Random Roller Stragegy Testing

Setting and influencing the dice roll is just part of the picture. To beat the dice you have to know how to bet the dice. Whether you call it a "system," a "strategy," or just a way to play - this is the place to discuss it.

Moderators: 220Inside, DarthNater

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Random Roller Stragegy Testing

Post by sharkbyte » Mon Dec 26, 2011 11:50 am

I will be the first to admit that I am not fully disciplined when I am at the Craps table.

The reason is that with only 2 casinos (currently...Massachusetts will be building 3 in the coming few years!!!), there are rarely tables that do not have at least five players, and I just cannot seem to stand there and not play against every shooter.

Rather than beat myself up about this I have always worked at finding a strategy that can either grind out small wins, or at least allow me to tread water as the dice work around to me.

To this end, I have put together a group of strategies. Some designed for RRs, a couple others seem geared towards DIs, which means they fail miserably. I have generated 5000 random rolls, broken them into 100-roll (~2 hour) sessions, and have begun calculating the results. The table below shows the first 5 sessions. I also discuss each strategy at the end of this post.

I'm asking for 2 things, from this group. First, any RR strategies you have, and that you would like to see tested. Just describe and I will add them. Second, I would be interested in beneficial discussion(s) regarding the pros/cons of any of these.

Also, it is not necessary to point out that the best strategy is not to play RRs. :P

[tr][td][b][color=orange] Strategy [/color][/b][/td][td][b][color=orange] Results [/color][/b][/td][td] [tr][td][b] DP - 3-step Martingale - 1 bet only [/b][/td][td][b] $110.00 [/b][/td][td] [tr][td][b] DP - 10-5-5 3-bid - Do not replace bets [/b][/td][td][b] $15.00 [/b][/td][td] [tr][td][b] No 4/10 after 2 - No 5/9 after 3 [/b][/td][td][b] $320.00 [/b][/td][td] [tr][td] -----Double if knocked off [/td][td] $520.00 [/td][/tr] [tr][td][b] No 4/10 after 1 hit - No hedge [/b][/td][td][b] $180.00 [/b][/td][td] [tr][td] -----Double if knocked off [/td][td] $760.00 [/td][/tr] [tr][td][b] No 4/10 always working [/b][/td][td][b] $880.00 [/b][/td][td] [tr][td] -----Double if knocked off [/td][td] $2,840.00 [/td][/tr] [tr][td][b] Single Pass - place 2 inside [/b][/td][td][b] $(459.00) [/b][/td][td] [tr][td][b] $50 No 4/10 work, dbl on hit - $18 6/8 $15 5, $10 field - WOTCO [/b][/td][td][b] $2,135.00 [/b][/td][td] [tr][td] -----$18 6/8 $15 5 - WOTCO [/td][td] $(1,290.00) [/td][/tr] [tr][td] -----$10 field [/td][td] $(70.00) [/td][/tr] [tr][td] -----$50 No 4/10 work, dbl on hit [/td][td] $3,175.00 [/td][/tr] [tr][td][b] $50 No 4/10 work, dbl on hit - $18 6/8 $15 5, $10 field - no WOTCO [/b][/td][td][b] $2,462.00 [/b][/td][td] [tr][td] -----$18 6/8 $15 5 - no WOTCO [/td][td] $(1,008.00) [/td][/tr] [tr][td] -----$10 field [/td][td] $(70.00) [/td][/tr] [tr][td] -----$50 No 4/10 work, dbl on hit [/td][td] $3,175.00 [/td][/tr] [tr][td][b] Sure Win 1 [/b][/td][td][b] $(170.00) [/b][/td][td] [tr][td][b] Modified Dave's System [/b][/td][td][b] $845.00 [/b][/td][td] [tr][td][b] 1-1-2-3 Labouchere - Pass Line [/b][/td][td][b] $(775.00) [/b][/td][td] [tr][td][b] 1-1-2-3 Labouchere - DP [/b][/td][td][b] $910.00 [/b][/td][/tr]
Don't Pass - 3-step Martingale (10-20-30) This is a low-volatility (-40 to +100) grinder. Only your first 2 bets turn a profit. The third is to try and break even. Restart on series win, wait until next shooter on series loss.

Don't Pass - 10-5-5 3-bid series - Do not replace bets An even tougher grind, or more for treading water (-20 to +30).

$40 No 4/10 after 2 show, $30 No 5/9 after 3 show - flat bet replace if knocked down One that I've used in the past, and tested with relative success. Every session positive. I also tested this strategy but doubling-up on any hit, so each series at least broke even. ie 40-80-240. Can get expensive when having to reload for 3 or more hits.

