1. Welcome to the #1 Gambling Community with the best minds across the entire gambling spectrum. REGISTER NOW!
  2. Have a gambling question?

    Post it here and our gambling experts will answer it!
    Dismiss Notice
  3. Discussions in this section are assumed to be EV- as they are outside of the Advantage Play section. For EV+ discussions, please visit the Advantage Play section.
    Dismiss Notice

Craps Would like comments on Craps strategy (maybe winning)

Discussion in 'Craps Forum' started by gerdm07, Aug 16, 2022.

  1. gerdm07

    gerdm07 New Member

    Joined:
    Aug 14, 2022
    Likes:
    0
    Location:
    USA
    First post! I started studying the math for craps during the pandemic and eventually came up with this strategy.

    COME OUT

    It’s well known that the pass line bet is favorable to the better and in the long run you will get a return of about 11.11%. In other words, if you bet $10 you will make about $1.11 per roll. The problem is after a point is established the house is favored.

    My strategy is to bet $10 on the pass line and once a point is established, place equal odds behind the pass bet and Lay $50 on either 4 or 10. The Lay bet will pay 12:25 (Bovada online) and will cover the $20 bet on the pass line. The only time you lose is if the Lay bet number is rolled.

    Here’s the math for the Pass Line bet

    Roll...Winnings...Ways...Total
    2..........-10.........1.......-10
    3..........-10.........2.......-20
    7..........+10........6......+60
    11........+10.........2......+20
    12.........-10.........1.......-10

    We win a total of $40 every 36 rolls and 40/36 is 1.11111. We win $1.11 per roll! We are a winning player at this point.

    POINT ESTABLISHED

    Add $10 odds behind the Pass Line bet.
    If the Point is 6, 5, or 4, Lay $50 on 10
    If the Point is 8, 9 or 10, Lay $50 on 4

    Here is the the math if the Point is 6 and we make the above bets:

    Roll...Winnings...Ways....Total
    6.......10+12........5.......+110
    7.......+24-20.......6.......+24
    10.......-50...........3........-150

    So we lose $16 every 36 rolls and -16/36 is -0.444. We lost $0.444 per roll when a 6 or 8 is the point number. Doing the math for each point, we get

    Point...... Lose.......Per Roll
    6 or 8......-16........-0.444
    5 or 9......-26........-0.772
    4 or 10....-36.........-1.000

    If we lose our Lay bet, we make the same Lay bet because we know this will cut our variance.
    If we hit the point (best result), we take down the Lay bet and start fresh. We want to preserve the advantage on the Come Out.

    We are a winning player if we compare wins and losses per roll in the various situations. However, you will not roll the same number of Come Out rolls and Point rolls.

    29.6% OF ROLLS ARE COME OUT

    Craps simulators have demonstrated that the Come Out roll will occur about 29.6% of the time. Let’s use this and do some more math.

    We will win $1.11 for each $10 Pass Line bet and this should occur 29.6% over a long period. Therefore, our winnings are 0.296*1.11, or $0.329 per roll.

    The 6/8 should be the point number 27.7% of the time. (10 out of 36)
    The 5/9 should be the point number 22.2% of the time.
    The 4/10 should be the point number 16.6% of the time.

    Our per roll losses during a Point should be

    27.7% of 70.4% of 0.444, or $0.087 loss for 6/8
    22.2% of 70.4% of 0.772, or $0.113 loss for 5/9
    16.6% of 70.4% of 1.000, or $0.117 loss for 4/10

    0.329 – 0.087 – 0.113 – 0.117 = $0.012

    This strategy should win a little more than 1 cent per roll! (A 1.2% ADVANTAGE)

    QUESTIONS

    - Is my math correct?
    - Is this really a winning strategy?

    NOTES
    • The math does not change if you decrease or increase your odds bet because that bet gets true odds. You would just be increasing your variance with larger odd bets.
    • I’ve been using this strategy with play money at Wizard of Odds with success. What I really like about the strategy is most of the time you win, either with a 7-out or the point number hitting. Once I build up a little extra money, I make a big Lay on 4 or 10 and hope. This makes it a little more exciting and I can end up with a good stack with some luck. If I lose that bet, I grind away and build my stack again.
    • Obviously, there are occasions where the Lay bet might lose 2, 3 or even 4 times in a row and you would be a loser that day. However, this should not happen too often. (I have not done the math for this)
    Thanks for your comments!
     
  2. Chip Magnet

    Chip Magnet Active Member

    Joined:
    Aug 19, 2019
    Likes:
    31
    Location:
    NYC
    I don't trust the Wizard of Odds simulator. At all.

    Invest in WinCraps Pro, for a good simulator.

    Concentrate on lowering the house edge.

    You started out with a pass Line bet, that has a house edge of 1.41%. Then you lowered the overall HE to about 0.8%, with single odds.

    Then you more than doubled the overall HE, to 1.88%, with a Lay on the 4 or 10.

    I ran your strategy with WinCraps, using 3,150 real dice table rolls.

