989456454.12 feelings and thoughts that pass through them instead of doing the math to see that it makes no difference.


The valid point you have brought up is, when playing with a fixed bankroll, or Sally saying one is playing scared, is when one weighs all the possible real world results, the odds either 1X or up to 1000X, for any number is a fair bet. So you bet less and the point 4 hits.
Do you now still feel better because your ideas just cost you money?

The winning probabilities for 9 trials are posted above in Sally's post.
Those probabilities are not way off the charts or even stiff odds as you have stated.
33% vs 45.45% are not at all stiff odds because when the 4/10 do win, on the right side, they pay way more.
Do you not want to win way more, even after a bunch of bets? Yes you do and if one just does some simple math on this, one will see it makes no difference even in short term play to bet less taking odds on the 4/10 vs taking more on the 6&8.


IF one only thinks that you win more by betting more on the odds for the numbers that have a higher winning probability,
just like the don't bettors that only lay the odds on the 4/10 and stay away from the odds on the inside point numbers,
that player and those that think that way just do not understand the math, for more than one bet at a time.
They just use their thoughts on this, and those easily will let one down.


In a short session, that is really part of just one big lifetime session,
and betting less on the odds for outside numbers makes no (math) sense IMO.
I personally saw last week a 2 hour session where not one point of a 6 or 8 hit?
Was it a bad bet?

People kept betting big on the 6&8, I did also, we were all really pissed off at the 6&8,
when it was the point saying they are so due to win! They can't keep losing all the time.

Well, they did, and took all those players down with them.
I survived because I kept my bets the same and stopped chasing the losses.

Quote: AlanMendelson

There I would have to judge my "chance of winning" and I would be more likely to bet 1,000X odds on 6 and 8 than on 4 and 10.

When I gamble sometimes I have to put chance of losing ahead of payoffs on winning.

And as I said before, am I the only one who thinks this way???

No. There are many, wrongfully-math based, thinking that way.
You need to be more consistent IMO in how you play Craps.
You play by feeling only?

Are you over 90% successful in your play by using your feelings?
That is so cool if it is true.

It seems to me you do as does many craps players playing by their feelings.
If that makes you have more fun playing or makes you think you can win more or lose less or makes you happy,
go for it,
since you can not show with math or any simulations the expected value of your type of play based from feelings and hunches about future random events.

That brings up another subject, most Craps players do not believe that each dice roll and each outcome is random.


You are only judging your chances of winning instead of calculating.
Two different methods and two different returns.

Here is some winning math by EV only the odds on the
4/10 .
They win, on average 1/3 and win $2 for every $1: EV = 1/3*$2 = $2/3 or .67 or 67 cents per bet made

5/9,
They win, on average 2/5 and win $1.5 for every $1: EV = 2/5*$1.5 = $3/5 or .60 or 60 cents per bet made

6/8,
They win, on average 5/11 and win $1.2 for every $1: EV = 5/11*$1.2 = $6/11 or .5454 or 55 cents per bet made

So the math says to make more bets on the 4/10 to win more money.

Point out the errors in my math, not just my judgments.
I watch many place bettors after a few wins reduce the money on their place bets and then the numbers keep hitting.
I say that is always wrong.
They would have won more money.
But they justify their actions now by saying the can and do lose more money if they leave their bets too high.

This tells me they are playing scared.
Scared of losing and scared of winning.

" mustangsally "

Hopefully named after 1965 Mark Rice recording and not the horrid 2006 Movie ? just curious !

I have to admit it. I'm a math guy. In fact, I have an advanced degree in math. Now, all math guys know that there is a built in house advantage so that they can pay for the beautiful cocktail waitresses, the drinks, the horrible carpets and all the other stuff that casinos have as well as providing a dividend to the stockholders. So, mathematically speaking, in the long run, you will *NOT* beat the house. Math guys play craps and other games of chance anyway, for a wide variety of reasons. It is possible to beat the house in the short run though, and craps with its low house advantage on the pass line and the 6/8 place bets allows that to happen; but not consistently.

This thread is quite relevant to me because my lady friend claims that I use too much math and that I should just use intuition. Well maybe I'd do better - but I seriously doubt it. In fact, she said that "scared bettors never win." I've heard that a few times. Yes, there's a good reason NOT to bet on 4 or 10, contrary to the bad advice on this thread. No amount of intuition, or I should, pride, is going to change that.

removed
silly

I have two questions, if there is anyone left who has not blocked me. LOL

We talk a lot about a session, usually referring to the time frame from our buy-in to the cash out, be it 20 mins or 8 hours. And then we talk about how anything can happen for that session, due to variance.
That is then followed up with the statement, "but make 10,000,000 rolls and the math will work out as predicted". This statement I understand.
And of course we also recognize that your lifetime of gambling should also be considered to be just one long session, with breaks along the way.


What is the minimum number of rolls required for there to be a reasonable chance of the expected distribution? (Sorry, I am really struggling on how to word this.)
Is it 1,000 rolls, 10,000 rolls, more ?

Example. If I roll the dice 36 times, I do not expect to see one -2, Two 3s, Three 4s, etc.
36 rolls is obviously a very short time frame where most anything could occur.

As the number of rolls gets higher, does the chance of variance decrease?

If I understand this at all, I would think it does. In other words, "if I roll the dice 36 times, my number distribution can be most anything, which would mean a very high variance"? Did I say that right?

If that statement is correct, then is it also correct to say, "if I roll the dice 10,000,000 times, I would expect my number distribution to fall very close to expected results, thus a there would be a lower variance"?

Damn MustangSally, that is a lot of good information. THANK YOU!