$40 No 4/10 after single hit - flat bet reload Similar to above, lower trigger point, to bet, and drop the 5/9 lay. Again, also tested with double-up.

$40 No 4/10 - always working Again, as above. This has the potential for the highest volatility, particularly when looking at the double-up side. However, this is a strong position play assuming you can afford to reload against multiple hits. The casinos know this, and have structured this bet to make it almost unplayable to certain levels.

Single pass - Place 2 inside (JP recommendation for points of 6 (5 & 8) or 8 (9 & 6)) Does not hold up well against RRs. Posted a loss each session. Will likely not continue testing.

$50 no 4/10 working, double-up if hit; $10 field; $18 6/8 $15 5 regardless of point, WOTCO Someone posted this, on this board (Americraps maybe?). Was surprised to find this making money until I looked further and discovered the No 4/10 was carrying it. I have broken each piece out, to show the impact on this strategy.

$50 no 4/10 working, double-up if hit; $10 field; $18 6/8 $15 5 regardless of point, no WOTCO Same as above.

Sure Win 1 (pass line) Posted by SteveO. While not a big loser, I really dislike this because of keeping track of where you are in the betting series, and I've never been a real fan of parlaying a win, particularly after several losers in a row...While I'm going to drop this from the Right side, I will retest against the Dont's. It may show better there.

Dave's System (modified?) Posted by $5Bill. 1 loss (-135), a couple modest wins (+45), and a couple large (+345, +545). I say modified because it was not clear if naturals counted toward the 2 hits. In testing I chose not to count them.

1-1-2-3 Labouchere - Pass line Just not going to work with RRs

1-1-2-3 Labouchere - DP Nice, consistent wins (+110 to +250), but you have to be prepared for some big bets. High bet was 295. A couple in the low 100's. The series worked out each time, though.

So there you have it. Let's hear everyone's comments.

Golfer
Posts: 892
Joined: Thu Jul 28, 2011 10:12 pm

Re: Random Roller Stragegy Testing

Post by Golfer » Mon Dec 26, 2011 12:34 pm

For Dave's system he always replaces after a natural (7 & 11) loss on the comeout. I play a modified Dave, usually limiting myself to 1 dp and 2 dc. Replace once and take down odds layed after 2 hits. Never odds on 4 and 10. If bankroll is there and table is cold continous DC bets w/odds fills the rails nicely.

Good Luck
Golfer

User avatar
heavy
Site Admin
Posts: 10654
Joined: Mon Nov 22, 2010 11:46 am
Location: Texas
Contact:

Re: Random Roller Stragegy Testing

Post by heavy » Mon Dec 26, 2011 12:50 pm

Those Lay bets look pretty interesting. On the $50 no 4/10 Lay bets with the double up after hit - are you saying if you win the no 4/10 bet at $50 (winning $25) that your next bet is to lay $100? Just want to be sure I'm on the same page with you. Essentially it looks like you are power pressing the Lay bet on a win.

The strategy that Midgley came up with after a fair amount of computer testing was Pass with Odds, Place the six and eight, then continuous Come bet with Odds. Six and eight Place bets come down as they are replace by Come bets with Odds. You might try that. Say a $5 game with 5X odds. Place six and eight for $30 each. Continuous $5 Come Bets with $25 odds ($24 on five and nine).

Another strategy you might like is SIA's One Hit - Can't Miss play. Don't Pass for $12. Once the point is established Place the six and eight for $12 each. Let 'em ride until the cows come home.
"Get in, get up, and get gone."
- Heavy

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Mon Dec 26, 2011 1:51 pm

Heavy wrote:Those Lay bets look pretty interesting. On the $50 no 4/10 Lay bets with the double up after hit - are you saying if you win the no 4/10 bet at $50 (winning $25) that your next bet is to lay $100? Just want to be sure I'm on the same page with you. Essentially it looks like you are power pressing the Lay bet on a win.
Actually, just like the others, I'm doubling up when it gets knocked off. $50-100-300-900-2700... Always playing a hit to come even. Also, all P/DP bets were $5 unless otherwise noted.

I'll take a look at your other suggestions, though I'm not too sure about Midgley's. Obviously, it's a small sample size, but so far I'm only seeing about 2 decent "rolls" per session. And the Right side plays don't seem to have the juice to overcome the other rolls.

