Keno explosion gaffe?
Poll
 | 1 vote (3.44%) | ||
| | 2 votes (6.89%) | ||
| | 3 votes (10.34%) | ||
| | 4 votes (13.79%) | ||
| | 2 votes (6.89%) | ||
| | 6 votes (20.68%) | ||
| | 5 votes (17.24%) | ||
| | 1 vote (3.44%) | ||
| | 16 votes (55.17%) | ||
| | 3 votes (10.34%) |
29 members have voted
Yesterday I played Keno Explosion to gather information for an analysis. Here are is an executive summary of the rules.
1. The game selects three numbers to place with bombs among those not chosen by the player.
2. If the ball draw hits a bomb, then it will explode into 1 to 3 fragments.
3. Fragments may only hit the 57 numbers that were not in the ball draw nor marked by a bomb.
4. If a number picked by the player is hit by a fragment, it shall count as a catch.
In a small sampling, I determined the average number of fragments per bomb is about 1.85.
However, the interesting thing is bombs seemed to be going off an unusually high number of games. In particular all three bombs. In 55 games I recorded, 7 resulted in all three bombs exploding. If the game followed natural odds, the probability of catching all three bombs would be 1.39%. The expected number of games to catch all three bombs in 55 games played would be 0.763. The probability of catching all 3 bombs 7 or more times in 55 games is 1 in 89,202.
I'd like to emphasize that if this game is gaffed, it is done so in the player's favor. A simulation shows that if the game followed natural odds, the return would be only 82.5%. On the same machine and denom, other keno games were set to 89.5%.
I document all this more carefully in a new page on Keno Explosion. Before I announce this page officially, I welcome all comments.
The question for the poll is do you think the game is gaffed and does it bother you?
Quote: DRichI hate that you use the word gaffed. Without knowing the algorithm and the rules is it really gaffed which I interpret as cheating ? My guess is that it is not statistically probable based on your assumptions which may or may not be correct.
If I wanted a stronger word, I would have said "cheat." I mean it in the same way an electronic or digital prize wheel might not land on the jackpot slice as much as slices with a lower prize. In other words, something that is technically legal, but still deceptive.
Quote: billryanIs 55 games enough to even begin to analyze a new game?
Sure. Zero games is enough to analyze most keno games that operate by way of a draw that corresponds to the random probabilities. You just need to know the paytable and free games rules. It used to be you could take for granted that all numbers picked are random.
Quote: billryanIs 55 games enough to even begin to analyze a new game?
If you flipped a coin 55 times and it was tails for all of them, would you be suspicious or would you need to flip it more?
There is a YouTube video of someone playing it at Wynn: 38 plays has 19 0s, 3 1s, 9 2s, and 7 3s.
Perhaps hidden somewhere in the help screens is the probability of getting particular numbers of bombs hit? (Don't MultiStrike Poker machines usually say what the probability of a Free Ride is in their help screens?)
Most people call it rigged. Look in slots forum, I posted about a rigged slot. Either I'm the unluckiest and luckiest or it is rigged. If you're playing pool with someone and they let you win for $10, then hustle you down the line $1000, being rigged is another word for hustled. Of course maybe your keno game is rigged in your favor, but what if there was another version rigged the other way, or one people can't predict so easily.Quote: WizardIf I wanted a stronger word, I would have said "cheat." I mean it in the same way an electronic or digital prize wheel might not land on the jackpot slice as much as slices with a lower prize. In other words, something that is technically legal, but still deceptive.
Quote: WizardIf you flipped a coin 55 times and it was tails for all of them, would you be suspicious or would you need to flip it more?
If someone came on and said they played 55 hands of BJ and didn't get a single BJ, would you say the game was gaffed or tell them they needed a few thousand more hands to decide? How is this any different?
You don't know, that's so cute. Adde: To be fair, everyone doesn't need to know everything, you have knowledge about comics and collectibles, and probably other things to shame us all. Low probability doesn't need a lot of trials as something with a higher probability. Yet, somehow a coin can flip heads or tails 999/1000 or 99/100 or 9/10.Quote: billryanIf someone came on and said they played 55 hands of BJ and didn't get a single BJ, would you say the game was gaffed or tell them they needed a few thousand more hands to decide? How is this any different?
