Pass-Odds only Scheme

Last night returned to this tried and true winner. This has been a consistent winner most(not all of course) sessions, and overall a good net winner.

Simply bet pass-odds only. My theory is that the longer the roll continues post-point, the more likelihood of making the point, and do not lose much on the PSO trends. Also it is much easier to win most session with making one good bet(pass-odds only) than the typical multiple place bets per roll.
Start every point sequence with $5 pass and the following odds: $5 odds if the point is 4,10, $10 odds if the point is 5,9, and $15 odds if the point is 6,8. Then as each inside number(5,6,8,9) is rolled post point, then up the odds bet one unit($5). Sometimes will get up to $50 odds or more, but not often, so it is better to play at joints with at least 10x odds and preferably 20x odds. $600 for day session bankroll. You will be surprised how well this works overall if you stick with it. Have played it for many hrs thusfar and several different versions, and this one is the best.

Of course it is not a 'tried and true winner'. However, since a large percentage of your bets are EV 1 bets (the odds bets) and your only other bet is a passline bet (small negative EV) you are probably making about the closest to even money bets you can make in a casino. But you must know your 'system' will not produce a profit over the long (infinity) run.

If I was going to do this, I'd do it on the Don't. The only reason I don't usually play the Don't is because I like playing numbers and Come Bets, and I don't like doing that if I'd going the wrong way. But if you are disciplined enough to not want to make any bet other than the line and odds, it should probably be Don't Pass.

Club: There is no meaningful difference between the donts or dos, so might as well bet with the shooter. Also, it is very difficult to win with multiple bets, the 7 is just too powerful wiping out all your bets many times with 0 or too few hits, been there and done that way too often. Much easier to win with one good bet, and hope for the best.

As to my comment as to tried and true winner, admittedly it does NOT win every session and the house edge is still there. But have been a net winner with it over many sessions and many hrs, so for me it is a net winner. Everytime I get bored with it and go to the usual multiple bet inside, across place bets or multiple come-odds betting, my bankroll suffers.

I agree, it's not a tried and true winner, but it gives you your best chance of being ahead after a long period of time. If you can stand the downtime, I'd say you've come upon the best craps strategy, bar none (except not playing or just playing someone's odds).

Quote: teddys

I agree, it's not a tried and true winner, but it gives you your best chance of being ahead after a long period of time. If you can stand the downtime, I'd say you've come upon the best craps strategy, bar none (except not playing or just playing someone's odds).

Best craps startegy is to own the craps table and the humongous building around it (most of the time) :P

Update: Have made some improvements with my pass-odds only scheme, latest 5 sessions are all winners in the $100 to $200 range, conservative with minimal downside per session thusfar.

*Start every pass point sequence with $5 pass and $15,16 odds. If the point is 6,8 then as every INSIDE number is rolled post point then up the odds bet one unit, $5, and stop at max odds of double the starting odds bet, meaning max odds bet is $30. If the point is 4,5,9,10, then the odds bet starts at $15,16 then up the odds bet one unit as every OUTSIDE number is rolled, stopping at max odds bet of $30.

As this has been working so well, now progressing to starting odds bet of $20, with max odds bet of $40 using this same scheme involving increasing the odds bet per point sequence as box numbers are rolled after the point is established. $600 day session bankroll. If this also continues winning over the next seveal sessions, will gradually progress the odds bet, with the next one being $25 odds to start the point progressing to max odds of $50. Any questions or comments, let me know..

Posts about "my system of play" often have NO data attached, no win goals, no time of play and no exact actual (empirical) data about winning and losing.
Sure systems can and do win in any trial or session or even a sequence of trials or sessions.
But with actual raw data missing, one can NOT run simulations to find out the actual results than can be seen from a system.

My questions would be(in order to run a simulation):
1) you never bet the pass line on the come out roll?

Most important:
2) what is your session win goal?

Very important:
3) IF no session win goal, what determines your length of play? A certain number of points established? A certain amount of action?

Remember:
It is difficult to program feelings or hunches, a hot table or a cold table needs to have "exact" terms defined.