User avatar
heavy
Site Admin
Posts: 10654
Joined: Mon Nov 22, 2010 11:46 am
Location: Texas
Contact:

Re: Random Roller Stragegy Testing

Post by heavy » Mon Dec 26, 2011 3:48 pm

I'll take a look at your other suggestions, though I'm not too sure about Midgley's. Obviously, it's a small sample size, but so far I'm only seeing about 2 decent "rolls" per session. And the Right side plays don't seem to have the juice to overcome the other rolls.

In that case, we might want to consider a regression strategy. Maybe wait one toss after the point is established, then place $640 across - regress to $110 inside after one hit - same bet on second hit - press every other hit afterward. Something along those lines.
"Get in, get up, and get gone."
- Heavy

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Mon Dec 26, 2011 8:14 pm

More test results. First five sessions, only.
[tr][td][b][color=orange] Strategy [/color][/b][/td][td][b][color=orange] Results [/color][/b][/td][td] [tr][td] SIA $12 DP - $12 6/8 [/td][td] $(342.00) [/td][/tr] [tr][td] Heavy's Regression [/td][td] $(5,110.00) [/td][/tr] [tr][td][b] Sure Win 1 - DP [/b][/td][td][b] $(25.00) [/b][/td][/tr]

SIA's Not sure if there are any changes that could bring this into the black.

Heavy's Regression - $640 for 1, regress to $110 Got hit hard with point-number-7s. 6 in-a-row for session 2.

Sure Win #1 - DP Moving to the DP did improve the results. I think it still lost money due to betting series not fully accounting for previous losses. For example: a win @ step 7, with parlay, is $50. But you have lost $65 up to that point. I still don't like keeping track of such a long progression, but I still may work the numbers; even them out some; and see if I can't make this one profitable.

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Tue Dec 27, 2011 4:19 pm

Spent the day programming the 1-1-2-3 Labouchere to test automatically. I will eventually do this with all the promising systems.

By putting them in the system I can test them across many sessions, in an extremely short amount of time. For example, the 47 sessions, below - almost 5000 random tosses - takes less than 10 seconds.

A couple of notes on the results: I did not build in a loss limit. I let each session run to completion, as is. I will now go back and add a -50% loss limit, to end a session, and see the results.

Lowest point, for any session was -271 units (-$1805 for $5/unit). All sessions combined finished -321 units. (I had to edit this when I realized my mistake of counting my original bankroll into the total units.)

Once I get the others programmed in i will start testing at about 50K tosses, and see how it does.

