# Measuring One's Success at Craps

Discussion in 'General Craps Discussion' started by KokomoJoe4, May 14, 2021.

1. KokomoJoe4, May 14, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
This can be done any number of ways, the simplest of which is to keep track of each and every session gain or loss. The reason we gamble is to give ourselves a chance to win money. Mathematics of the game suggests that over the very long haul, most players will have lost more than they have won.

There is no question that, when playing craps, the only thing that is important is that the player be able to add the spots on top of each die when the dice come to rest after a toss.

Depending on how the player has wagered, the end result seen will either make him happy or upset.

We all know that by design, the game favors the house, not the player. This is because the house pays ALL winning bets unfairly, except for a few: (1) odds bets, (2) no vig, 2:1 payment on '4' and '10' , and (3) Field bet paying triple on both '2' and '12'. Two of these three "fair" wagers are so rare they may no longer exist anywhere.

The "odds" of winning at craps are based entirely on the unfair casino payments, which includes vigorish.

Simple mathematical formulas are used to determine how "well" a player has done. So-called DI's might say that these same formulas can be used to determine how "good" they are, and of course this is true, provided that the calculation is correctly done. I will come back to some examples using a few of these formulas later.

What needs to be realized is that our bottom line when in the casino depends on every wager that we make and which goes to a decision. If multiple wagers are being made, the calculation for each is different, and depends on which wager was made. Obviously, calculated results made for one box number will be different than those for a different, non-sister box number. Maybe one of these calculations will show a result better than expected. Maybe not.

The point is, the ONLY thing that matters when using math to calculate how a shooter is performing with respect to "expectation for the bet in question" is (a) the number of wagers made on the box number in question (b) the number of times the number shows, and (c) the number of hand ending sevens showing when these wagers were in action.

Since I know that the baron can hardly wait, I will put up a few examples later.

#1
Last edited: May 14, 2021
HornHiYo, amzngguy and yacraps like this.
2. KokomoJoe4, May 14, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
Any shooter can analyze his results for every type of bet he makes. To do this accurately, all he needs to do is keep track of every bet he makes and record results of every throw he makes.

Since we are concerned only with how "good" this particular shooter is, we are only tracking HIS bets during HIS rolls.

As we all know, anything can happen in the short term. The more results that are analyzed, the more we expect these results will look as predicted by the math, but there are never any guarantees ----> the math SHOWS you where you are with respect to expectation.

As an example to show a calculation. Let's say we have a so-called DI who shoots out of the 3V and hopes to toss a bunch of those "on-axis" sixes and eights that he sees during practice sessions. Because he also tosses a bunch of nines, when this shooter gets the dice, he will be going out on the Pass with minimum required bet, establish a point, and then place the 6,8 and 9. He is betting on what his practice results show him he should.

Calculations for the sisters six and eight are:

HE = (5/11 x 7) - (6/11 x 6) / 6 x 100 = -1.52% or (7/6) - (6/5) x 5/11 x 100 = -1.52%

Same calculation for the place nine is:

HE = (4/10 x 7) - (6/10 x 5) / 5 = -4.00%. or (7/5) - (6/4) x 4/10 x 100 = -4.00%

The IMPORTANT numbers in these calculations are the ratio of numbers EXPECTED for BOTH the box number being bet and the hand ending seven.

Obviously, expectation for both the six and eight is FIVE of each sister for every SIX sevens.

Also evident in the calculation is the unfair, lower than fair payment for the bet being made. In the six/eight calculation, the house pays us \$7 for every \$6 risked (7/6 = 1.167) instead of \$6 for every \$5 risked (6/5 = 1.200). These unfair payments are exactly what makes each and every HE calculation positive for the house and negative for the player.

The ONLY way for a SHOOTER to overcome this is to toss more of the box number he is betting on and/or less of the hand ending sevens.

Nearly completed and to be continued.

#2
Last edited: May 14, 2021
HornHiYo, amzngguy and yacraps like this.
3. von duck, May 14, 2021

### von duck Member

Joined:
Feb 10, 2017
Messages:
14,467
3,069
Gender:
Male
It does you no good to throw less 7s, you have to throw more of your box numbers.

#3
4. DeMango, May 14, 2021

### DeMango Member

Joined:
Dec 20, 2010
Messages:
6,063
2,631
Gender:
Male
That may entail di, which I’m sure he will deny.