I enjoy running simulations, but I am missing important parts of your "type of play" in order to program a bet file.
Good luck with your improvements with your "pass-odds only scheme"

Quote: dwm

My theory is that the longer the roll continues post-point, the more likelihood of making the point,

The above 'theory" is wrong.
I answered this question for a few craps friends recently and give my table below from a 1 million roll simulation.

The answer is the longer a hand in craps goes without a decision, it favors neither side.
(Please correct my if I am wrong by simulation or math only, but the results from my simulation, as long as my programming was correct, and appears to be so, show I am correct because of the simulation.)

The reason is because the probability of a win or a loss always remains the same, regardless of the number of rolls it takes to make a decision, the probability is always constant depending on the point number that has been established.

(In games like Roulette, this would not hold true. Some players bet on the last 14 spins, that in ,say, the next 6 spins one number will repeat. Well the probability of that can and does change. One could see, (in 14 spins)14 unique numbers, 13 unique numbers, 12 unique numbers, etc. so the probability of a repeat could be 14/38, or 13/38, or 12/38 etc.)

Example from table below:
When a point lost, 62.7631% of the time the frequency, or when it happened, was 3 rolls or less after a point had been established.
When a point won, 63.3236% of the time the frequency, or when it happened, was 3 rolls or less after a point had been established.
The longer the rolls go, the cumulative frequencies are almost the same at each step of the way.
(The reason the % between lose and win is not closer is there are more losing point than winning points in 1 million dice rolls.
When the same value of point wins equals the same value of point losses, the percentages are within .1%.)

What the table does show is most point wins (63.3236%) happen within 3 rolls of establishing a point.
That means 36.6764% (slight more than 1 in 3) of point wins happen after 4 rolls or more.

	117773	points lose				80166	points win
roll	absolute frequency	relative frequency	cumulative frequency	.	roll	absolute frequency	relative frequency	cumulative frequency
1	32934	27.9640%	27.9640%	.	1	22806	28.4485%	28.4485%
2	23828	20.2321%	48.1961%	.	2	16451	20.5212%	48.9696%
3	17156	14.5670%	62.7631%	.	3	11507	14.3540%	63.3236%
4	12325	10.4650%	73.2282%	.	4	8425	10.5094%	73.8330%
5	8613	7.3132%	80.5414%	.	5	6036	7.5294%	81.3624%
6	6511	5.5284%	86.0698%	.	6	4184	5.2192%	86.5816%
7	4355	3.6978%	89.7676%	.	7	3090	3.8545%	90.4361%
8	3327	2.8249%	92.5925%	.	8	2210	2.7568%	93.1929%
9	2405	2.0421%	94.6346%	.	9	1564	1.9510%	95.1438%
10	1717	1.4579%	96.0925%	.	10	1105	1.3784%	96.5222%
11	1291	1.0962%	97.1887%	.	11	761	0.9493%	97.4715%
12	944	0.8015%	97.9902%	.	12	534	0.6661%	98.1376%
13	646	0.5485%	98.5387%	.	13	450	0.5613%	98.6989%
14	519	0.4407%	98.9794%	.	14	291	0.3630%	99.0619%
15	324	0.2751%	99.2545%	.	15	222	0.2769%	99.3389%
16	243	0.2063%	99.4608%	.	16	137	0.1709%	99.5098%
17	167	0.1418%	99.6026%	.	17	95	0.1185%	99.6283%
18	150	0.1274%	99.7300%	.	18	87	0.1085%	99.7368%
19	95	0.000807	0.998107	.	19	56	0.0006986	0.9980665
20	61	0.000518	0.998624	.	20	40	0.0004990	0.9985655
21	45	0.000382	0.999007	.	21	27	0.0003368	0.9989023
22	40	0.000340	0.999346	.	22	21	0.0002620	0.9991642
23	14	0.000119	0.999465	.	23	19	0.0002370	0.9994012
24	12	0.000102	0.999567	.	24	7	0.0000873	0.9994886
25	19	0.000161	0.999728	.	25	17	0.0002121	0.9997006
26	7	0.000059	0.999788	.	26	3	0.0000374	0.9997380
27	6	0.000051	0.999839	.	27	6	0.0000748	0.9998129
28	6	0.000051	0.999890	.	28	3	0.0000374	0.9998503
29	5	0.000042	0.999932	.	29	3	0.0000374	0.9998877
30	0	0.000000	0.999932	.	30	3	0.0000374	0.9999252
31	4	0.000034	0.999966	.	31	2	0.0000249	0.9999501
32	2	0.000017	0.999983	.	32	0	0.0000000	0.9999501
33	1	0.000008	0.999992	.	33	0	0.0000000	0.9999501
34	1	0.0000085	1.000000	.	34	4	0.0000499	1.0000000