Here are the test results:
[tr][td][b][color=orange] SessionID [/color][/b][/td][td][b][color=orange] StartUnits [/color][/b][/td][td][b][color=orange] EndUnits [/color][/b][/td][td][b][color=orange] LowPoint [/color][/b][/td][td] [tr][td] 1 [/td][td] 100 [/td][td] 136 [/td][td] 30 [/td][/tr] [tr][td] 2 [/td][td] 100 [/td][td] 124 [/td][td] 44 [/td][/tr] [tr][td] 3 [/td][td] 100 [/td][td] 125 [/td][td] -19 [/td][/tr] [tr][td] 4 [/td][td] 100 [/td][td] 124 [/td][td] 69 [/td][/tr] [tr][td] 5 [/td][td] 100 [/td][td] 131 [/td][td] 86 [/td][/tr] [tr][td] 6 [/td][td] 100 [/td][td] 105 [/td][td] 87 [/td][/tr] [tr][td] 7 [/td][td] 100 [/td][td] 131 [/td][td] 76 [/td][/tr] [tr][td] 8 [/td][td] 100 [/td][td] 131 [/td][td] 17 [/td][/tr] [tr][td] 9 [/td][td] 100 [/td][td] 124 [/td][td] 60 [/td][/tr] [tr][td] 10 [/td][td] 100 [/td][td] 117 [/td][td] 84 [/td][/tr] [tr][td] 11 [/td][td] 100 [/td][td] 110 [/td][td] 29 [/td][/tr] [tr][td] 12 [/td][td] 100 [/td][td] 79 [/td][td] 64 [/td][/tr] [tr][td] 13 [/td][td] 100 [/td][td] 94 [/td][td] 83 [/td][/tr] [tr][td] 14 [/td][td] 100 [/td][td] 136 [/td][td] 66 [/td][/tr] [tr][td] 15 [/td][td] 100 [/td][td] 93 [/td][td] -13 [/td][/tr] [tr][td] 16 [/td][td] 100 [/td][td] 125 [/td][td] 82 [/td][/tr] [tr][td] 17 [/td][td] 100 [/td][td] 100 [/td][td] 14 [/td][/tr] [tr][td] 18 [/td][td] 100 [/td][td] 48 [/td][td] 26 [/td][/tr] [tr][td] 19 [/td][td] 100 [/td][td] 145 [/td][td] 81 [/td][/tr] [tr][td] 20 [/td][td] 100 [/td][td] 23 [/td][td] -28 [/td][/tr] [tr][td] 21 [/td][td] 100 [/td][td] 138 [/td][td] 92 [/td][/tr] [tr][td] 22 [/td][td] 100 [/td][td] 145 [/td][td] 91 [/td][/tr] [tr][td] 23 [/td][td] 100 [/td][td] 114 [/td][td] 83 [/td][/tr] [tr][td] 24 [/td][td] 100 [/td][td] -12 [/td][td] -82 [/td][/tr] [tr][td] 25 [/td][td] 100 [/td][td] 120 [/td][td] 56 [/td][/tr] [tr][td] 26 [/td][td] 100 [/td][td] 114 [/td][td] 70 [/td][/tr] [tr][td] 27 [/td][td] 100 [/td][td] 108 [/td][td] 90 [/td][/tr] [tr][td] 28 [/td][td] 100 [/td][td] 69 [/td][td] 46 [/td][/tr] [tr][td] 29 [/td][td] 100 [/td][td] 143 [/td][td] 63 [/td][/tr] [tr][td] 30 [/td][td] 100 [/td][td] 118 [/td][td] 70 [/td][/tr] [tr][td] 31 [/td][td] 100 [/td][td] 124 [/td][td] 81 [/td][/tr] [tr][td] 32 [/td][td] 100 [/td][td] 37 [/td][td] -143 [/td][/tr] [tr][td] 33 [/td][td] 100 [/td][td] 125 [/td][td] 78 [/td][/tr] [tr][td] 34 [/td][td] 100 [/td][td] -198 [/td][td] -271 [/td][/tr] [tr][td] 35 [/td][td] 100 [/td][td] 125 [/td][td] 90 [/td][/tr] [tr][td] 36 [/td][td] 100 [/td][td] 127 [/td][td] 62 [/td][/tr] [tr][td] 37 [/td][td] 100 [/td][td] 120 [/td][td] 96 [/td][/tr] [tr][td] 38 [/td][td] 100 [/td][td] 55 [/td][td] -21 [/td][/tr] [tr][td] 39 [/td][td] 100 [/td][td] 120 [/td][td] 96 [/td][/tr] [tr][td] 40 [/td][td] 100 [/td][td] -118 [/td][td] -174 [/td][/tr] [tr][td] 41 [/td][td] 100 [/td][td] 104 [/td][td] -88 [/td][/tr] [tr][td] 42 [/td][td] 100 [/td][td] 143 [/td][td] 96 [/td][/tr] [tr][td] 43 [/td][td] 100 [/td][td] 63 [/td][td] 38 [/td][/tr] [tr][td] 44 [/td][td] 100 [/td][td] 24 [/td][td] -10 [/td][/tr] [tr][td] 45 [/td][td] 100 [/td][td] 86 [/td][td] 71 [/td][/tr] [tr][td] 46 [/td][td] 100 [/td][td] 60 [/td][td] 42 [/td][/tr] [tr][td] 47 [/td][td] 100 [/td][td] 124 [/td][td] 81 [/td][/tr]
Last edited by sharkbyte on Tue Dec 27, 2011 4:26 pm, edited 2 times in total.

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Tue Dec 27, 2011 4:20 pm

sharkbyte wrote:Sure Win #1 - DP Moving to the DP did improve the results. I think it still lost money due to betting series not fully accounting for previous losses. For example: a win @ step 7, with parlay, is $50. But you have lost $65 up to that point. I still don't like keeping track of such a long progression, but I still may work the numbers; even them out some; and see if I can't make this one profitable.
Altering this progression isn't going to work. The only way for the numbers to work is when the series becomes a Martingale, and we all know how well a true Martingale performs...