    It lost $600. Total amount bet was $44,580.

    Instead of laying the 4 or 10 for $50, you can use that money to take higher odds on your pass line bet.

    Odds have a zero house edge.

    A $10 pass line bet with 3x odds ($30), and no other bet, won $213 over the same 3,155 rolls. An $800 improvement.

    House Edge for 3x, was 0.47% vs 1.88%.

    Total amount bet was $28K vs $44K

    4x odds won $264

    5x odds won $315
     
  3. gerdm07

    gerdm07 New Member

    Joined:
    Aug 14, 2022
    Likes:
    0
    Location:
    USA
    Thanks for the reply. I downloaded WinCraps and am learning how to use it.

    The Pass Line bet is a winning bet for the Come Out with a player advantage of 11.1%. It's the best bet in any Casino. When the point is established then the house has the advantage.

    Higher odds bets only changes your loss percent or win percent. It does not change your per roll win or loss because they pay true odds.

    How is my math incorrect?
     
  4. TwoUp

    TwoUp Well-Known Member

    Joined:
    Nov 30, 2020
    Likes:
    328
    Occupation:
    Unknown
    Location:
    Nowhere
    Craps is an interesting game.

    Your first error is you have not accounted for all outcomes:
    27.7+22.2+16.6 ≠ 100​

    The next error related to the first is diluting the losses, when you multiply by 70.4% (100 - 29.6%).

    The third error is that 29.6% is a statistical figure about rolls, not point cycles.

    You knew that you need to account for the comeout but using your statistical number still doesn't equal 100%, all outcomes should always add up to 1 (100%):
    29.6+27.7+22.2+16.6 = 96.1 ≠ 100​

    You already established there are 12 combinations on the comeout that end the point cycle, the remaining 24 combinations set the point. Of those 24 combinations they resolve to a net loss on average (which you have shown), those losses are greater than the advantage on the comeout. All 36 combinations are accounted for.

    I develop craps strategies for entertainment, here is one you might like which is similar, as it hedges the point but it is played from the don't. I have recorded 4 sessions to get a feel for it under both adverse and favorable conditions.

     
    Last edited: Sep 7, 2022
  5. Alan Shank

    Alan Shank Member Founding Member

    Joined:
    Dec 31, 2014
    Likes:
    17
    Occupation:
    retired software engineer
    Location:
    Woodland, CA
    The fact that the "right" player has an advantage on the comeout (twice as many ways to win as to lose) is really irrelevant to a strategy, because it is a contract bet that you cannot take down, so you have to analyze both parts of the bet. The ev, BTW, is +.33 just looking at the comeout roll. Once a point is established, the ev's are:
    4/10 -.33
    5/9 -.20
    6/8 - .909
    If you put them all together, the ev is -1.414%, so you have an expected loss of $.1414 for a $10 bet.
    Here's a handy way to look at the passline bet -- the "perfect 1980":
    For every 1980 passline decisions, the expectations are:
    win comeout 440
    lose comeout 220
    point 6 or 8 550 (win 250, lose 300)
    point 5 or 9 440 (win 176, lose 264)
    point 4 or 10 330 (win 110, lose 220)

    Add 'em all up, you get -28, divide by 1980 = -.01414
    Adding up the losses on points, you get 784, so 784 of 1980 decisions are seven-outs, or 39.6%.
    1 / .396 = 2.52 decisions per hand (shooter).
    The average number of rolls to a decision is 3.375, so a hand (shooter) averages 8.525 rolls.

    It looks like the lay bet pays $24, so that's a $1 vig, but if you only lose $50 that means they are only charging the vig on winning bets, which is good, and they're also rounding down the 5% vig to $1 from $1.20. This means the ev for the lay bet is -.66, and the house advantage is just -.1333.

    So, bottom line, your calculations are in error, and you do not have an advantage, as Chip Magnet and TwoUp demonstrated.

    The simulation Chip Magnet did on WinCraps is not very meaningful IMO, because it's not a big enough sample. If you learn to write autobet files in WinCraps you can try it with a couple of million rolls. I run simulations in individual sessions, with defined and points, like a certain number of rolls (simulation time at the table), a loss limit or a win goal, or all three. Then I run 10,000 sessions and examine the results; this gives you a very good picture of what your chances are of having a winning session. Of course, the dice will select one of those possibilities for you.
    Good luck,
    Alan Shank
     
  6. Alan Shank

    Alan Shank Member Founding Member

    Joined:
    Dec 31, 2014
    Likes:
    17
    Occupation:
    retired software engineer
    Location:
    Woodland, CA
    I will simulate this for you on WinCraps, running 10,000 sessions. Give me:
    your starting bankroll for a session
    one or more ending conditions for a session, eg, a time limit (WinCraps uses 120 rolls/hr, so the limit will be a certain number of rolls), a loss limit (if any) and/or a win goal (if any).

    The sim will show you a large range of possible outcomes of any given session; the dice pick one out for you when you play.
    Cheers,
    Alan Shank
     

Share This Page