Posted by odiousgambit
Feb 13, 2011

Variant Field Bets Explored

The lack of symmetry on the Field bet, with '5' losing and '9' winning, does bug me. Of course, the come-out is not symmetrical either, as the '11' has the same chance of coming out as '3' does, but one wins and the other loses. I've gotten used to that it seems, and never think about it.

But the equally likely '5' and '9' having different field payoffs, grrrr. Probably it's 'just me', but I had to toy around with this; after all, different casinos have different field bets. And "how things came about" is of interest to me.

Here's what I have so far, and to make a stir, if I owned a casino, I'd use and promote #1 at least some of the time. The excitement factor I think is pretty good!

The desired HE aside, at this forum I realize I will be up against plenty of boos, hisses, and catcalls [g]. And I too would really like something more simple than #3, not sure I can.

Wincraps should have these right, although I sometimes seem to run into a glitch with "bet results". At first I thought it was kind of weird that zero HE would sometimes be what I got, toying around, but now I think I get it.

    variant #1- house pays 3x on '2','3','11', & '12', while '5' and '9' loses [as well as the other inside numbers]. Zero HE. Because of the symmetry, but also excitement factor, I like this better for a zero HE promotion. I have to admit as casino owner I'd probably only have something like that on a low-max-bet table myself.

    variant #2 - house pays 3x on '2','11', &'12' but pays 2x on '3', while '5' and '9' etc. loses. This one is -5.56 HE. The "stingy casino" house edge with symmetry, except a lack of symmetry in what pays in multiples.

    variant #3 - house pays 4x on '12', 3x on '2' or '11' and 2x on '3' while 5 and 9 etc. loses. Has a -2.78 HE. The most standard HE, now with symmetry and same problem as #2.

I havent been able to concoct a field bet that has -2.78 HE without a lack of symmetry at least in the unsymmetrical multiples pay. I think the problem is that one 36th is .0278. Thus that is the most fine tuning you can pull off one way or the other?

My guess on how the field bet got to its various current formations: All field bets with '7' paying are rejected, and inside numbers in general are rejected. Then having the '5,6,7,8,9' inner numbers losing, with no bet paid in multiples, was considered at the beginning? Finding no one to bet such a rip-off, the '9' was chosen to pay I think, why the '9' over the '5' I know not. Indeed a truly symmetrical way to give better odds perhaps cannot be found. Paying double on 2 and 12 and then triple on 2 or 12 surely came about in an effort to out-compete another casino by offering better odds yet, as also paying on the '5' sends it to player advantage. It took me a while to realize the outside numbers are for fine tuning, as the inside number diddling is big whacks.


kenarman Feb 13, 2011

variant #1 is already on the table, although with different odds, in the horn bet

guido111 Feb 13, 2011

No need to guess on how the field came into being.

From John Scarne on Dice 1974.

Pages 103 and 104 have photos of old layouts.

The field bet started with the numbers 12,11,10,9,5,3,2 and paid even money.

changing the 5 for the 4 and adding the double 2 and 12 was just a way to adjust the HA and get more action against other casinos as you have mentioned..

I remember years ago in Miss. that I saw a field with a centerfield 5, now I know why.

7winner Feb 13, 2011

I have read in Scarne's book that the 2 and 12 paid 3 to 2 on bets over $1. The other numbers 11,10,9,5 and 3 paid even money.

He has some funny stories in the "Betting Systems" chapter.

The field has always been an action bet.

odiousgambit Feb 14, 2011

thanks for all the input

Posted by odiousgambit
Feb 02, 2011

Craps Question Was Eating At Me

This will strike many of you as a profoundly idiotic question, but it has been bugging me for some time. If taking free odds in Craps, you increase the variance to the point that the significance of winning or losing at the pass line [or don't pass] becomes smaller; at something like 100x odds the pass line bets are almost immaterial for a session. I also had noticed during a session that I seemed to win a pretty normal percentage of times when the bet was resolved by rolling 7, 11, or craps on the come-out, but that my luck varied wildly on whether I would make my points, the very time that I had my big bets up with the free odds.

I couldn’t figure out if somehow wincraps kept track automatically of how often a come-out got resolved immediately, or how to keep track of that using chipracks in wincraps.