Quote: ThatDonGuyIs MultiStrike Poker considered "gaffed" since the Free Ride card shows up more than 1/53 of the time? I would consider Keno Explosion the same way.
There is a YouTube video of someone playing it at Wynn: 38 plays has 19 0s, 3 1s, 9 2s, and 7 3s.
Perhaps hidden somewhere in the help screens is the probability of getting particular numbers of bombs hit? (Don't MultiStrike Poker machines usually say what the probability of a Free Ride is in their help screens?)
The “Free Ride,” is not a card that would normally be in a deck of playing cards, totally different and separate concept. This Keno game might be, based on Wizard’s findings, and likely is choosing specific numbers for bombs and the bomb hits, all evidence suggests, do not correspond to the probabilities of those numbers hitting in twenty draws.
There are ways around having to actually use the numbers. For example, one Quick Hit variation has three fire symbols that appear away from the reels and lead to Free Games if all light up. This game could have done something similar to that.
While the inconsistency may work in the player’s favor for this game, it perhaps opens the door for inconsistencies in keno games that do not work in the player’s favor.
Quote: billryanIf someone came on and said they played 55 hands of BJ and didn't get a single BJ, would you say the game was gaffed or tell them they needed a few thousand more hands to decide? How is this any different?
I would say it was gaffed, based on their version.
Quote: billryanIf someone came on and said they played 55 hands of BJ and didn't get a single BJ, would you say the game was gaffed or tell them they needed a few thousand more hands to decide? How is this any different?
There is a concept called statistical confidence and it can be calculated. It allows you to say that the evidence cited implies some level of confidence, say 20% or 82%, or 96% or 99.93%, that the underlying mechanism is not random.
Wizard cites something that should logically happen 1.39% of the time and yet it occurs 7 times in 55 trials. There is a high confidence ( I don't know the level) that the action occurs more frequently than 1.39%. It doesn't require many trials to gain confidence if the action you are observing should only occur rarely and yet it occurs very frequently in a few trials.
It would be helpful to see all of the help screens too.
Edit: I see the help screens. It does not say that a fragment cannot hit another unexplored bomb. Seems like there may be a chain reaction?
Quote: CrystalMathThis is illegal, even if it’s in the player’s favor.
Can you quote me a regulation? I thought the one about natural odds applied to dice and cards only. I'd be happy to bring a formal complaint about this if I had a legal leg to stand on.
It used to be 14.040.2(b).
For gaming devices that are representative of live gambling games, the mathematical probability of a symbol or other element appearing in a game outcome must be equal to the mathematical probability of that symbol or element occurring in the live gambling game.
I watched some videos, and the game is gaffed, in my opinion, and it really makes me mad.
Quote: CrystalMathI believe NGCB 14.040.5
It used to be 14.040.2(b).
For gaming devices that are representative of live gambling games, the mathematical probability of a symbol or other element appearing in a game outcome must be equal to the mathematical probability of that symbol or element occurring in the live gambling game.
I watched some videos, and the game is gaffed, in my opinion, and it really makes me mad.
It’s going to make you more mad when a Keno game is gaffed to look like it would pay in the high-90’s if the special feature actually corresponded to natural probabilities.
Maybe not today, maybe not tomorrow, but sometime.
Or Caveman Plus Keno, except the game decides to put the eggs in spots after the fact of the draw and the probability of an egg landing on one of the twenty numbers is less than a Keno game would otherwise dictate.
Are you really ever happy bringing a formal complaint? I've always felt the casinos would retaliate and have never done it. The one time I did complain, they just showed their ignorance and I got tired of fighting them, not wanting to go through it all again with a supervisor.Quote: WizardCan you quote me a regulation? I thought the one about natural odds applied to dice and cards only. I'd be happy to bring a formal complaint about this if I had a legal leg to stand on.
Quote: CrystalMathI believe NGCB 14.040.5
It used to be 14.040.2(b).
For gaming devices that are representative of live gambling games, the mathematical probability of a symbol or other element appearing in a game outcome must be equal to the mathematical probability of that symbol or element occurring in the live gambling game.
Thank you.