The table above is a relative/cumulative frequency of actual data.
One should know how to make and interpret a table like the one above.
Many sites can do a better job than I for this and one is here:
http://stattrek.com/AP-Statistics-1/Cumulative-Frequency-Plot.aspx?Tutorial=Stat

For those of you that like averages:
The averages from the table above are:
win
average rolls 3.5136
average rolls including point 4.5136
lose
3.5808 average rolls
4.5808 average rolls including point

Nope: I ALWAYS bet the pass($5) on every come out roll, then take triple odds after the point is established, must have not made myself clear on that issue. Then increasing my odds bet as box numbers are rolled per my odds scheme spelled out above. Win goal is at least $100 at the end of the shooter's hand. I agree that hunches or feelings are not a good way to play, that is why I play a pre-determined betting scheme that takes the guess work out of it.

Last night upped my starting bet to $20 odds with max odds bet of $40, meaning if the point was 6 or 8, then increased my odds bet one unit($5)as each inside number is rolled. If the point was an outside number, then increased the odds bet one unit as each outside number is rolled. Then start over from the beginning on the next point sequence. Ended up winning $70 in a marathon 8 hr session, decided to quit before the $100 win goal. I know that after now 6 consecutive wins with this odds scheme for a overall net win of about $800, 0 session losses, that a loss is due. Last night was down about $450 at the worse playing with a $600 session bankroll, so was no doubt lucky to pull out a minimum win. Probably should have a $700 bankroll for the $20 to $40 odds spread to give a better cushion for the downdrafts. BUT this has been a very consistent winner thusfar which is difficult in this craps game as you know. Agree with your conclusion per the math as to the probability of the point being made as the roll continues, maybe just lucky thusfar, but I better ride my "theory" as long as it wins and often save considerable $ on those all too frequent PSO trends compared to the level odds player..

thank you for your update.
I am in Tunic Ms. currently, and being from Toledo Spain,(I know that does not sound exciting at all coming from a town named Toledo!) this area by the Mississippi River is so different from anything I have ever seen in person.I'm taking in the Crapless Craps games, since almost every casino has at least one table and at night they are very busy tables. I play roulette, but have friends that play craps a lot.

Your "betting system" can do well since you are increasing your bets on a 0% HA odds bet. As from other posts it is nice to see simulations since they can give you so much more information than just winning and losing probabilities.
When I return home, I will share with you some simulation results.
Good luck to you!

Nope: I call it a betting "scheme", seems more appropriate as afterall we are talking here about gambling and betting on those dice that seem to have a mind of their own. Random game but there are often trends. Will look forward to your simulation results.

"What the table does show is most point wins (63.3236%) happen within 3 rolls of establishing a point.
That means 36.6764% (slight more than 1 in 3) of point wins happen after 4 rolls or more."

Question for Nope27, or anyone else.

It seems that based on this calculation that a good system might be to start with a doey-don't and take odds on the pass for the first three rolls and then remove them and lay odds until a point or seven out is made????

Everything I've read about playing for comps discusses Blackjack or Video Poker. But this seems like it could be a promising method, and more fun than sitting at a VP machine all day.
What kind of comps have you been getting at this level of play?

Dude: Comps are still decent, I get free meals and decent coupons of about $100 per week total involving several different casinos, about the same as when I was playing inside and across betting schemes.

Quote: wrongway

"What the table does show is most point wins (63.3236%) happen within 3 rolls of establishing a point.
That means 36.6764% (slight more than 1 in 3) of point wins happen after 4 rolls or more."

Question for Nope27, or anyone else.

It seems that based on this calculation that a good system might be to start with a doey-don't and take odds on the pass for the first three rolls and then remove them and lay odds until a point or seven out is made????

