444 Craps System

Craps Strategy: Al Kaufman's 444. I don't recall the specifics of the system, but it was something like $32 across (Level 1), increased to $192 across on a 7 out. Craps is not a game that can be beaten by using betting systems. If the game is random, as the casino desires it to be, you will lose whatever the house edge is on the total amount of money you bet. That is an uncomfortable fact that many craps players and some craps authors just can't seem to assimilate. Craps is instead a game of shooters. A craps game is run by two dealers, a stickman and a box man. There's one dealer for each side of the table. Each side of the layout is a mirror image of the other.

Although I have tested a lot of systems, I don't need to test all of them to know they are all worthless. No system can ever pass the test of time. It is not unusual to win for a while with a system, but if you keep playing the odds will eventually catch up to you and you will fall behind.
For more information about the futility of betting systems, please see The Truth about Betting Systems.

That is known as 3-4-5X odds, and is now pretty common. The following table shows all the possible outcomes, for the pass and odds combined, with full odds.

Return Table with 3-4-5X Odds

Level Four Craps System

Unlike most gambling writers, I don't put much emphasis on betting strategies. Assuming the same game and bet, there is no one right or wrong strategy. They all behave differently in the short run, but in the long run you will give the house the same percentage of total money bet.

This is similar to a question I got last week. Yes, it is true that there are ten ways to roll a 6 or 8, and six ways to roll a 7. However, one must not look at the probabilities alone, but weight them against the payoffs. The place bet on the 6 and 8 pays 7 to 6 odds when fair odds would pay 6 to 5. By making six unit place bets on the 6 and 8, and taking the other down if one wins, the probability of winning 7 units is 62.5% and the probability of losing 12 units is 37.5%. If the player must cover both the 6 and 8, then the place bet is the way to go. This rate of return isn't bad but could be better. For the player who puts a priority on minimizing the overall house edge, the best strategy is to make combinations of pass, don't pass, come, and don't come bets, and always take the maximum allowable odds.

The better system is to bet on the don't pass only and take full odds. Yes, betting on both does increase you chances of winning on any one bet. However you are suffering a higher combined house edge by betting on both the pass and don't pass and it will cost you in the long run.

Yes, it was luck. It helped that you stuck to the low house edge bets. However, next time, make the line bets with odds only, and don't bet the field, especially if it pays 2 to 1 only on both the 2 and 12.

No combination of bets can give the player an advantage. In your example you would lose one unit for every 12 on the come out roll. You don't make up for it laying the odds. While you usually win laying the odds, you have to risk more. In the end, laying the odds has zero house edge.

As long as you are backing up your pass and come bets with full odds, it doesn't make any difference how many come bets you make. However, it does reduce the overall house edge to keep the odds on your come bets working on the come out roll.

You should never remove a don't pass bet after a point is made! Once a point is made of 6 or 8 the don't pass has equity of 9.09% of the bet amount, which you would be throwing away by taking the bet down. The equity of a don't pass bet on a point of 5 or 9 is 20%, and on a 4 or 10 is 33.33%.

Thanks for the compliment on my site. The best thing I can say about this system is that it composed of low house edge bets. Yes, a 12 will lose the pass bet and push the don’t pass on the come out roll, this is where the house edge is. By making the pass bet you are increasing the overall house edge. If you’re afraid losing you shouldn’t be playing at all. Never hedge your bets. So my advice is to stick to just the don’t pass and laying odds. Yes, you’ll lose some on the come out roll. However if you don’t lose on the come out roll the don’t pass bet will usually win.

1. Wait until the shooter establishes a point.
2. Play both come/don’t come (same amount). Until you have a maximum of four numbers
3. After the shooter has rolled five times without rolling a 7, take odds on all your numbers on the front side.

The rationale: Limit your exposure until you find a 'qualified' (five rolls without a 7) shooter. Only betting the odds so there is no 'house edge'! Can you compare this system with just playing pass/come and taking the odds?