But I thought it was possible that a fairly lengthy session might involve a situation where playing free odds meant a losing session could turn into a winning one, or vice versa. The question became, does betting free odds tend to emphasize winning or losing in a session? The intuitive answer would seem to be “yes” but I decided to use wincraps and do some trials.

So what we have below are the results for 20 sessions of 300 rolls, long sessions of approximately 3 hours in real time each. The *same set* of random numbers for each trial was duplicated to see how it turned out betting differently. It pretty much confirms that major winning or losing gets boosted, not reversed, with the free odds. In other words common intuition on the matter seems correct. If, however, a session was close to break-even, then the free odds seem to have an unpredictable influence.

Some sessions, like #4, turned out quite interesting.

PS: the betting is by one unit, with a bankroll of zero to start.
PS: if I get around to the other stats, it will be a chart below this chart

Session Number Pass Line no odds 1 Pass line with 5x4x3 odds Don't Pass no odds Don't Pass lay 5x4x3 odds
#1 -13 -58 +10 +55
#2 - 20 - 93 +17 +90
#3 +8 +52 -11 -55
#4 -1 -45 -1 +43
#5 -2 0 -2 -4
#6 -8 -14 +6 +12
#7 +9 +57 -12 -60
#8 -5 -5 +2 +2
#9 -1 -18 -1 +16
#10 -6 -51 +2 +49
#11 +6 +33 -10 -37
#12 -2 +8 +2 -8
#13 -10 -70 +10 +58
#14 -14 -56 +10 +52
#15 -3 -3 +1 +1
#16 +8 +72 -10 -74
#17 -5 -45 +3 +43
#18 +1 +52 -4 +49
#19 +9 +80 -12 -83
#20 +6 +24 -7 -25


teddys Feb 02, 2011

Not an idiotic question at all. I've often wondered about it myself. Even single odds will increase your variance to the point where the pass line bet really doesn't matter. The odds tend to emphasize losing, because you will have fewer points lost than won. However, variance goes both ways, and you also have the best chance of winning when you take odds.

Ayecarumba Feb 02, 2011

In reality, I don't think you would flat bet if you won with a 7 or 11 on the comeout multiple times, or made multiple passes. A progressive betting scheme would further increase your variance, but also increase your total win. Alternately, your could bust out sooner if your betting scheme does not include a "reset to base amount" following a loss, or string of losses.

i note that of the 20 passline trials, taking odds would "save" the player only twice (#5 and #12). In other words it turned a losing session into a breakeven, or plus. This is as expected, however, I wonder what the peak bankroll and max loss were for each session?

odiousgambit Feb 03, 2011

I believe max and min bankroll are preserved figures. If I add columns it might get too big, but I'll at least post that for you in comments if you check back.

I don't see a way to vary betting and benefit myself.

RaleighCraps Feb 03, 2011

I had been struggling with the same thoughts, which is how I came up with the big bet idea. Free odds works great when numbers are being rolled (or 7s if you are dark side), but they cause large pain if you are on the wrong side of the resolution.

You can see the number of instant resolves by having your autobet file only play 1 game of 300 rolls. Just end the Hyper Roll in the auto bet file after 300 rolls, plus resolve the final action by testing for a new comeout roll. If you do that, then the Summary page will show you the number of shooters, and the number of points made. It also shows the complete number of passes. so number of passes - number of points made = number of comeout winners. That data is not persistent though, so if you reset the table, you lose that info.

odiousgambit Feb 03, 2011

Raleighc., you are right, that has it all, including bankroll high and low. When I can get to it, I'll add these other stats, I've started the column already.

dm Feb 04, 2011

Session 4 - quite interesting? To everyone? I assumed it would be totally not interesting to me. I know ways to give money away and I don't need any more. But, by all means, keep posting this craps or crap or whatever. In general, advantage players are not interested in this game.

dm Feb 04, 2011

My post was uncalled for - sorry.

Posted by odiousgambit
Dec 23, 2010

Boardless Diceless Clue

Christmas week is a very slow time for us at work, yet we are expected to show up and fill the days somehow. Gambling would be frowned upon, so after our recent discussion, I decided a game of Clue could be played. So here is the idea and how it went.

