Sam Grafstein's methods
Dan Moore
Colorado expanding its gaming laws, casinos with craps games are just an
hour and a half away.
Previously, I used to play PL w/odds with two come bets w/odds. That
boom or bust method was OK for the occasional vacation, but I quickly
learned that I could not sustain that type of betting if I were going to
play more often locally.
So I began to educate myself -- reading books, doing the math, writing
computer programs, dealing at parties.
Since adopting the Dice Doctor's methods earlier this year, overall I
have been on the plus side of things (playing 2-3 times a month). Now
that may just be statistical variation, but so far it has been working.
I starting this thread to discuss Sam's methods in more detail, get some
clarification and opinions.
- Progression
I've seen an analysis of where sticking with a constant flat bet vs.
increasing the bet upon wins. For a shooters with a few passes vs. many,
a person is better off with the constant bet. I acknowledge that, but
I'm on the side of progression -- wanting to take advantage of that long
winded shooter when he or she comes along.
The portion of the progression that I'm interested in is:
Flat bet increase of last bet
5
7 40%
10 43%
15 33%
25 67%
35 40%
50 43%
75 50%
125 67%
Well I have to modify this table since the limit of an individual bet in
Colorado is $100. So I bring the $125 down to $100 and the $75 to $70,
resulting in this:
Flat bet increase over last flat bet
5
7 40%
10 43%
15 33%
25 67% <-----------
35 40%
50 43%
70 40%
100 43%
Moving from $25 from $15 does seem that I'm not retaining enough of my
last winnings.
Can somebody suggest an alternate progression? ...hopefully one that
won't annoy the dealers too much with dollar chips.
- Single odds values
I've chosen to go with single odds instead of double odds. Several
factors for this:
* keeps my session stake lower
* I not allowed make a full double odds bet on any flat bet above $50
(there's that Colorado $100 limit again).
* The house advantage between 0.8% and 0.6% is negligible
In one place in the book, Sam writes something to the effect that we
shouldn't favor one number over the other when placing odds. (like
placing a smaller odds bet on 4 and 10 vs. 6 and 8 due to the lower
probably of hitting a 4 or 10 vs. a 6 or 8) He writes that the laws of
probabilities don't mean much on a short term basis -- I agree with him.
However, it seems a bit contradictory in his odds table. For instance:
flat bet 4/10 odds 5/9 odds 6/8 odds
================================================
15 15 20 25
...
35 35 40 50
Does anyone have a theory on this? ...or have adopted an alternate
"single" odds schedule.
Dan
Alan Shank
>I've been a once or twice a year craps player until recently. With
>Colorado expanding its gaming laws, casinos with craps games are just an
>hour and a half away.
>
>Previously, I used to play PL w/odds with two come bets w/odds. That
>boom or bust method was OK for the occasional vacation, but I quickly
>learned that I could not sustain that type of betting if I were going to
>play more often locally.
What odds multiple were you taking? If you were not maxed out on odds,
you could have taken more odds instead of making the come bets,
avoiding the 1.4% "tax" on the flat portion of the come bets. It would
still be a pretty high-variance system, however.
>
>So I began to educate myself -- reading books, doing the math, writing
>computer programs, dealing at parties.
>
>Since adopting the Dice Doctor's methods earlier this year, overall I
>have been on the plus side of things (playing 2-3 times a month). Now
>that may just be statistical variation, but so far it has been working.
Which methods? He has lots of different methods in "The Dice Doctor".
>
>I starting this thread to discuss Sam's methods in more detail, get some
>clarification and opinions.
>
>
>- Progression
>
>I've seen an analysis of where sticking with a constant flat bet vs.
>increasing the bet upon wins. For a shooters with a few passes vs. many,
>a person is better off with the constant bet. I acknowledge that, but
>I'm on the side of progression -- wanting to take advantage of that long
>winded shooter when he or she comes along.
>
>The portion of the progression that I'm interested in is:
>
>Flat bet increase of last bet
> 5
> 7 40%
> 10 43%
> 15 33%
That's 50%, right?