As I stated in the other craps strategy question you are only mixing another house edge bet into the game by betting on both the pass and don’t pass, or come and don’t come. It is also not going to help to wait until a shooter hits five points. The probability of making a point is the same for me and you as it is for somebody who just threw 100 points in a row. In other words, the past does not matter. As I stated to the person who asked the other question (whom I think may also be you) don’t make opposite bets, just stick to either the do or don’t side and always back up your bets with the odds.

Unless bankroll preservation is very important to you then Kelly betting won’t help. I would just flat bet. Nice strategy to milk the comp system.

• 5- place $5
• 6- place $6
• 8- place $6
• field- $5
• total= $22

Good question. To confirm their math I made the following table, based on a field bet paying 3 to 1 on a 12. The lower right cell does shows an expected loss of 25 cents over $22 bet. So the house edge is indeed .25/22 = 1.136%.

Mensa Anything but Seven Combo

The reason the overall house edge appears to be less than the house edge of each individual bet is because the house edge on place bets is generally measured as expected player loss per bet resolved.
However, in this case the player is only keeping the place bets up for one roll. This significantly reduces the house edge on the place bets from 4.00% to 1.11% on the 5 and 9, and from 1.52% to 0.46% on the 6 and 8.
For you purists who think I am inconsistent in measuring the house edge on place bets as per bet resolved (or ignoring ties) then I invite you to visit my craps appendix 2 where all craps bets are measured per roll (including ties).

Craig from Los Angeles

No. I had to Google this to find out what this is. This appears to me to be an amusing urban legend about some young scientists who developed a winning craps system. The story is told at Quatloos. I would file this under other fictional stories that have become mistaken for fact, like Joshua’s missing day. As I have said hundreds of times, not only can betting systems not beat games like craps, they can’t even dent the house edge.

If the player bets $5 on the field and 5, and $6 on the 6 and 8, then he will have a net win of $2 on the 5, 6, and 8, $10 on the 2, $15 on the 12, and $5 on the other field numbers, assuming that the 12 pays 3 to 1 on the field. The player will lose $22 on a 7. On a per roll basis, the player can expect to lose 25 cents compared to $22 in bets, for a house edge of 1.136%.

This begs the question, why is this lower than the individual house edge of each bet made? It’s not. The reason it seems that way is the result of comparing apples to oranges. The house edge of place bets is usually expressed as the expected loss per bet resolved. Looking at the individual bets on a per-roll basis, the house edge on the 5 is 1.11%, and on the 6 and 8 is 0.46%, according to my craps appendix 2. Comparing apples to apples, the house edge is a weighted average of the house edge on the field, 5, 6, and 8, on a per-roll basis, or (5/22)×2.778% + (5/22)×1.111% + (6/22)×0.463% + (6/22)×0.463% = 1.136%.

For the benefit of other readers, the 5-Count is a method of slow-playing craps, as discussed in ’Golden Touch Dice Control Revolution’ by Frank Scoblete and Dominator. As the book states, it is a way of betting nothing on some rolls, reducing your expected loss on random shooters, while still getting the full comp value of table time.

444 Craps System Free

The way the 5-Count works is you start counting rolls as soon as a new shooter throws any point number. When you get to five rolls after you start counting, the shooter is deemed worthy, and you start betting. However, you if the 5th roll is not a point number, it doesn’t count.

The book says you will only be betting 43% of the time, which I agree with. It is common for craps players to not bet, bet small, or bet the don’t pass on new shooters, as a way to qualify him. Once a shooter has made a point, or thrown lots of point numbers, the other players will gain confidence in him, and start betting with him. So, this kind of strategy seems natural. When casinos rate your average bet, they don’t lower the average for betting nothing some of the time. However, sometimes they will dock your time, especially if you are betting big.

An alternative strategy is to wait until the shooter makes a point. Under this strategy you will only be betting 40.6% of the time, less than the 43.5% with the 5-Count.