It looks to be 14.040.5 (page 8).
Quote: Regulation 14.040.5For gaming devices that are representative of live gambling games, the mathematical probability
of a symbol or other element appearing in a game outcome must be equal to the mathematical probability
of that symbol or element occurring in the live gambling game.
Quote: Mission146It’s going to make you more mad when a Keno game is gaffed to look like it would pay in the high-90’s if the special feature actually corresponded to natural probabilities.
Maybe not today, maybe not tomorrow, but sometime.
Yes. This game is opening that door.
Quote:The shall choose 20 numbers, without replacement, from 1 to 80.
typo
Quote:If the ball draw by the game matches one of the three numbers marked by a bomb, then that bomb shall explode. After the explosion, one to three fragments will land on one of the 57 numbers not chosen by the game or marked with a bomb.
Bombs can't land on bombs confirmed by this statement. I find this sentence confusing following the sentence before, because "drawn" is right above "draw".
Quote: rule screen picA maximum of 9 fragments can appear in a single game
Explosions and fragments are evaluated after all numbers are drawn...
Wins are shown in credits unless marked as currency.
Bombs cannot land on marked numbers, correct? Fragments cannot land on marked numbers, correct? This is why "explosions and fragments are evaluated after all numbers are drawn".
This post event evaluation is common wording on some slot machines which require it, cannot think of an example. I believe usually to do with pays, just so small pays aren't awarded in addition to a larger and improved pay by some game feature. I find the phrase fair and excusable, though an extra or hidden meaning could exist. I'm doubtful.
Does the game force you to bet in even amounts of credits or is the player not paid fully with half credits?
Quote:The following table shows an example pay table. Note that many pays end in 0.5. To keep the total pays in round numbers, the player must bet an even number of coins.
Same question as above.
Maybe. There would be 20 balls drawn to account for hitting the numbers and 3 for the bombs, unless the bombs cut into the 20 to make it 17 for numbers and 3 for bombs. You have to test play to see if the bombs are...I'm going to keep this train of thought here...I looked at the pictures and the bombs do correspond to drawn numbers. This is screwy and complex. So it isn't just getting 3 of 20, it is representing getting 3/14 for a 6 spot. I see the wording now being strange and not innocent as I assumed. Why are they even correlating the drawn numbers with the bombs for explosions, if that's not how the game is really working? I see the entire point now I have read this completely. Are the bombs matching the drawn numbers what causes them to explode or what places them on the board? Maybe I am thinking this because I haven't slept in over 24 hours, but can you play the game again and just test the first stage of the game for being legit with a 2 spot before the bombs come into play?Quote:It should be noted the first rule screen says, "Explosions and fragments are evaluated after all numbers are drawn." While this is open to interpretation, it suggests to me everything that goes on with the bombs is determined independently and not necessarily by the random draw of balls.
Quote: WizardThank you.
It looks to be 14.040.5 (page 8).
And it appears that the 20 balls being drawn do have an equal chance of being any of the C(80,20) ways to do it. Since drawing the numbers for the bombs has nothing to do with "live" Keno, I don't think the way those three numbers are selected has to be uniformly random to comply with the regulation. See my earlier comment about the "free ride card" in MultiStrike Poker for a similar case.
Here's a question: once a bomb releases fragments, do the numbers that the fragments have to land on have to be uniformly random?
Quote: ThatDonGuy
Here's a question: once a bomb releases fragments, do the numbers that the fragments have to land on have to be uniformly random?
I would say yes, and the Wizard also made this assumption. But, I don’t trust anything on this game to work fairly.
I found this in a NGCB notice:
“ Gaming devices which do not utilize elements commonly used in live games may alter the probability of game elements appearing in a winning or losing combination in a bonus game if it is clear to the patron that a change has been made.”
This game uses elements commonly used in live games, so this does not apply.
. You are the expert. I’m a layman. But once you introduce ‘bombs and fragments’ they can argue those are NOT elements commonly used in live games.Quote: CrystalMathI would say yes, and the Wizard also made this assumption. But, I don’t trust anything on this game to work fairly.
I found this in a NGCB notice:
“ Gaming devices which do not utilize elements commonly used in live games may alter the probability of game elements appearing in a winning or losing combination in a bonus game if it is clear to the patron that a change has been made.”