#4
5. von duck, May 14, 2021

### von duck Member

Joined:
Feb 10, 2017
Messages:
14,467
3,069
Gender:
Male
Well, "DI" is easy to deny and difficult to prove. In any case, if you threw a lot of your box numbers, it would pay the same, if it was done by luck, or by skill. Seems to me, that it would be to a "so called, DI's" benefit, if everybody thought it was impossible. You know, sort of "fly under the radar".

#5
6. KokomoJoe4, May 14, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
Ok, now finally, our check to see if the so-called DI is actually an advantaged player.

Wow, for this session, where he held the dice for over a half hour on one of his shoots, the numbers look fantastic. His four other shoots were not this good but were not too bad. He had over 40 rolls on his one shoot, and a total of 78 rolls for his five shoots.

Results recorded show he made 8 sixes, 5 eights and 4 nines. Guess how many hand ending sevens he had? Dave?

So for this session, his results are:

For the six:

scDI = (7/6) - (5/8) x 5/11 x 100 = + 24.6%

For the eight:

scDI = (7/6) - (5/5) x 5/11 x 100 = + 7.58%

And finally the nine:

scDI = (7/5) - (5/4) x 4/10 x 100 = + 6.00%

Negative expectation my ass. Maybe for me and for you, but not for the scDI, who's numbers prove that we can drop the sc from his title.

Forward fifteen years and lots and lots of sessions and shoots, many of which were not quite as stellar as the one described above. Some good ones sure, but we must also include the bad ones.

We are actually not even concerned about the sessions, but rather only with roll results for our scDI.

Over many years of play, the six calculation becomes:

scDI = (7/6) - (192/163) x 5/11 x 100 = - 0.51%

For the eight:

scDI = (7/6) - (192/167) x 5/11 x 100 = + 0.77%

And the nine calculation:

scDI = (7/5) - (192/130) x 4/10 x 100 = -3.08 %

These results are much worse than those calculated for the stellar session, but that is expected because no one shoots that well every time. In spite of this, the calculated results for this shooter are still BETTER THAN EXPECTATION for all three numbers, and are even POSITIVE for the eight.

Does this verify that our so-called DI is a DI? Some would say yes. I would say that they show he is a good shooter.

Obviously the numbers posted above are all hypothetical. I have NOT analyzed the roll data for this shooter. Rather, I simply made the results up.

My point with this is simply to show, that IF you track your every bet and your every result when shooting, you can plug in your numbers into the simple formula that applies to the number in question, and that the ONLY THING THAT MATTERS IN ANY OF THESE CALCULATIONS IS THE RATIO OF HAND ENDING SEVENS TO THE BOX NUMBER BEING CONSIDERED.

#6
HornHiYo and yacraps like this.

7. ### Liman Member

Joined:
May 4, 2013
Messages:
9,945
9,877
I measure my success by trips to the atm

#7
HornHighBLEVE and yacraps like this.
8. KokomoJoe4, May 15, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
As shown in post 6, if the scDI tracks his wagers made and his outcome results for each toss, he does NOT need a spreadsheet to determine how his results for every one of his wagers compares to expectation, which is ALWAYS negative in a casino game of chance.

Since the numbers he inputs to do the calculation CHANGE every time he plays, this calculation is ALWAYS a work in process.

IF calculations show a result that is a more positive number than expected, even if it is still a negative number , you might be a scDI. COMPARISON OF THIS NUMERIC RESULT TO EXPECTATION IS WHAT DEFINES THE "SKILL" OF THE SHOOTER.

Our scDI in post 6 beats expectation for all three bets that he makes on his signature numbers since with the six, -0.51% > -1.52%, and for the eight +0.77% > -1.52% and for the nine -3.08% > -4.00%.

We do NOT need to do statistical analysis of this data to show it is better than random. The calculated results TELL us this. Our scDI is not a whole lot better than random, but he is better

The way in which a Chi squared calculation supplements this information is very simple.

A Guassian curve for each of our shooters wagers/results is made for each bet. For the Place six and eight, the center of this curve is at -1.52%. For the nine, center is at -4.00%.

Typically and almost infallibly, 67% of all results are within +/- one standard deviation of the mean (-1.52% for the 6 & 8). 95% of all results are within +/-2 std's and 99% within +/-3 std's.

With respect to "proving" that a scDI's results are significantly better than expectation, Chi Squared wants them to be at least +1 std,

AFAIK, only Dave, UCDT and LID are in the > +3 std 0.1% range.

#8
yacraps likes this.
9. Dave G Ct, May 15, 2021

### Dave G Ct Member

Joined:
Mar 15, 2015
Messages:
8,245
1,485
Gender:
Male
Koko

You have not addressed the so called " lucky shooter" who has decent rolls just randomly tossing!