Yes! I’ve said many times that betting systems not only can’t beat a house edge game, they can’t even dent it. That includes denting it in the house’s favor. In other words, even if he tried to lose, he still only gives up 0.18% over the long-run, under your assumptions. Over a shorter time, he probably could do this, but not over 'years.' Some might argue that to deliberately lose, the player should do an anti-Martingale, where the player kept pressing his bets until he lost. However, a problem there is that a winning player will eventually reach the table maximum, which is rather low in craps. It just goes to show how futile betting systems are.

Disclaimer

One of the most hotly debated topics among advantage gamblers is whether or not it is possible to influence the throw of the dice in craps. Personally, I'm still skeptical. I don't rule out the possibility, but I'm not convinced. Most casinos happily allow it. If I ran a casino, I would allow it too, because I think the number of people who can influence the dice (if any) is far outweighed by the number who think they can, but can't. Still some people I respect do believe in it, mainly Stanford Wong, who trained under Golden Touch Craps. My craps appendix 3 outlines evidence I have gathered. That said, everything else in this page is from the hypothetical perspective that the dice can be influenced. The purpose of this page is to quantify the player advantage, according to his skill level, and recommend the best dice settings.

How Dice Control Allegedly Works

Even the strongest believers in dice control will admit that most throws, even of the best shooters, are still random. However, it takes a small percentage only of precise throws to overcome the house edge. What is happening on these successful throws? There are two schools of thought, or types of shooters. Both types of shooters set the dice in a certain way, hoping to keep them on axis, and together in rotation, as if the dice were glued together. Two things can go wrong after the dice leave the shooter's hands, and that is what divides the two types of shooters.

The first type of shooter is what I'll call the 'correlation shooter.' The correlation shooter does no better than a random shooter at keeping the dice on axis. However, when the dice do stay on axis, the rotations of the two dice are correlated. For a random shooter, if the dice did stay on axis, there would be a 25% chance of their landing with the same faces together, as when the dice were in the shooter's hand. The correlation shooter hopes to increase this probability above 25%, by reducing the probability of the dreaded double-pitch throw. A double-pitch is when both dice stay on axis, but one die rotates 180 degrees more than the other. Likewise, a single-pitch is when both dice stay on axis, but one die rotates 90 degrees more or less than the other. Based on a careful reading of Wong on Dice, for purposes of this page, I assume the correlation shooter shoots randomly, except the probability of zero-pitch throws is a certain percentage higher than expectations, at the cost of an equal reduction in double-pitches. Wong never states this is exactly what happens with correlation shooters; it is my own simplified interpretation. To get specific results from my analysis, I had to put in some specific assumptions.

The second type of shooter is what I'll call the 'axis shooter.' Not only can the axis shooter keep the dice correlated when they do stay on axis, but he can keep them on axis more than the expected 44.44% of the time of a random shooter.

Stanford Wong writes in 'Wong on Dice' that most careful shooters he observed were not keeping both dice on axis more than the random expectations, but were achieving influence through correlation. Based on my faith in Wong, the following tables are all under the assumption of correlation shooting only.

Dice Settings

There are 84 distinct ways to set the dice. In my analysis for this page, I examined all 84 sets, and noted the best set for each bet. The following sets are the only ones that came up as the best for the bets studied. For practical purposes, the only sets you need to know are Hard Ways set #1 and Sevens set #1. The other sets are either equally as good in some situations, or optimal for bets you shouldn’t be making, because better bets are available.

Dice Settings

Skill Factor

The skill factor is defined as the percentage of double-pitch throws that the skillful shooter turns into zero-pitch throws. A skill factor of zero would apply to a random shooter, where the probability of both a zero-pitch and double-pitch throw are each (2/3) × (2/3) × (1/4) = 1/9 = 11.11%. A skill factor of 12%, for example, would move 12% of double-pitches into zero-pitches. In this case, the probability of a double-pitch would be 11.11% × 88% = 9.78%, and for a zero-pitch would be 11.11% × 112% = 12.44%. All other outcomes would be the same as that of a random shooter.

