Golden Touch Craps

The poster I wrote about last month in the article http://www.goldentouchcraps.com/skinny0015.shtml had a follow up question. Here is that question along with my response.

Hi Skinny,
 
Thanks for responding.  The response was very well written. 
 
I figured the HA would still take over on the dp/dc and you would still lose 1.36% regardless of the five count.  I am not a frequent dark side bettor but if the table is cold and I want to bet, I always limit it to one bet. If I lose it, then I don't make it again on that shooter.
 
Here is my thought.  If you eliminate 57% of the losing rolls with the five count, wouldn't the count just start over again and the next 57% of the rolls from a 6-10 count be losers?  I think people get the five count confused and think they will win money on the five count but that is not the truth.  The only truth is that it will save you money because you don't bet on everyone.  I would like to know, what % of rolls after the five count would make it to 10, 15, 20, or 30 rolls?  I think I had read before that only 30 out of 1000 rolls ever make it to a count of 30.  Is that true?  I think I figured it more around 26-27 rolls.
 
Next thought, using the 5 count on a controlled shooter, can we figure what percentage of the rolls would be losing rolls based on their SRR?  Since Dr. Don Catlin used a standard 1 in 6 rolls has a 7 probability.  Let’s up it to 1 in 6.5 or 1 in 7 and so forth.  Here is what I am thinking.  What is the SRR needed to change the five count figure of 57% losing rolls into anything under 50%?  When we have established a SRR high enough to break this, wouldn't we in fact want to bet on all of our rolls and no longer use the five count?
 
Thanks again,
 
GF

 

 

 

Your welcome GF,

 

I am pleased you liked it.

 

"I figured the HA would still take over on the dp/dc and you would still lose 1.36% regardless of the five count.  I am not a frequent dark side bettor but if the table is cold and I want to bet, I always limit it to one bet. If I lose it, then I don't make it again on that shooter."

 

Making one bet on random rollers is always a good idea.  It does not matter if it is on the do or the don't.  Whatever makes you feel comfortable is the best way to go.  BTW, I suspect you realize there is no such thing as a cold table.  The random nature of dice cause streaks to occur, which we as humans define as hot, cold and choppy.  But the streak can only be stated after the fact, not before it occurs.  It does not matter what happened in the past.  If a table has been cold, there is no mathematical reason for it to continue or change.  That is why I say, do whatever makes you feel comfortable.  There is no mathematical reason to bet either way.

 

"Here is my thought.  If you eliminate 57% of the losing rolls with the five count, wouldn't the count just start over again "Yes" and the next 57% of the rolls from a 6-10 count be losers?"  "NO"

 

That is one of the points I was trying to make.  Just because 57% of the rolls for random rollers do not make it past the five count, it does not mean you can win a greater percentage of your bets on the don't.  The shooter can throw a 7 or 11 on the come out before throwing his first point number which is when the five count would begin.  You will lose your don't bet on a come out 7 or 11 even though the shooter does not make it past the five count.  That will happen often enough so that you will still lose your don't bet even on those 57% of the rolls that don't make it past the five count.

 

57% of the rolls of a random roller not making it past the five count just means 57% of the rolls by a random roller will be a seven out before completing the five count.  It does not mean you are going to win a don't bet 57% of the time.  That is because for that 57% of rolls, the first roll will be a don't loser 8/36, don't winner 3/36 and don't tie 1/36 of the time.  That means 22.2% of the time you are going to lose your don't bet on the first throw, win it 8.3% of the time and tie 2.8% of the time.  Your don't bet gets resolved 30.6% of the time on the first throw and the don't loses 8/11 of those.  Of course you will have the advantage on the other 24/36 rolls when a point number is established.  But it will not be enough of an advantage to bring your winning percentage on the don't up above 50%.  It will only bring it up exactly to the HA.  For random rollers you can expect to lose $27 for every $1980 you bet on the don't or 1.36% of all the money you ever bet on the don't.  For random rollers you can expect to lose $28 for every $1980 you bet on the do or 1.41% of all the money you ever bet on the do.

 

The above is the crux of my argument.  If it is not clear or you still question it please let me know.  It is important you understand and believe it otherwise you will be missing the point.

 

That is also why I say you can bet on the do or the don't for random rollers.  Go whichever way makes you feel more comfortable.  Your peace of mind and comfort at the table is more important than the difference of $1 out of every $1980 you bet on random rollers to make the don't a better bet than the do and vice versa.

 

"I think people get the five count confused and think they will win money on the five count but that is not the truth.  The only truth is that it will save you money because you don't bet on everyone."

Agreed.

 

"I would like to know, what % of rolls after the five count would make it to 10, 15, 20, or 30 rolls?  I think I had read before that only 30 out of 1000 rolls ever make it to a count of 30.  Is that true?  I think I figured it more around 26-27 rolls."

Here are two charts by Dan Pronovost that give you some information about your question.

 

http://www.goldentouchcraps.com/CrappyMath/math0004.shtml

http://www.deepnettech.com/smartcraps009.shtml

 

"Next thought, using the 5 count on a controlled shooter, can we figure what percentage of the rolls would be losing rolls based on their SRR?  Since Dr. Don Catlin used a standard 1 in 6 rolls has a 7 probability.  Let’s up it to 1 in 6.5 or 1 in 7 and so forth.  Here is what I am thinking.  What is the SRR needed to change the five count figure of 57% losing rolls into anything under 50%?  When we have established a SRR high enough to break this, wouldn't we in fact want to bet on all of our rolls and no longer use the five count?"

 

This has been discussed this from time to time.  There is no doubt if a player has an edge that player should bet on all his own rolls.  As for other players who have an edge the discussion gets murky.  In theory if a player has an edge you should bet on that player all the time without using the five count.  But you would have to bet on that player all the time in order to be able to take advantage of his edge.  Since you would not be able to bet on anyone other than yourself every time the person shoots, I think it only makes sense to do that on yourself.  I think even for other players who have an edge you would be wise to use the five count on that shooter until you have seen evidence of that player being "on".  You may know the players you play with well enough to be able to tell when they are consistently throwing their A shot.  Bet early on a player you know if you think that player is shooting well.  But that is just my opinion.  As I said in theory, if a player has an edge you should always bet on that player without using the five count.  Doing so would increase your variance because you are wagering more money.  Your 401K and mental health need to be able to withstand the downswings which will happen as a result of that variance.

 

Skinny