This game uses elements commonly used in live games, so this does not apply.
I know Axel must be salivating..... the question for you guys is this.... has this gaffe left a player advantage, and if so, what is the best way to exploit it?
Quote: SOOPOO. You are the expert. I’m a layman. But once you introduce ‘bombs and fragments’ they can argue those are NOT elements commonly used in live games.
I know Axel must be salivating..... the question for you guys is this.... has this gaffe left a player advantage, and if so, what is the best way to exploit it?
That’s where I see the fault in the rule. Where do you draw the line? I think most people would expect them to be randomly distributed because it’s a keno grid. This game, however, has also altered the initial distribution of the 20 balls, which are elements used in live games.
I’ve designed maybe 50 keno games and it would never cross my mind to do something like this. If they need more explosions, have 4 bombs or increase the average number of fragments, but don’t alter natural probabilities.
Quote: CrystalMathWhere do you draw the line?
Its all about what the gaming control board approves, and they know everything about every game that goes into their casinos... or am i wrong to assume that?
how about artificial intelligence in a table card game?
https://scholars.law.unlv.edu/cgi/viewcontent.cgi?article=1045&context=glj
Quote:When Shuffle Master submitted the game Dealer Bluff Six Card Poker for
approval, the Gaming Control Board warned the manufacturer that the game
might violate the text of NRS 465.075.99 The problem was that Dealer Bluff
included a small element of artificial intelligence, designed to make it more
interesting for players.100 The Gaming Control Board and Nevada Gaming
Commission described the game as fair,101 and maybe fairer because of the
artificial intelligence element.102
Quote: CrystalMathI would say yes, and the Wizard also made this assumption. But, I don’t trust anything on this game to work fairly.
I found this in a NGCB notice:
“ Gaming devices which do not utilize elements commonly used in live games may alter the probability of game elements appearing in a winning or losing combination in a bonus game if it is clear to the patron that a change has been made.”
This game uses elements commonly used in live games, so this does not apply.
It's also not "in a bonus game."
Somebody who's in Vegas (or anywhere else where there's a machine with this game) needs to scan the help/payout screens and see if any probabilities are mentioned.
Quote: ThatDonGuyIt's also not "in a bonus game."
Somebody who's in Vegas (or anywhere else where there's a machine with this game) needs to scan the help/payout screens and see if any probabilities are mentioned.
The Wizard did include some photos of the help screen here:
https://wizardofodds.com/games/keno/keno-explosion/
IGT could have designed a fair game, so why didn’t they? Is it to obscure the fact that two machines with the same pay table have different RTP? Is the intention just to push the rules and open Pandora’s box?
The game was obviously approved by NGCB, but it still violates the intent of the law.
It’s somehow better when you say it vs billryan.Quote: MoneyK1ngIs 55 games enough to even begin to analyze a new game?
Quote: onenickelmiracleBombs cannot land on marked numbers, correct?
Correct.
Quote:Fragments cannot land on marked numbers, correct?
Correct.
Quote:This is why "explosions and fragments are evaluated after all numbers are drawn".
I don't understand what is meant by that statement either. In my opinion, it is their admission the game is gaffed, but what player will understand it?
Quote:This post event evaluation is common wording on some slot machines which require it, cannot think of an example. I believe usually to do with pays, just so small pays aren't awarded in addition to a larger and improved pay by some game feature. I find the phrase fair and excusable, though an extra or hidden meaning could exist. I'm doubtful.
I'm not sure I see your point there.
Quote:Does the game force you to bet in even amounts of credits or is the player not paid fully with half credits?
It makes you bet in even numbers of credits. I think 2, 4, 6, 8 or 10.
Quote:Maybe. There would be 20 balls drawn to account for hitting the numbers and 3 for the bombs, unless the bombs cut into the 20 to make it 17 for numbers and 3 for bombs. You have to test play to see if the bombs are...I'm going to keep this train of thought here...I looked at the pictures and the bombs do correspond to drawn numbers. This is screwy and complex. So it isn't just getting 3 of 20, it is representing getting 3/14 for a 6 spot. I see the wording now being strange and not innocent as I assumed. Why are they even correlating the drawn numbers with the bombs for explosions, if that's not how the game is really working? I see the entire point now I have read this completely. Are the bombs matching the drawn numbers what causes them to explode or what places them on the board? Maybe I am thinking this because I haven't slept in over 24 hours, but can you play the game again and just test the first stage of the game for being legit with a 2 spot before the bombs come into play?