Rolls to Sevens Ratio

Most of the time the shooter is going to want to avoid sevens. By far, the most common metric for measuring dice control is the 'Sevens:Rolls Ratio,' or RSR. As defined in 'Wong on Dice,' the RSR is the ratio of rolls to sevens. I believe that acronym is a misnomer, because the rolls to sevens ratio should be abbreviated RSR. So, I am going to break with convention and call it that. For a random shooter, the probability of a seven is 1/6, so the RSR would be 6. A skillful shooter should be able to throw fewer sevens, and thus increase the RSR above 6. As a basis of comparison to other sources, I will include the RSR in my house edge tables. The RSR's mentioned in 'Wong on Dice' are usually in the range of 6.3 to 7.0.

Pass Line with 3-4-5X Odds

The following table shows the player advantage on the pass line bet, with 3-4-5X odds, according to skill factor. I measured the house edge two different ways. The column for the house edge with the Hard Way set (HW#1), is the house edge if the shooter always uses the Hard Way set #1, even on a come out roll. The column for the house edge with the Hard Way and Seven sets (HW#1 & 7#1) applies when the shooter uses the Seven set #1 on a come out roll, and the Hard Way set #1 otherwise. The reason for listing the house edge for the Hard Way set alone is that many shooters also make come bets, which would lose on a seven on a come out roll. I've observed some so-called skillful shooters using the Hard Way set on a come out roll, even with no come bets. I believe the reason for this is ease in record keeping.

Pass with 3-4-5X Odds

Don't Pass Line with Laying 3-4-5X Odds

The following table shows the player advantage on the don't pass line bet, with 3-4-5X odds, according to skill factor. As with the pass bet, I measured the house edge two different ways. The first applies if the shooter uses the Seven set #2 on every throw. The second applies if the shooter uses the Hard Way set #2 on the come out roll, and the Seven set #2 on all other throws. Comparing this table to the above table, the player advantage on the pass bet is greater with a skill factor of 0.01 or greater. So, this table is not of much practical value.

Don't Pass, Laying 3-4-5X Odds

Place and Buy Bets

The next table shows the house edge for placing the 5, 6, 8, and 9, and buying the 4 and 10. When buying the 4 and 10, it is under the rule of paying the commission always. As the table shows, the greatest advantages are on the 6 and 8. If the shooter always used the Hard Way set #1 on the pass line, and had a skill factor of 0.17 or greater, then his advantage would be greater placing the 6 and 8 than the pass line bet with 3-4-5X odds.

Place and Buy Bets

Hop Bets

When I set out on this analysis I thought I would find that the advantage on hop bets was more than on the pass with odds. However, for the most part, it isn't. The only exception would be if your skill factor is at least 12%, and you are throwing at a table in the United Kingdom or Australia, where they pay 16 to 1 on easy hops, and 33 to 1 on hard hops.

Under the UK/Australian rules, my advice on setting for the hop bets is to set anything, and then make hop bets on each of the different combinations showing on the adjacent faces. For example, with Hard Ways set #1, bet on 2-2, 3-3, 4-4, and 5-5. With Sevens Set #1 bet on 1-6 and 2-5.

Hop Bets

Links

• Wong on Dice, by Stanford Wong.
• Golden Touch Dice Control Revolution!, by Frank Scoblete.
• Get the Edge at Craps, by Sharpshooter.

• Golden Touch Craps

Internal Links

• How the house edge for each bet is derived, in brief.
• The house edge of all the major bets on both a per-bet made and per-roll basis
• Dice Control Experiments. The results of two experiments on skillful dice throwing.
• Dice Control Advantage. The player advantage, assuming he can influence the dice.
• Craps variants. Alternative rules and bets such as the Fire Bet, Crapless Craps, and Card Craps.
• California craps. How craps is played in California using playing cards.
• Play Craps. Craps game using cards at the Viejas casino in San Diego.
• Number of Rolls Table. Probability of a shooter lasting 1 to 200 rolls before a seven-out.
• Ask the Wizard. See craps questions I've answered about:
• Simple Craps game. My simple Java craps game.

444 Craps System Youtube

External Links

• Las Vegas craps survey — The max odds bet allowed at each casino.