forced martingale as a game mechanic?
January 29th, 2014 at 9:59:56 AM
permalink
A couple of days ago I wrote about how a friend lost a bit of money in shady carnival game. It was similar to a game other members pointed out called Razzle, and had what I thought was an interesting mechanic of forcing bet doubling if you draw certain balls, along with an increase in the big prize. I was thinking, even as he was playing, that it was a nice game design. Obviously, at this point, I think it was likely rigged(beyond the math), but I believe it would've been profitable even if it hadn't been. Is there a reason casino's wouldn't want to have a game using a forced martingale? Off the top of my head, possibilities include:
1) association with the shady game? (But are razzle-like games well known among the public?)
2) tough to get a compelling run going if played honestly?
3) not wanting to put players in an uncomfortable position or force players out?
4) simple bias against persistence features?
I think I can soften 2 and 4 with some thoughts I have on design. I'm most concerned about casino's simply not wanting to deal with #3.
thoughts?
1) association with the shady game? (But are razzle-like games well known among the public?)
2) tough to get a compelling run going if played honestly?
3) not wanting to put players in an uncomfortable position or force players out?
4) simple bias against persistence features?
I think I can soften 2 and 4 with some thoughts I have on design. I'm most concerned about casino's simply not wanting to deal with #3.
thoughts?
January 29th, 2014 at 10:28:07 AM
permalink
A fundamental tenet of current casino games is that the player always has the ability to stop betting, even if it's not in his or her best interests from an EV standpoint. I don't believe regulators would allow a game that requires an additional bet after a game has started, as opposed to a larger bet up front that can be partly removed (e.g., blackjack surrender, Let It Ride) or forfeited (e.g., most SHFL games). Plenty of games allow optional increases in bets, but I'm not aware of any that require an increased wager.
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice."
-- Girolamo Cardano, 1563
January 29th, 2014 at 11:10:34 AM
permalink
Hmm. I was thinking that one option might be allowing an partial equity cashout at any time. Obviously, this takes away some of the punch from the original design, but the momentum and excitement of reaching higher levels and playing for more would, hopefully, still have value, even though it woudn't be required. So the persistence feature becomes more persistence-lite?
I recently listened to a dancer podcast where it sounded like casino's didn't like persistence features but that there wasn't anything categorically excluding them. It sounded like the problem was that any equity in the machine had to be played out, and that would be awkward if someone walked away and left a partial game because the game couldn't be reset without losing the quity. But I could be way off.
Maybe there's and associated problem that by allowing people to cash out at any time that can avoid -ev runs in a game that balance things out? Which makes design hard?
I guess I'm not actually sure how that'd work... If the persistent game state were Level+Points... then, you have to map the cashouts, on those 2 parameters, so that EV is constant?
I recently listened to a dancer podcast where it sounded like casino's didn't like persistence features but that there wasn't anything categorically excluding them. It sounded like the problem was that any equity in the machine had to be played out, and that would be awkward if someone walked away and left a partial game because the game couldn't be reset without losing the quity. But I could be way off.
Maybe there's and associated problem that by allowing people to cash out at any time that can avoid -ev runs in a game that balance things out? Which makes design hard?
I guess I'm not actually sure how that'd work... If the persistent game state were Level+Points... then, you have to map the cashouts, on those 2 parameters, so that EV is constant?
January 30th, 2014 at 9:05:21 PM
permalink
Quote: MathExtremistA fundamental tenet of current casino games is that the player always has the ability to stop betting, even if it's not in his or her best interests from an EV standpoint. I don't believe regulators would allow a game that requires an additional bet after a game has started, as opposed to a larger bet up front that can be partly removed (e.g., blackjack surrender, Let It Ride) or forfeited (e.g., most SHFL games). Plenty of games allow optional increases in bets, but I'm not aware of any that require an increased wager.
Contract bets in Craps can't be removed, and in All-In WPT you must wager 3 times your ante bet to stay in the game..
February 5th, 2014 at 3:11:16 PM
permalink
Put it this way - in a Martingale strategy, most people either make a small amount of money or lose it all. Casinos work a lot better when people either lose a small amount of money or win a lot. This is why raffles, lotteries, etc. work so well - throw in a few bucks for a small chance at a crap-ton of money.