For the record, I have gone out to wikipedia and other internet sources, trying to understand SD. I know more than I did, but I am far from understanding it. But I think understanding it is crucial to my understanding the potential impact to me for my style of craps play.

I also need to clear the air. At NO time did I ever mean to imply that Place/Buy betting is a better bet than Pass/Come with Odds. the math is indisputable.
I may have used a poor choice of words that left that impression, but that was not my intention.

I think you mentioned that I am 68% likely to fall in the distribution from -1 SD to +1 SD. Is that correct?

So in the graph above, after 5000 rolls,
if I am CB with 345x odds, I expect to be -424.2, but I have a 68% chance of ending anywhere from -2,509 to +1,661.
if I am Placing the 6 then, I expect to be -1,717, but I have a 68% chance of ending anywhere from -3,446 to +12

Am I reading your information correctly?

The bell curves do not look like they would line up on top of each other, and the spread of -1 SD to +1 SD bears witness to that. The CB appears to be wider, and have a slightly flatter top. Is that because the lower HE of the CB leads to a wider number of possible results?

I know the COR has a 2-1 advantage of an instant win, but that is not my personal experience. I lose so many COR bets to craps it is not funny. I bet I am running closer to 1:1. Perhaps the dice hate me? Or my mind is just programmed to remember the losses and not the wins.

Quote: mustangsally

In my example, You have about a 36.7275811% chance of showing a better outcome than just come bets.

This is some of the information I am looking for. What I also was looking for is what percentage chance do I have of ending equal to the CB? Since both areas of the curves that overlapped was 74%, can I simply subtract the 36% of beating the CB to arrive at a 38% chance of being equal to the CB result?

What I was trying to do was assess the most likely outcome (-1SD to +1SD) of the Place bet vs the CB, to see how bad it is, and I have to say it is much worse than I expected. The +12 for 1 SD of the place 6 was an especially ugly number to me.

Quote: RaleighCraps

What I also was looking for is what percentage chance do I have of ending equal to the CB? Since both areas of the curves that overlapped was 74%, can I simply subtract the 36% of beating the CB to arrive at a 38% chance of being equal to the CB result?

What I was trying to do was assess the most likely outcome (-1SD to +1SD) of the Place bet vs the CB, to see how bad it is, and I have to say it is much worse than I expected. The +12 for 1 SD of the place 6 was an especially ugly number to me.

This thread is now reading like a text book with out chapters and an index.

When you look at the bell curve in the photo, the probabilities for the y-axis (going up) is not there.
The values are probably very small.

Draw a horizontal line with your eyes and where each line crosses that HL,
they have the same probability.

Example:
There are 4 points on the line that has the same probability.


That does not answer a question about the chances of being more than or less than any value.
You would then need to work with the EV and SD using normal distribution functions in Excel or another program.

ev, variance and standard deviation is basic prob/stats stuff.
Working with them to answer higher level questions can take higher level understanding.

Quote: RaleighCraps

I know the COR has a 2-1 advantage of an instant win, but that is not my personal experience. I lose so many COR bets to craps it is not funny. I bet I am running closer to 1:1.
Perhaps the dice hate me? Or my mind is just programmed to remember the losses and not the wins.

The 2 to 1 advantage only relates to the ev. You need to variance and the sd (=/- vaule)
simple math
ev = N*P
var = N*P*Q
sd = square root of the variance

N= # of trials
P= prob of success
Q= 1-P

I still say your cor ratio would be closer to 2:1 than 1:1

If you would have tracked your dice rolls, you would have that answer,
instead you have to rely on your selective memory only and the feelings associated with that.
When players lose for any reason, they remember it more than when they win for any reason.

There are theories about this concept.
Try this: Cognitive bias
"A cognitive bias is a pattern of deviation in judgment that occurs in particular situations,
leading to perceptual distortion, inaccurate judgment, illogical interpretation, or what is broadly called irrationality."

Quote: RaleighCraps

Am I reading your information correctly?

:D

Quote: mustangsally

In my example, You have about a 36.7275811% chance of showing a better outcome than just come bets.

Quote: RaleighCraps

This is some of the information I am looking for.
What I also was looking for is what percentage chance do I have of ending equal to the CB?

The point where both curves cross.
In Excel =NORMDIST(-1131,-1717.4,1729.4,0) gets very close.
0.000217795
or 1 in 4591.5

Quote: RaleighCraps

The +12 for 1 SD of the place 6 was an especially ugly number to me.

It can be for most that do the math.
-$3446.8 has the same probability.

That has to make the +$12 look even more better ;)

Sally

Would someone please respond to BrooklynJake's comment here? It doesn't appear to me to be in line with "group thought" and for that reason I'd like to know a little bit more about what he is saying. Thanks.

Quote: BrooklynJake

I have to admit it. I'm a math guy. In fact, I have an advanced degree in math. Now, all math guys know that there is a built in house advantage so that they can pay for the beautiful cocktail waitresses, the drinks, the horrible carpets and all the other stuff that casinos have as well as providing a dividend to the stockholders. So, mathematically speaking, in the long run, you will *NOT* beat the house. Math guys play craps and other games of chance anyway, for a wide variety of reasons. It is possible to beat the house in the short run though, and craps with its low house advantage on the pass line and the 6/8 place bets allows that to happen; but not consistently.

This thread is quite relevant to me because my lady friend claims that I use too much math and that I should just use intuition. Well maybe I'd do better - but I seriously doubt it. In fact, she said that "scared bettors never win." I've heard that a few times. Yes, there's a good reason NOT to bet on 4 or 10, contrary to the bad advice on this thread. No amount of intuition, or I should, pride, is going to change that.

e431312