Yes, a bomb explodes if one of the 20 balls drawn by the game matches the number a bomb is on. Consider the example in my page.
Bombs were put on 1, 36, and 41. You can see those numbers were drawn, as evidenced by the balls on the right and that the squares they are on are red. Actually, the 20th ball drawn was 41 and you can see it only half turned in the screenshot, but you can see it all the way in the image age the end of the game.
I apologize if this reply didn't hit all your points. I wasn't sure what all of them were.
Quote: ThatDonGuyAnd it appears that the 20 balls being drawn do have an equal chance of being any of the C(80,20) ways to do it.
No! The numbers with bombs have a much greater chance of being drawn by the 20-ball draw of the game. I would agree that if you tracked balls numbers drawn, they would probably be equal, because they randomize where the bombs are put. If the player could choose the bomb locations, for example 1 to 3, then 1 to 3 would be hit more than 20/80 of the time each.
Quote:Since drawing the numbers for the bombs has nothing to do with "live" Keno, I don't think the way those three numbers are selected has to be uniformly random to comply with the regulation. See my earlier comment about the "free ride card" in MultiStrike Poker for a similar case.
I disagree with your opinion. As to MultiStike, the "free ride card" isn't really a card. It's a random feature they display to the player as something that looks like a card.
Quote:Here's a question: once a bomb releases fragments, do the numbers that the fragments have to land on have to be uniformly random?
I am fine with the rule that fragments can't hit numbers already drawn, have a bomb, or were already hit by an earlier fragment. Otherwise, I think all should be equally likely to be compliant with the spirit of fair and honest gambling in Nevada.
if we allow slot machines to weight their symbols and how often they come up
why is this an issue - can they not choose to weight the symbols to whatever they please?
slot machines in real life dont have the amount of stops they have on in the digital world and the normal games would not meet the requirements set by law so we allow them to add digital stops in order to make these machines pay out exactly as we want
Quote: heatmapserious question
if we allow slot machines to weight their symbols and how often they come up
why is this an issue - can they not choose to weight the symbols to whatever they please?
slot machines in real life dont have the amount of stops they have on in the digital world and the normal games would not meet the requirements set by law so we allow them to add digital stops in order to make these machines pay out exactly as we want
The difference, or at least why people would expect this to correspond to natural probabilities, is because video keno is a representation of the live game of keno.
It’s the same thing as VP representing physical playing cards.
Whereas, slots aren’t an electronic representation of anything physical with set probabilities.
Quote: Mission146The difference, or at least why people would expect this to correspond to natural probabilities, is because video keno is a representation of the live game of keno.
It’s the same thing as VP representing physical playing cards.
Whereas, slots aren’t an electronic representation of anything physical with set probabilities.
i dont know why i asked this i kind of knew the answer i guess i just wanted confirmation - i also forgot keno was an actual game that was played before video gaming terminals lol
Quote: Mission146Whereas, slots aren’t an electronic representation of anything physical with set probabilities.
I realize you know this, but just to clarify for others if a slot machine contains a feature using an electronic representation of anything physical with set probabilities it must adhere to those probabilities. There are several slots that use dice outcomes and those outcomes must adhere to the real probabilities.
Quote: sabreI realize you know this, but just to clarify for others if a slot machine contains a feature using an electronic representation of anything physical with set probabilities it must adhere to those probabilities. There are several slots that use dice outcomes and those outcomes must adhere to the real probabilities.
This is not true in Nevada. I did a slot machine with dice as some of the reel symbols and they did not need to be evenly weighted because the game itself was not mimicking another game.
Quote: WizardNo! The numbers with bombs have a much greater chance of being drawn by the 20-ball draw of the game. I would agree that if you tracked balls numbers drawn, they would probably be equal, because they randomize where the bombs are put. If the player could choose the bomb locations, for example 1 to 3, then 1 to 3 would be hit more than 20/80 of the time each.