One would also be "giving up" so to speak the 36% of point winners in that happen after the 3rd roll. At least the system of play would be based off of math and that would be as far as I can see as being good.


I have run a few, not many, simulations to see results of something close to what you propose. But then just a few results is not a good thing to make conclusions with.

What appears to work best (from only my observations) is waiting (and your wait may be a long one) until the actual number of point wins or losses is 2.5 standard deviations or more from expectation, then start betting that it will not stay 2.5 SDs or more in the next coming trials(at least more than 1).
This means you must keep accurate running stats, a hard thing to do with pencil and paper in a casino.

Even this "system" does not succeed 100% of the time and would bore most to tears and retirement.

Look at this way. The odds that the point is 4 or 10 is 6/24. The odds that the point is 5 or 9 is 8/24. The odds that the point is 6 or 8 is 10/24.

If the point is 4 or 10, the odds it will be resolved after n rolls is: 1-(1-(3+6)/36)^n. After 1 roll, it's .25, after 3 rolls, it's .578125.

If the point is 5 or 9, the odds it will be resolved after n rolls is 1-(1-(4+6)/36)^n. After 1 roll, it's .27888, after 3 rolls it's .623285.

And it the point is 6 or 8, the odds it will be resolved after n rolls is 1-(1-(5+6)/36)^n. After 1 roll it's .305556 and after 3 rolls it's .665102.

Put it all together:

The odds of resolving any point after n rolls is:

6/24*(1-(27/36)^n)+8/24*(1-(26/36)^n)+10/24*(1-(25/36)^n).

So for example, after three rolls after a point is established, the odds that the point will be resolved, one way or another is .629419. There is no simulation required.

Taking your odds bet on after 3 rolls and then putting it to the Dark Side on the 4th is meaningless if you believe (as any mathematician should) that past results do not affect future results in the roll of the dice. All the Doey Don't does for you is assure the loss when the 12 is thrown on the comeout. After that, you have the option of guessing pass or don't pass. Even if after 20 rolls the point is still not made, moving it from the pass to the don't has no effect on whether the next roll is going to be Big Red or the point, nor does the probability or cumulative probability change on that roll, because the past doesn't matter. You're just guessing.

Quote: boymimbo

Put it all together:

The odds of resolving any point after n rolls is:

6/24*(1-(27/36)^n)+8/24*(1-(26/36)^n)+10/24*(1-(25/36)^n).

So for example, after three rolls after a point is established, the odds that the point will be resolved, one way or another is .629419. There is no simulation required.

Very nice to see the math. Thank you boymimbo.

I had a bet file already in WinCraps and had run a simulation for a few friends of mine. Sometimes simulations drive home the point better to those of us with less math skills than others like yourself.

update: it only takes a minute in excel to create a nice table from boymimbo's formula.
sure nice to know how to do math well.thanks again!

So you math wizards, what would be the better way of increasing or decreasing your odds bet after the point is established, if you had to make a choice based on cumulative probability? It would appear to me that the math says higher odds for the first 3 rolls, and lower odds after that, per point sequence.

Quote: dwm

So you math wizards, what would be the better way of increasing or decreasing your odds bet after the point is established, if you had to make a choice based on cumulative probability? It would appear to me that the math says higher odds for the first 3 rolls, and lower odds after that, per point sequence.

Doesn't matter. The odds are the same every roll after the point is established, so bet whatever you feel like on the odds (you're already locked into the line contract bet). There is no "cumulative probability" unless you believe some of these roulette threads.

If it makes you feel better to lower or raise your odds as the roll continues, by all means do so. Just know that the probability of rolling a seven or the point will not increase or decrease ever. Whatever mental gymnastics you use to keep your bets limited to the line+odds is a great "system."

Sorry, I don't mean to sound condescending in this post, but this is a concept that needs to be re-emphasized often.

Quote: teddys

Doesn't matter. The odds are the same every roll after the point is established, so bet whatever you feel like on the odds (you're already locked into the line contract bet). There is no "cumulative probability" unless you believe some of these roulette threads.