Put it this way: I propose to you two different games. In Game A, you buy a ticket for $10 and compete with, say, 1,023 other people. One person wins $10,000 and the others lose their $10, just for a smallish house edge of 2.34%, which is well within the acceptable range for a typical gambler. You're probably going to lose, but it's just $10, no big whoop. However, the alternative is the "dream scenario" - walk away with an extra ten grand.
Now let's flip it: In Game B, I hand you $10, but you hand me a check for $10,000, and I do the same with 975 other people for a similar house edge. In this game, however, out of 976 checks, I randomly burn 975 of them and cash the other, so you're most likely going to win $10, which is hardly anything to write home about, but it carries the "nightmare scenario" of losing $10,000.
Most people would rather play the first game. This explains why poker-style carnival games work - you can root for that big hand but will more than likely lose a smallish amount. So a game with Martingale as an element probably wouldn't work too well. Besides, if a gambler has the expectation of losing that small amount, it's OK if they do and fantastic if they win big. In a Martingale-style game, they have the opposite expectation - a small win. So if they win small, it's no big deal; it's expected. But if they lose catastrophically? Oh damn. I wouldn't want to be the dealer in the game.
Put it this way: I propose to you two different games. In Game A, you buy a ticket for $10 and compete with, say, 1,023 other people. One person wins $10,000 and the others lose their $10, just for a smallish house edge of 2.34%, which is well within the acceptable range for a typical gambler. You're probably going to lose, but it's just $10, no big whoop. However, the alternative is the "dream scenario" - walk away with an extra ten grand.
Now let's flip it: In Game B, I hand you $10, but you hand me a check for $10,000, and I do the same with 975 other people for a similar house edge. In this game, however, out of 976 checks, I randomly burn 975 of them and cash the other, so you're most likely going to win $10, which is hardly anything to write home about, but it carries the "nightmare scenario" of losing $10,000.
Most people would rather play the first game. This explains why poker-style carnival games work - you can root for that big hand but will more than likely lose a smallish amount. So a game with Martingale as an element probably wouldn't work too well. Besides, if a gambler has the expectation of losing that small amount, it's OK if they do and fantastic if they win big. In a Martingale-style game, they have the opposite expectation - a small win. So if they win small, it's no big deal; it's expected. But if they lose catastrophically? Oh damn. I wouldn't want to be the dealer in the game.
February 5th, 2014 at 3:28:28 PM
permalink
"Matingale" might not have been quite right. Both the bet and prize escalate in this game as they do in the Razzle game it seems to be loosely based on.
I went ahead and coded a demo for a design I modeled on the previously discussed game, and I was pleasantly surprised, as I wasn't particularly optimistic to start with. I think the balance between risk/return and escalation is pretty good. Unfortunately, I'm not sure I can afford to pursue it, starting with hiring a lawyer for another provisional patent again so soon... so it might sit on the shelf for a while.
I went ahead and coded a demo for a design I modeled on the previously discussed game, and I was pleasantly surprised, as I wasn't particularly optimistic to start with. I think the balance between risk/return and escalation is pretty good. Unfortunately, I'm not sure I can afford to pursue it, starting with hiring a lawyer for another provisional patent again so soon... so it might sit on the shelf for a while.
February 5th, 2014 at 4:37:03 PM
permalink
Quote: MathExtremistA fundamental tenet of current casino games is that the player always has the ability to stop betting, even if it's not in his or her best interests from an EV standpoint. I don't believe regulators would allow a game that requires an additional bet after a game has started, as opposed to a larger bet up front that can be partly removed (e.g., blackjack surrender, Let It Ride) or forfeited (e.g., most SHFL games). Plenty of games allow optional increases in bets, but I'm not aware of any that require an increased wager.
https://wizardofodds.com/games/ultimate-three-card-poker/
Ultimate three card poker.
Your play option is to raise or fold.
February 6th, 2014 at 10:39:22 AM
permalink
Quote: Dalex64https://wizardofodds.com/games/ultimate-three-card-poker/
Ultimate three card poker.
Your play option is to raise or fold.
Bolded text is my point. There are no games, to my knowledge, that require a raise after the game has started without the option to fold. As I mentioned above, it's well known that basically all SHFL games allow the player to forfeit (fold) their hand, but none require you to put in more money without another option.
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice."