Texas Tosser

Re: Random Roller Stragegy Testing

Post by Texas Tosser » Tue Dec 27, 2011 4:32 pm

So you lost 321 units out of 4700. That's a 6.83% house edge. Sounds like you might as well just bet the field... :P

Tex

User avatar
heavy
Site Admin
Posts: 10654
Joined: Mon Nov 22, 2010 11:46 am
Location: Texas
Contact:

Re: Random Roller Stragegy Testing

Post by heavy » Tue Dec 27, 2011 4:46 pm

Well, I would never have lost $640 across more than three times back to back due to my "lose three in a row - time to go" rule. But it would still suck to get tagged that way. The big danger in regression style betting is not surviving long enough to get that regression in.
"Get in, get up, and get gone."
- Heavy

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Tue Dec 27, 2011 10:37 pm

Adding in the loss limit didn't improve the outcome, at any level from 50% to 90% of buy-in. So I'm on to the next strategy...

User avatar
heavy
Site Admin
Posts: 10654
Joined: Mon Nov 22, 2010 11:46 am
Location: Texas
Contact:

Re: Random Roller Stragegy Testing

Post by heavy » Wed Dec 28, 2011 2:29 am

Hmmm. Are you using a RNG or are you using the same numbers for all trials? Just curious. Also, I'm still trying to figure out how SIA's One Hit - Can't Miss strategy went that much upside down. Any insights?
"Get in, get up, and get gone."
- Heavy

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Wed Dec 28, 2011 7:37 am

Heavy wrote:Hmmm. Are you using a RNG or are you using the same numbers for all trials? Just curious. Also, I'm still trying to figure out how SIA's One Hit - Can't Miss strategy went that much upside down. Any insights?
I built an RNG and generated 5000 results. Currently, I am running all tests against those initial 5000. Once I have the tests automated I will bump the test results up to 50k.

I'm not too sure about SIA's play. Let's make sure I played it correctly: $12 DP, once a point is set $12 6 & $12 8. Play until there is a decision on the DP. Rinse and repeat.

I also tried a modified version where I was placing only a $6 6&8. It did a little better since a 7-out was a wash instead of losing $12. I didn't experiment with 1 hit and down, or any other options.

It took a big hit in session #3 with a lot of comeout 7s. Looking back, it only had one really bad session, so I will add that back in to my testing and see what we find.

Also remember, none of my by-hand testing included loss limits. I played 100 rolls to a seven-out. As I design these into the computer I will be adding a loss limit function, as well, to allow for testing at multiple loss-limit percentages.

Mad Professor
Posts: 1830
Joined: Fri Jul 29, 2011 12:15 pm

Re: Random Roller Stragegy Testing

Post by Mad Professor » Wed Dec 28, 2011 8:20 am

You could copy and paste those 5,000 rolls into WinCraps, and it will do all of those things for you.


MP

User avatar
heavy
Site Admin
Posts: 10654
Joined: Mon Nov 22, 2010 11:46 am
Location: Texas
Contact:

Re: Random Roller Stragegy Testing

Post by heavy » Wed Dec 28, 2011 10:20 am

Yeah, WinCraps also has a couple of different RNG's you can select from. It also allows you to "skew" the results if you wish to adjust the outcome to reflect some measure of dice control. Without a doubt, the best $20 craps testing program out there. For those of you who are unfamiliar with Steen's work - check out www.cloudcitysoftware.com.
"Get in, get up, and get gone."
- Heavy

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Sun Jan 01, 2012 12:57 am

So, as we usher in the new year, I have some results from testing against 50000 random rolls.

As expected, several strategies failed miserably when calculated over extended play. There were 4 that were close to even, 2 "mid" range that came in just negative, and another at a "high" range that still averaged a loss of less than 1 unit per session.

My goal has not been to necessarily make money, but to find a strategy that can be empoloyed while waiting for the dice to circle the table. Because not all of us have the luxury of finding too many empty tables to play at.
[tr][td][b][color=orange] System [/color][/b][/td][td][b][color=orange] Sessions [/color][/b][/td][td] Units [/color][/b][/td][td] [tr][td] 10-5-5 DP A [/td][td] --- [/td][td] --- [/td][/tr] [tr][td] 10-5-5 DP B [/td][td] --- [/td][td] --- [/td][/tr] [tr][td] 3 DP/DC [/td][td] 479 [/td][td] -72 [/td][/tr] [tr][td] 3-Step Martingale [/td][td] 479 [/td][td] -422 [/td][/tr] [tr][td] Dave's System [/td][td] 770 [/td][td] -314 [/td][/tr] [tr][td] Lay 4/10 1-hit, Reload Dbl [/td][td] 508 [/td][td] -234 [/td][/tr] [tr][td] Lay 4/10 2-hit, Reload Dbl [/td][td] 489 [/td][td] -199 [/td][/tr]
First thing, all testing (except Dave's) was done w/ 100 unit buy-in, and a 100% loss limit. Each session was a minimum of 98 rolls, but played until a true seven-out. If a system never reached a loss limit there were 479 possible sessions.