That assumes that the bomb locations are selected before the numbers are drawn. I have a feeling it is the other way around. Among other things, it makes it a lot easier to control how many bombs are hit.
Quote: ThatDonGuyThat assumes that the bomb locations are selected before the numbers are drawn. I have a feeling it is the other way around. Among other things, it makes it a lot easier to control how many bombs are hit.
I think you are right, but the game doesn’t display it that way. I think the game draws the 20 random balls and then decides where to put the bombs in order to have them hit. Either way, it’s deceptive.
May as well randomly display 0-3 bombs off the grid, then nobody would expect anything.
https://www.pacodeandbulletin.gov/Display/pacode?file=/secure/pacode/data/058/chapter810/s810.5.html
Quote:(d) The scaling method may not compromise the cryptographic strength of the random number generator. The scaling method must preserve the distribution of the scaled values. For example, if a 32-bit random number generator with a range of the set of integers in the closed interval [0, 232-1] were to be scaled to the range of the set of integers in the closed interval [1, 6] so that the scaled values can be used to simulate the roll of a standard six-sided die, then each integer in the scaled range should theoretically appear with equal frequency. In the example given, if the theoretical frequency for each value is not equal, then the scaling method is considered to have a bias. Thus, a compliant scaling method must have bias equal to zero.
also - this process isnt necessarily UNFAIR right?
because thats what were essentially worried about when weve created these laws
but this process somehow increases the players advantage?
which isnt necessarily unfair?
or since its NOT random it IS unfair?
Quote: CrystalMathI think you are right, but the game doesn’t display it that way. I think the game draws the 20 random balls and then decides where to put the bombs in order to have them hit. Either way, it’s deceptive.
I agree.
Quote: heatmapPennsylvania's Law which pertains to what were talking about - at least i think this is what were talking about algorithmically - maybe not though
That is not what we're talking about.
Direct: https://www.youtube.com/watch?v=pNIU6JyYCNQ
For those who couldn't stand my sample size of 55 before, despite being overwhelmingly conclusive, I am now up to 263 games played and recorded. Here are the updated totals.
| Bombs | Obervations | Expected |
|---|---|---|
| 3 | 37 | 3.65 |
| 2 | 38 | 36.49 |
| 1 | 34 | 113.32 |
| 0 | 154 | 109.54 |
| Total | 263 | 263.00 |
A chi-squared test shows the probability of results this skewed or more are 1 in 958,700,425,938,185,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000. Is more data required?
.Quote: WizardHere is my video documenting my case that Keno Exposion is gaffed and a violation of NRS 14.050.5 .
Direct: https://www.youtube.com/watch?v=pNIU6JyYCNQ
For those who couldn't stand my sample size of 55 before, despite being overwhelmingly conclusive, I am now up to 263 games played and recorded. Here are the updated totals..
Bombs Obervations Expected 3 37 3.65 2 38 36.49 1 34 113.32 0 154 109.54 Total 263 263.00
A chi-squared test shows the probability of results this skewed or more are 1 in 958,700,425,938,185,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000. Is more data required?
I forget how exactly YOU define gaffed. If it as simple as ‘the outcomes don’t match what the actual physical game would be’ then you have proven it is gaffed. If you include in the definition somehow that the win percentage is changed you have NOT proven that. In the 263 games the TOTAL expected number of bombs seems about on expectation, just more ‘extremes’ (0 or 3)
Quote: SOOPOOI forget how exactly YOU define gaffed. If it as simple as ‘the outcomes don’t match what the actual physical game would be’ then you have proven it is gaffed. If you include in the definition somehow that the win percentage is changed you have NOT proven that. In the 263 games the TOTAL expected number of bombs seems about on expectation, just more ‘extremes’ (0 or 3)
I could prove that, but I'm afraid it would muddy the waters. If I programmed this game, assuming no gaffe, it would show a return of of about 80%, when in fact it is probably about 90%. Part of the reason is there are 12% more bombs hit than expected. Another his the bombs hit tend to be clumped in the same game. It would be better to play one game with 22 balls drawn and one with 20 than two with 21.