-- Girolamo Cardano, 1563
February 6th, 2014 at 10:59:26 AM
permalink
Quote: MathExtremistQuote: Dalex64https://wizardofodds.com/games/ultimate-three-card-poker/
Ultimate three card poker.
Your play option is to raise or fold.
Bolded text is my point. There are no games, to my knowledge, that require a raise after the game has started without the option to fold. As I mentioned above, it's well known that basically all SHFL games allow the player to forfeit (fold) their hand, but none require you to put in more money without another option.
I don't think that that was ever suggested. The suggestion seems to be something where, after not winning, you can bet more (say, double your last bet) for a chance at a disproportionately larger prize, or quit, forfeiting the pot that you have built up.
Games with a similar mechanic (minus the requirement that the bet increases) already exist. Eg, versions of wheel poker where you don't get any cash for your wheel slices if you quit. Or ultimate X poker (to a lesser degree since it only carries over for one hand). Or slot machines with locking wilds that carry over between spins. Or, you could argue, any progressive jackpot VP or slot game.
February 6th, 2014 at 11:08:57 AM
permalink
Quote: AxiomOfChoiceQuote: MathExtremistQuote: Dalex64https://wizardofodds.com/games/ultimate-three-card-poker/
Ultimate three card poker.
Your play option is to raise or fold.
Bolded text is my point. There are no games, to my knowledge, that require a raise after the game has started without the option to fold. As I mentioned above, it's well known that basically all SHFL games allow the player to forfeit (fold) their hand, but none require you to put in more money without another option.
I don't think that that was ever suggested. The suggestion seems to be something where, after not winning, you can bet more (say, double your last bet) for a chance at a disproportionately larger prize, or quit, forfeiting the pot that you have built up.
Perhaps I misread the OP then; what you're describing isn't much different than a kill pot in poker, where a 10/20 game suddenly turns into a 20/40 game.
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice."
-- Girolamo Cardano, 1563
February 6th, 2014 at 11:19:47 AM
permalink
I may have worded it badly or not yet have had a clear idea of what I meant.
The idea, as it stands, is that in the course of the game, bets and stakes are doubled if you pull certain random #'s. So it's forced, but not really, since you can reset the game to the initial stakes whenever you want (or leave), without being forced to put more money into the round.
The idea, as it stands, is that in the course of the game, bets and stakes are doubled if you pull certain random #'s. So it's forced, but not really, since you can reset the game to the initial stakes whenever you want (or leave), without being forced to put more money into the round.
February 6th, 2014 at 11:36:59 AM
permalink
Ok, so, imagine a simple 1-player-vs-the-house dice game, played with a single 6-sided die. You make a bet, and the dealer adds in 1.2x your bet, to form the pot.
You roll the die. On a 1, 2, or 3, you lose. On a 5 or 6, you win.
On a 4, the pot stays out there. You may either quit and forfeit the pot, or you may add double your last bet to the pot (which the dealer will also match 1.2x) and play again.
There is obviously a house edge here, since, even if you never quit, you still lose 60% of the time, and all your bets pay 6:5 instead of the fair price of 3:2.
However, once you have made the first bet, any time a 4 is rolled, it is +EV to keep playing, since you are getting more than 2-1 on your bet (since everything in the pot is dead money if you quit) and you have a 1/3 chance of winning on the next roll.
Eg:
You bet $10. The dealer adds $12. Pot = $22.
You roll a 4. In order to keep going, you must bet another $20. Dealer will add $24. Pot will be $66. This is +EV!
Roll another 4. In order to keep going, you must bet another $40. Dealer will add $48. Pot will be $154. This is hugely +EV!
And it keeps going. Every time you roll a 4 before the bet is resolved, the EV of continuing rises drastically. The entire game is still -EV though.
I think that this is the type of concept that the OP was referring to (not that this is a great implementation of it)
You roll the die. On a 1, 2, or 3, you lose. On a 5 or 6, you win.
On a 4, the pot stays out there. You may either quit and forfeit the pot, or you may add double your last bet to the pot (which the dealer will also match 1.2x) and play again.
There is obviously a house edge here, since, even if you never quit, you still lose 60% of the time, and all your bets pay 6:5 instead of the fair price of 3:2.
However, once you have made the first bet, any time a 4 is rolled, it is +EV to keep playing, since you are getting more than 2-1 on your bet (since everything in the pot is dead money if you quit) and you have a 1/3 chance of winning on the next roll.