Dave's System was played as it was described: +40 units also ended a session. There were 403 sessions that reached the win limit. 93 where the complete buy-in was lost; 142 where at least 75% was lost (including total loss); and that jumps to 213 sessions if loss level is raised to 50% of buy-in. I may re-test with a 50% loss limit, and see if there is any measurable change.

10-5-5 DP Edit: Further testing, and code correction, has pushed this strategy into the negative.

Lay 4/10 1-hit & Lay 4/10 2-hit These plays are replacing the bet, when it is knocked off, but at double the original bet. So after the first bet level, you are playing for 0. Ie $40, 80, 240, 720... so the bet can push you past your loss limit pretty quick. It should be noted that both these bets become losers because of the vig.


So there you have it. Do with it what you will. If anyone has any other suggested plays, let me know and I can test them.

From here I'm going to expand my testing to 500K rolls, and perhaps test some slight variations on the plays.

Happy New Year, folks.
Last edited by sharkbyte on Mon Jan 02, 2012 8:46 pm, edited 1 time in total.

acpa
Posts: 594
Joined: Mon Sep 05, 2011 10:45 am

Re: Random Roller Stragegy Testing

Post by acpa » Sun Jan 01, 2012 1:16 am

Heavy wrote:Yeah, WinCraps also has a couple of different RNG's you can select from. It also allows you to "skew" the results if you wish to adjust the outcome to reflect some measure of dice control. Without a doubt, the best $20 craps testing program out there. For those of you who are unfamiliar with Steen's work - check out http://www.cloudcitysoftware.com.
it's amazing what that WinCraps can do.

however, I have never found a bet type that is a winner on random rollers.

Noah

Jonah

Re: Random Roller Stragegy Testing

Post by Jonah » Sun Jan 01, 2012 2:58 pm

sharkbyte wrote:[snip]

10-5-5 DP The difference between these is that the "A" system includes naturals as part of the betting series and "B" treats them separately. But it is a simple system, that looks like you could stay pretty close to even (long-term) by employing this against random rollers.
[snip]
Could not find an explanation of your 10-5-5 DP.

Can you explain it for us Homer Simpson's?

Thanks!

sharkbyte
Posts: 165
Joined: Wed Dec 21, 2011 1:40 pm

Re: Random Roller Stragegy Testing

Post by sharkbyte » Sun Jan 01, 2012 4:54 pm

Jonah wrote:Could not find an explanation of your 10-5-5 DP.
Jonah:

Edit: I found an error in the code, pushing this strategy into the negative. Average loss is ~1.5 units per session.

al_falcons
Posts: 183
Joined: Fri Jul 22, 2011 2:48 pm

Re: Random Roller Stragegy Testing

Post by al_falcons » Mon Jan 02, 2012 9:47 pm

acpa wrote:
Heavy wrote:Yeah, WinCraps also has a couple of different RNG's you can select from. It also allows you to "skew" the results if you wish to adjust the outcome to reflect some measure of dice control. Without a doubt, the best $20 craps testing program out there. For those of you who are unfamiliar with Steen's work - check out http://www.cloudcitysoftware.com.
it's amazing what that WinCraps can do.

however, I have never found a bet type that is a winner on random rollers.

Noah
Noah, I would beg to differ with you on never finding a betting method that is a winner on random rollers in WinCraps. It depends on how you use it. About 6 years ago when I started to play craps and bought WinCraps I created 100 "sessions" of random rolls and then ran them through many betting methods. Each session had between 600-900 rolls. At that time I had just finished reading a few of John Patrick's books on craps and betting methods, so I put win and loss limits into a lot of the betting methods. With some of these in place, I did find that several methods would win over my 100 random session.

I would agree with you if you said, that a created betting method was run for a simulated 100 hours of random rolls straight through, all of them would come up losers. That is called the house advantage, ;) However we don't play craps for 100 hours straight, so that isn't a real world simulation, breaking down into smaller sessions is more like it. You can win in the short run with some conservative betting methods on random rollers. I will have to look back at my stats, because I have not done any work on this since I began attempting to DI, but some of the best best methods I found were these:
$66 inside, one hit and down
$5 pass, $20 odds.
$5 don't pass, $10 odds.
Just for grins, I just ran the $66 inside method on random rolls and it held its own for about 6000 rolls before it started to lose serious money and then at about 8000 rolls, lost my theoretical $1000 bankroll.

Post Reply