Golden Touch Craps

A poster on the private pages of the GTC website asked me a good question which I thought would make a perfect topic for an article. Following is the question and my answer.

Hi Skinny,
 
GTC teaches people that the best or safest bet on a random roller is a pass/come bet with full odds or no odds. But as I was looking over the odds, I found that it isn't true.  The ideal bet on a random roller, if we want to bet, is the don't pass/don't come with full odds or no odds.
 
Now I am sure you have done all the research on this and that is why I am asking you.  In the five-count, how is 57% of the losing shooters determined and when does a random roller becoming a winning shooter, since this is technically not possible in the long run based on the facts of the HA?
 
I hope all has been well and your rolls have been successful.
 
GF

________________________________________________________________

 

Hi GF,

 

This is a great question.  I am going to use it for an article in my column.

First, Dr. Don Catlin did the analysis of the five-count by running simulations.

He determined that 57% of the rolls by a random roller would not make it past the five-count.  This is different than saying 57% of the shooters won't make it past the five-count.  I have no idea what percentage of shooters do not make it past the five-count.  But it does not really matter.

 

A random shooter can throw 7's and 11's before the five-count starts and that will lose your dp/dc bet.  So even if 57% of the shooters did not make it past the five-count, that would not mean you would win the dp/dc, bet 57% of the time.  Remember, there are 24 ways out of 36 to throw a point number.  With the other 12 non-point numbers that can be thrown on the come out, the dp/dc bet loses to 2/3 of those 12 rolls!  That is a whopping 2 to 1 disadvantage for the dp/dc bet on 1/3 of the come out rolls for a random shooter.  This should show you how you can still lose your dp/dc bet on a random roller even if he does not make it past the five-count.  Not getting past the five-count is no guarantee for the dp/dc bet.

 

The only way to analyze the dp/dc bet versus the pass/come bet is by looking at the house advantage (HA) for a random roller.  If a random roller were to bet $1 on the pass/come 1980 times, he could expect to win his line bet 976 times and lose it 1004 times.  Thus he would lose $28 out of the $1980 he bet.  28 divided by 1980 = 1.41%.  That is the HA on the pass/come bet.

 

If a random roller were to bet $1 on the dp/dc 1980 times, he could expect to win his bet 949 times, lose it 976 times and tie 55 times.  Thus he would lose $27 out of the $1980 he bet.  27 divided by 1980 = 1.36%.  That is the HA on the dp/dc bet.

 

It does not matter at what point in time you place your bet on the random roller.  You can do it the moment he starts to roll, wait for him to complete the five-count or wait even longer.  But the minute you put your money on the line on a random roller you are facing a HA of 1.41% on the pass/come and 1.36% on the dp/dc.  The HA works on every bet you make on a random roller and you can expect to lose that HA on all the money you wager on random rollers in the long run.  That is why the best thing to do with random rollers is to not bet on them at all.  Then you will not lose any money on them at all.  Next best is to wait for them to complete the five-count because you will not have your money wagered on 57% of the rolls thrown by the random rollers.  That in turn will save you 57% of the HA on the total money you wager on random rollers.

 

The only advantage you have on the dp/dc versus pass/come bets on a random roller is the difference in the HA.  This difference is a small .05% or $1 out of every $1980 you wager on those bets.  If you think that difference makes it worth it to you to be a don't bettor on random rollers that is fine.  But you are not going to break the bank by betting the don't versus the do on random rollers.

 

Most players would rather bet the do than the don't.  Don't bettors are not looked favorably upon by the majority of players at a craps table.  One needs a thick skin to bet the don't.  He can not participate in the camaraderie of wanting to see the shooter make his point or in the celebration after, if the point is made.  He can not rejoice if he wins his bet unless he is willing to accept the ire and potential harassment from the other players at the table.  Since the difference in the HA is so miniscule, GTC recommends the pass/come bet with or without odds on random rollers after the five-count is completed.  Next to no bets at all, I stand by that as the safest bet on a random roller one can make.

 

Skinny