Eg:
You bet $10. The dealer adds $12. Pot = $22.
You roll a 4. In order to keep going, you must bet another $20. Dealer will add $24. Pot will be $66. This is +EV!
Roll another 4. In order to keep going, you must bet another $40. Dealer will add $48. Pot will be $154. This is hugely +EV!
And it keeps going. Every time you roll a 4 before the bet is resolved, the EV of continuing rises drastically. The entire game is still -EV though.
I think that this is the type of concept that the OP was referring to (not that this is a great implementation of it)
February 6th, 2014 at 11:50:36 AM
permalink
Yes, to start with, I was thinking of a separate pot along with a point count as you progressed towards a BigPrize, but I didn't have a clear idea of how the 2 related to each other. Eventually, I kept the points and dropped the idea of having a separate pot. So, now, when you don't have to reset the level and points if you want to realize gains that are in the pot.
You do get to points where you essentially have positive equity due to the points, but no way to cash them out. For some reason, at first, I thought this was something I wanted to avoid, but I decided it didn't matter.
You do get to points where you essentially have positive equity due to the points, but no way to cash them out. For some reason, at first, I thought this was something I wanted to avoid, but I decided it didn't matter.
February 13th, 2014 at 3:12:11 PM
permalink
In the spirit of an idea not being worth that much, I'm abandoning the carny game idea. Any takers?
I feel like it may value, but I'm busy and I think it's very unlikely I will follow through at this point. Finding out that IGT is less approachable than previously thought makes me less eager to follow through as well. I am not abandoning the demo or code, just the idea. You're welcome to run with it. Presumably, a specific board/tile distribution could still be patented, or some other tile type added. If you want to use the demo (which would have to be adjusted once the math is ironed out), contact me.
http://frozen-gorge-8975.herokuapp.com/jtswebapps/liquidate
I'm still somewhat new to heroku. The javascript is not delivered under some circumstances, though if you give it a few seconds and try again, it works.
Gameplay is simple. You either either "play" or "reset". A play produces 4 types of results: $, Pnts, -Pnts, or LevelUp. Cost of "play" starts at 1 and doubles after hitting LevelUp("^Up^"). The BigPrize also doubles after hitting "^Up^". "reset" is free and sets Cost and BigPrize to original values. After the 5th "^Up^", they produce no effect. Points never go below 0, and points above 100 do not accumulate.
At the beginning of the game, play costs $1 credit and there is a $10 immediate win on the board. The BigPrize is $10 and there is also a 100Pnt space so you can immediately win that as well. As things are, a quick look at the math make me think optimal play requires 20 Pnts for Level 2, and goes up from there. At higher levels some of the decisions can be quite a bit worse than optimal if you're playing without many points which may not be immediately obvious since both Cost and BigPrize doubles as levels go up.
I feel like it may value, but I'm busy and I think it's very unlikely I will follow through at this point. Finding out that IGT is less approachable than previously thought makes me less eager to follow through as well. I am not abandoning the demo or code, just the idea. You're welcome to run with it. Presumably, a specific board/tile distribution could still be patented, or some other tile type added. If you want to use the demo (which would have to be adjusted once the math is ironed out), contact me.
http://frozen-gorge-8975.herokuapp.com/jtswebapps/liquidate
I'm still somewhat new to heroku. The javascript is not delivered under some circumstances, though if you give it a few seconds and try again, it works.
Gameplay is simple. You either either "play" or "reset". A play produces 4 types of results: $, Pnts, -Pnts, or LevelUp. Cost of "play" starts at 1 and doubles after hitting LevelUp("^Up^"). The BigPrize also doubles after hitting "^Up^". "reset" is free and sets Cost and BigPrize to original values. After the 5th "^Up^", they produce no effect. Points never go below 0, and points above 100 do not accumulate.
At the beginning of the game, play costs $1 credit and there is a $10 immediate win on the board. The BigPrize is $10 and there is also a 100Pnt space so you can immediately win that as well. As things are, a quick look at the math make me think optimal play requires 20 Pnts for Level 2, and goes up from there. At higher levels some of the decisions can be quite a bit worse than optimal if you're playing without many points which may not be immediately obvious since both Cost and BigPrize doubles as levels go up.