Boardless Diceless Clue. Some sequence of who goes first and who "is to the left" of who is modified from where people normally sit. I modified a deck of playing cards by writing names, weapons, and rooms on various and distributing them; the empty deck package held the 3 key cards determining whodunit etc.

A simple map of the board is provided to each player along with a piece of paper with the suspects, weapons, and rooms listed so that notes can be made. The map shows how you move from one room to the next, providing some manner in which you can cross the board as well as move from adjacent rooms and secret passages. There is no need to decide which suspect you are playing, each player is just himself.

Players could pick the room they would start in. To move, each player gets one move in one turn; in that move you may walk out of a room or into a secret passage. In the next move you may enter a room or continue to the next place on the map without entering. Rules about blocking a room and all that were not in effect. It was considered good if the game went pretty fast, as of course certain interruptions occur even on a slow day.

I anticipated a problem with people remembering where they were on the board, so the player "to the left" must help the current player remember. It turned out everybody started helping everybody remember. Otherwise it was standard Clue rules.

It went very well! For the second game players decided they wanted everybody to start in the same room, so I thought that was OK, if unnecessary. As is the nature of the game, sometimes it was over fast, and with moving so easy, it was really fast. But that was probably a good thing. I'm wondering now how well Boardless Diceless Clue would go as a parlor game at a party and might try it.


benbakdoff Dec 23, 2010

I think they should have thanked you for a great year and sent you home With Pay!!

Nareed Dec 26, 2010

You should pitch it to the manufacturer as "Rapid Clue" ;)

Ayecarumba Jan 03, 2011

There actually is a card based version of "Clue". It is designed for 2-4 players. You don't move around a board, just ask other players if it was "insert two rooms/weapons/suspects here". Sort of like "Go Fish". Your invention sounds more fun.

odiousgambit Feb 03, 2011

stay tuned, I think I will have to make a chart below the first chart to have room

Posted by odiousgambit
Oct 30, 2010

The Girl with the Dragon Tattoo [again]

Making this movie review a blog post [was a forum post]. Now that I hear hollywood will be making a movie, and the books are getting more popular, I thought this might be a better place for it.

Recently saw "The Girl with the Dragon Tattoo". This has been one of the more popular foreign movies recently, you should be able to catch it if a theater near you tends to show foreign or art movies.

I highly recommend it with some qualifications.

A Swedish movie, it is very feminist, and as a matter of fact the author of the book, Stieg Larsson, was a Communist. His villains were Capitalists; in the movie, though, the villains are not just Capitalists but Nazi sympathizers. No kidding. And we are talking murderers, serial killers, to boot. Of course all this makes having them be the bad guys palatable even for anyone who might otherwise not like the idea of portraying businessmen as the guys wearing the black hats. For balance one minor bad character is a social worker.

You might think you have no feminist sympathies, but by the time you experience how rotten these men are you are rooting for their demise, believe me.

The movie is very brutal, and this applies not just to the creeps, but to the heroine as well. I want to focus on her to keep this review from being too long. She is a new type we are seeing a little more of these days, a woman with a ton of metal on her face perhaps, plenty of tattoos, and Goth or Punk themed clothing etc. There just was not such a character in the media of yesteryear; now we see them portrayed sometimes, and this certainly suggests we find them intriguing. I wound up looking up the character Abby Sciuto in the TV series NCIS to see if this character could have been based on the movie or vice versa. Sciuto is described as been distinguished by a "gothic style of dress and her interest in death and the supernatural" in a wikipedia article. I concluded that it was possible Larrson was influential in bringing this type of character forward, but I havent read his books so can't be sure. Quite coincidentally I just saw another movie, "My First Mister," a dvd I picked up cheap that featured a Goth/Punk type teenager. This was a fair movie, with flaws, but my point is not to review it too, but to point out that these characters are multiplying. And why not, plenty of these types do exist in the real world.