Im confused about this cumulative probability. Thought I understood the table on page one.

then what is this:
http://stattrek.com/Help/Glossary.aspx?Target=Cumulative_probability
A cumulative probability refers to the probability that the value of a random variable falls within a specified range. Frequently, cumulative probabilities refer to the probability that a random variable is less than or equal to a specified value.

or here:http://stattrek.com/AP-Statistics-1/Cumulative-Frequency-Plot.aspx?Tutorial=Stat

maybe I need to re-read when Im not tired.

This site has some great math members and posts, miplet, MathExtremist and boymimbo just to name a few. great work.

But this site also has some crazy posts. The "hoax about 1.414%" is a classic thread. Ive never laughed so hard while I was learning something.

What I mean by cumulative probability is a series of decision results. Such as looking at this from the start per series in round numbers an example is there is about a 40% chance of making a point, but only a 20% chance of making consecutive points.

I have been increasing my odds bet after box numbers are rolled per point sequence with several different schemes thru the last several years with an overall minimal average win but still a net win. I did have a session loss last session, hardly no points were being made so doubt if any pass-odds scheme would have worked, just one of those bad days that happens with gambling. Still in the positive after 7 day sessions from the start of this thread for a minimal net $200 overall win.

I do think the best chance is betting the pass-odds only, and believe me have tried many a betting scheme. Do know that really no chance with any multiple bet such as inside, across betting schemes, the 7 is just too powerful knocking down all those bets too often with 0 or too few hits, much easier to recover from losing just one good bet, and no bet is better than the pass-odds bet. Now how to maximize the odds bet results, that is the big question???

Quote: dwm

What I mean by cumulative probability is a series of decision results. Such as looking at this from the start per series in round numbers an example is there is about a 40% chance of making a point, but only a 20% chance of making consecutive points.

I have been increasing my odds bet after box numbers are rolled per point sequence with several different schemes thru the last several years with an overall minimal average win but still a net win. I did have a session loss last session, hardly no points were being made so doubt if any pass-odds scheme would have worked, just one of those bad days that happens with gambling. Still in the positive after 7 day sessions from the start of this thread for a minimal net $200 overall win.

I do think the best chance is betting the pass-odds only, and believe me have tried many a betting scheme. Do know that really no chance with any multiple bet such as inside, across betting schemes, the 7 is just too powerful knocking down all those bets too often with 0 or too few hits, much easier to recover from losing just one good bet, and no bet is better than the pass-odds bet. Now how to maximize the odds bet results, that is the big question???

OK. that now makes more sense. betting for a series of outcomes instead of just one outcome. That is where the cumulative probability and binomial distribution comes into play.

How to maximize the odds bet according to how a bet wins.Interesting thought.

now that would take some real good math or simulations to solve that one.and more than likely it would show there is no better way to maximize the odds bet than having a big bankroll and always taking max odds, since that lowers the house edge the most.

Quote: dwm

What I mean by cumulative probability is a series of decision results. Such as looking at this from the start per series in round numbers an example is there is about a 40% chance of making a point, but only a 20% chance of making consecutive points.

Using cumulative probability as a guide to betting involves a basic misunderstanding. I suggest you read my blog on the subject:

https://wizardofvegas.com/member/goatcabin/blog/#post16

Quote: dvm

I have been increasing my odds bet after box numbers are rolled per point sequence with several different schemes thru the last several years with an overall minimal average win but still a net win. I did have a session loss last session, hardly no points were being made so doubt if any pass-odds scheme would have worked, just one of those bad days that happens with gambling. Still in the positive after 7 day sessions from the start of this thread for a minimal net $200 overall win.

I do think the best chance is betting the pass-odds only, and believe me have tried many a betting scheme. Do know that really no chance with any multiple bet such as inside, across betting schemes, the 7 is just too powerful knocking down all those bets too often with 0 or too few hits, much easier to recover from losing just one good bet, and no bet is better than the pass-odds bet. Now how to maximize the odds bet results, that is the big question???

Although the basis for increasing your odds bets is fallacious, the fact remains that you are betting the largest portion of your money on bets that carry no expected loss, which means you have a very small "handicap" to overcome to win. High variance combined with low expected loss yields the maximum probability of winning; however, variance works both ways, increasing losses when you have bad luck.
Cheers,
Alan Shank
Woodland, CA