> 25 67%
> 35 40%
> 50 43%
> 75 50%
>125 67%
>
>Well I have to modify this table since the limit of an individual bet in
>Colorado is $100. So I bring the $125 down to $100 and the $75 to $70,
>resulting in this:
>
>Flat bet increase over last flat bet
> 5
> 7 40%
> 10 43%
> 15 33%
> 25 67% <-----------
> 35 40%
> 50 43%
> 70 40%
>100 43%
>
>
>Moving from $25 from $15 does seem that I'm not retaining enough of my
>last winnings.
>
>Can somebody suggest an alternate progression? ...hopefully one that
>won't annoy the dealers too much with dollar chips.
15 50%
20 33%
25 25%
30 20%
40 33%
50 25%
etc., keep it between 20 and 33%.
>- Single odds values
>
>I've chosen to go with single odds instead of double odds. Several
>factors for this:
> * keeps my session stake lower
> * I not allowed make a full double odds bet on any flat bet above $50
>(there's that Colorado $100 limit again).
> * The house advantage between 0.8% and 0.6% is negligible
The house advantage is misleading; you should consider the "Expected
Value" instead. Those .8% and .6% figures are based on the expected
value divided by the combined flat and odds portions. The flat amount
is ALWAYS subject to the 1.4% advantage, the odds amount to none. I
don't know what the odds limit is where you play, but consider this:
$100 flat bet, no odds: expected value -$1.40
$50 flat bet with $50 odds: expected value -$.70
$25 flat bet with $75 odds: expected value -$.35
$10 flat bet with $90 odds: expected value -$.14
To the extent allowed, you can reduce your expected loss by minimizing
the flat portion, while keeping your volatility about the same. For
example:
$50 pass, $50 odds $25 pass, $75 odds
for 60 bets, about 200 rolls, about 2 hours (depending on conditions)
EV: -$42.42 -$21.21
SD: $733 $741
You "save" over $20 per 2-hour session.
>In one place in the book, Sam writes something to the effect that we
>shouldn't favor one number over the other when placing odds. (like
>placing a smaller odds bet on 4 and 10 vs. 6 and 8 due to the lower
>probably of hitting a 4 or 10 vs. a 6 or 8) He writes that the laws of
>probabilities don't mean much on a short term basis -- I agree with him.
>
>However, it seems a bit contradictory in his odds table. For instance:
>
>flat bet 4/10 odds 5/9 odds 6/8 odds
>================================================
> 15 15 20 25
>...
> 35 35 40 50
>
>Does anyone have a theory on this? ...or have adopted an alternate
>"single" odds schedule.
It looks like those are just adjusted for correct payoffs, but at the
5-unit level. IOW, the 5/9 bet must be divisible by 2, to get the 3:2
payoff, the 6/8 by 5 to get the 6:5. Also, many casinos now offer 3,
4, 5X odds, so the payoff is always 6 units.
The reason some people don't like to take the same odds on 4/10 is
that the variance is higher; the standard deviations are:
4/10 1.41 * bet
5/9 1.22 * bet
6/8 1.10 * bet
The more volatility you build into your system, the more extreme will
be the swings between good sessions and bad ones. You should
understand that progressive betting increases your volatility very
substantially over flat betting. Every betting method that varies the
bet amount is "chasing" a certain pattern. When you progress your
bets, you are chasing a winning streak; you might win more bets than
you lose but still come out behind.
Most books on craps do not get into volatility, concentrating on house
advantage and "money management". Several years ago, I wrote a
detailed review of "The Dice Doctor". If you are interested, I could
try to dig it up and e-mail it to you.
Cheers, and good luck,
Alan Shank
The Midnight Skulker
> I have been on the plus side of things (playing 2-3 times a month).
> Now that may just be statistical variation, but so far it has been
> working.
If it's not broken, don't fix it, but recognize that Dr. Sam's
methods do not defeat the house advantage. A positive attitude is
good; a positive expectation is not.