The Dragon Tattoo character, Lisbeth Salander, fits this profile; I don't exactly remember how much body piercing she has, but there is the Goth/Punk look, certainly tattoos, and with an emphasis for sure on being anti-social with a brutal if not sadistic side to her. I am also reminded of the character Felicia "Snoop" Pearson from the HBO series "The Wire". I promise you, even if you find this character type off-putting, that you will be totally won over to Salander and her cause . There is a distinct feminine appeal to her in spite of her efforts to have none. Truly fascinating, go see the movie if you get a chance.

links are all easily found in imdb dot com, and wikipedia.

Posted by odiousgambit
Sep 22, 2010

How Computer Games Cheat

This is not about Casino computers, which tower in honesty compared to general games on the home computer.

I am posting what I have posted before in other forums, new comments are in italics.

Others have made these observations, I like to consolidate them.



# A player learns to play without a using a rulebook, and no rulebook governs the play. Generally, you have to just take the plunge and see if you have the computer savvy to operate it.

# The Black Box Problem A. Just what exactly are the rules anyway? There will be no assurance anywhere that the most basic assumptions are accurate. As far as I know, you can't buy a computer game that simulates a simple game of solitaire in cards that will assure you that you are indeed playing with a deck of 52 cards. Not in the game when it's up, not on the box, not anywhere. You may assume so after playing it a while, but I have heard of some that are either harder or easier to win than using a real deck of cards.

# The Black Box Problem B. How it is determined that one side wins or loses a battle or campaign is anybody's guess. Common sense evaluation seemed to work sometimes, but for Boardgamers this is never going to be a satisfactory state of affairs. At some point you are in danger of finding that the computer has just declared itself the winner, unless you dial back the AI; then you are in danger of being bored by winning too easily.

# The Black Box Problem C. Once you increase the AI, you might be hoping that the vast computing power of your machine kicks in and starts to out-think you; like Spock playing the computer in chess on Star Trek, you'll be hoping that you'll just find a level of AI in your mightily-powered opponent that roughly matches you in pondering strategy and problems. Lo and behold, though, what you find is that the computer has simply increased the manpower of the opposition. There may be no sense whatsoever that the AI is actually getting better in its strategy. The Risk computer games seem to do this for sure

# The Black Box Problem D. After you begin to get a handle on what seems to be some rules in the Black Box, such as hey, you can't do this, you find that the computer opponent is not held to the same restriction. Often this is a matter of the player not being able to “see” various units for various reasons, but it becomes clear that the computer is not held to these same restrictions. The Modernized Risk Game, Risk II or whatever, just for the computer [that I have] with all sorts of variants to the original game, cheats like crazy in C and D problems.

# The Black Box Problem In General. The Computer simply cheats. This is manifested at various times and in various ways, and maddeningly there will be no acknowledgment of it. A player won't be able to ascertain what aspects of the cheating he can tolerate because what exactly he is up against can't be determined.

I have been told, "do not expect to see human-like AI in your lifetime".


CHESS COMPUTER GAMES CHEAT TOO [well, OK, in a tolerable manner perhaps]

I find for now it is telling that IBM has flatly refused to take Kasparov up on his very do-able challenge. Kasparov was finally beaten by their "Big Blue" of course, but K. claimed that they had used the well-known tactic of studying your opponent's games. He challenged them to go ahead and enter Big Blue in tournaments, then, to allow for the same counter-tactic, predicting (IIRC) it would not even be able to sustain Grandmaster ratings, much less win. Since this would be so clearly do-able, you have to figure the IBM team knew he was right, declining to do so though it has been something like 10-15 years since the challenge. For us humans, we have to cling to this to be able to claim that possibly some human somewhere is always going to be better than the machine, unless, well, something gets manipulated?

For your home game, computer chess has the black box as well, doing things you would never allow a human opponent to do. Would you let your opponent browse a huge databank of "opening book" while you are just on your own? Would you allow your opponent to re-arrange the board to his heart's content while pondering moves, while you are not even allowed to touch a single piece? IBM says " During the match with Kasparov, [Big Blue] averaged 126 million positions per second." Now that is what I call re-arranging the board!

Nonetheless Computer Chess probably shines as an example of a computer game that ultimately is tolerable in spite of how the AI goes about things.


Ibeatyouraces Dec 23, 2010

When I play minesweeper and am faced with a 50/50 shot at picking the square without the mine, it seems I hit the mine more than 75% of the time.