#9
10. KokomoJoe4, May 15, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
How would you describe the guy holding the dice for over 30 min whose session numbers are listed in the first part of post 6? You might say he is skilled, while I might say he is lucky.

#10
yacraps likes this.
11. Dave G Ct, May 15, 2021

### Dave G Ct Member

Joined:
Mar 15, 2015
Messages:
8,245
1,485
Gender:
Male
Yes you might. But if he is a DI or just lucky who cares? Everyone making \$

#11
12. KokomoJoe4, May 15, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
Everyone understands that the objective at craps is to win money, regardless of how it comes your way.

The entire POINT of my lengthy posting is to show that it is relatively simple to calculate how "good" or "bad" a shooters results are with the dice. We know what the negative expectation is for every bet that can be made, and we can easily calculate whether a certain shooters results are better or worse than normal/expected. We can determine this for a single session, and we can determine this for his entire "career".

All we need is an accurate account of his results for every bet he places and every toss he makes.

#12
yacraps likes this.
13. von duck, May 15, 2021

### von duck Member

Joined:
Feb 10, 2017
Messages:
14,467
3,069
Gender:
Male

#13
14. Dave G Ct, May 15, 2021

### Dave G Ct Member

Joined:
Mar 15, 2015
Messages:
8,245
1,485
Gender:
Male
Yes you can easily calculate that. Unless you are the Duck who would resort to silly disguises .

#14
FredP likes this.
15. twodicebilly, May 15, 2021

### twodicebilly Member

Joined:
May 9, 2016
Messages:
5,396
2,056
Gender:
Male
Location:
wisconsin
dave

The game of craps is played in the short run and it is played one of two ways. Either you think
you can beat odds and you bet more or only on yourself, or you bet on everyone. those are both
playing craps, and i am not saying one way is better than the other for other people...but they
are different.

If your a so called DI what ever that means and practices on your home table and throwing
station, keep track of you outcomes on your table as it is, then add 2 inches, then 4, then six
on your side wall where your shooing from, if your less than 6 2 you will see a change in your
results., then move from sl1 to sl2 to sl3 and do it, then add to the height of your side walls.
then record along the way and any honest player will tell there is going to be 1 or 2 spots and heights
that work best for you.

Now go pay some people to help you learn how to shoot, and a bunch get together so no one
knows where they are shooting from, then move from casino to casino in a group, then wonder
what in the hell you paid money for because you as a person is more than likely shooting
in unfavorable conditions.

I love taking the classes i have, not griping about any of them, but i offer some really important
advice, learn your best shooting position and understand the conditions you have at home, and make
an effort to match those conditions as best you can at the casino. Dont think your good enough to

Dicesetter

#15
16. DeMango, May 15, 2021

### DeMango Member

Joined:
Dec 20, 2010
Messages:
6,063
2,631
Gender:
Male
4% on the nine???? How about a \$36 nine paying \$53 after vig?

#16
17. KokomoJoe4, May 15, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
Yes, that is a much smarter way to play the five/nine, IF the casino operates that way. Mine in Pittsburgh does. Yours in Biloxi does. The great majority of places do not..

The \$53 winner on the \$36 wager calculates to -1.11% house edge, a bit better than the place nine.

You still need to make the nine before the hand ends, but lower edge against is always better.

#17
yacraps likes this.
18. DeMango, May 15, 2021

### DeMango Member

Joined:
Dec 20, 2010
Messages:
6,063
2,631
Gender:
Male
You crack me up with your math. So 1.11 is a bit better than 4.00? Almost a factor of four times worse?

#18
19. KokomoJoe4, May 15, 2021

### KokomoJoe4 Member

Joined:
Sep 28, 2013
Messages:
6,198
5,471
Show me where any of "my math" in this thread is incorrect.

Yes, if we are buying any of the outside or odd box numbers, buy calculations must be used that relate to the wager you are making. These calculations will differ, both for the 5/9 and the 4/10, depending on whether vig is paid on the bet or on the win.

None of this matters. Use the correct formula, put in YOUR numbers that relate to YOUR wagers and YOUR results.

Only things that matter: (1) record wagers and results accurately, (2) do the math properly and (3) the deciding factor, YOUR ratio of hand ending sevens to the box number.

Result of the calculation tells you either that you are a DI or you are not.

#19
Last edited: May 15, 2021
yacraps likes this.
20. von duck, May 15, 2021

### von duck Member

Joined:
Feb 10, 2017
Messages:
14,467
3,069
Gender:
Male
He did not say your math was incorrect. He said it cracked him up.

#20