> I've seen an analysis of where sticking with a constant flat bet
> vs. increasing the bet upon wins. For a shooters with a few passes
> vs. many, a person is better off with the constant bet. I acknowl-
> edge that, but I'm on the side of progression -- wanting to take
> advantage of that long winded shooter when he or she comes along.
As Alan noted, any deviation from flat betting looks for a pattern of
wins and losses. Winning progressions, where bets are increased
after a win and dropped back to some base level after a loss, look to
maximize profits at tables where wins come in streaks while accepting
losses, albeit small ones, at choppy tables, where flat betting might
have shown a small gain.
> The portion of the progression that I'm interested in is:
I think it might be of more interest to see how much you have made in
a series if you were to lose the bet you are making, so I have added
two columns to your tables.
Bet % increase net (progression) net (flat betting)
5 -5 -5
7 40 -2 0
10 43 +2 +5
15 33 +7 +10
25 67 +12 +15
35 40 +27 +20
50 43 +47 +25
75 50 +72 +30
125 67 +122 +35
To explain what I have done look at the fifth line, the one with the
$25 bet. My "net (progression)" column shows that if you lose this
bet you will still be up $12 for the series (because you won bets of
$5, $7, $10, and $15 previously). My "net (flat betting)" column
shows that had you not been progressing your bets, but betting $5
each time, you would be up $15. The table shows that your progres-
sion needs six consecutive wins to outperform flat betting, but the
difference between the two methods increases dramatically after that
point.
> Well I have to modify this table since the limit of an individual bet in
> Colorado is $100. So I bring the $125 down to $100 and the $75 to $70,
> resulting in this:
Bet % increase net (progression) net (flat betting)
5 -5 -5
7 40 -2 0
10 43 +2 +5
15 33 +7 +10
25 67 +12 +15
35 40 +27 +20
50 43 +47 +25
70 40 +77 +30
100 43 +117 +35
> Moving from $25 from $15 does seem that I'm not retaining enough of
> my last winnings.
The key question you must answer, then, is how much profit you want
after four consecutive wins. There is no rule that forces you to
increase your bets once you have started a progression. What if,
after four wins, you bet $15 again, progressing to $25 only if that
fifth bet won? Alternately, you could keep your own bet at $15 and
put the crew up for a dollar or two. I call this a "double clutch"
and use it in my place betting progressions as well as after natural
winners on the line.
> Can somebody suggest an alternate progression? ...hopefully one
> that won't annoy the dealers too much with dollar chips.
My own progression is 5, 8, 10, 15, 20, 25, 30, 40, 50, 75, 100.
> - Single odds values
>
> I've chosen to go with single odds instead of double odds. Several
> factors for this:
> * keeps my session stake lower
> * I not allowed make a full double odds bet on any flat bet above
> $50 (there's that Colorado $100 limit again).
> * The house advantage between 0.8% and 0.6% is negligible
Might I suggest progressing your odds from single to double before
progressing your flat bet? You don't necessarily have to go from
single to double in one jump. Maybe 5+5, 5+10, 7+10, 7+14/15,
10+14/15, 10+20/25, 15+20, 15+25/26, 15+30, etc. You would still be
taking advantage of a hot roll, but putting more of your money on a
bet with no house advantage.
1 2
| The Midnight Skulker
9 * 3 aka Van Lewis
aka cvle...@gmail.com
6
Nick Naim
"Dan Moore" <moor...@suncup.net> wrote in message
news:4b36ade5$0$48223$815e...@news.qwest.net...
Always use dx odds on all nbrs.
NEVER NEVER start scared using 1x odds to start off with.
5 5 7 10 15 20 25 ........ECT
> The two 5,s get you a profit however small if the shooter made his first
> point then sevened out on the second point.
The 20 inserted is for the typical NO GO brick wall
I made it once up to a $250 line bet @ 2x odds.
During this shoot there where no back down 2 3 or 12,s
Ye........my own
always use double odds
>
>
>
>
>
>
>
>
>
>