Lion's Share
Quote: DRichIs the 1 in 256 an assumption based on previous results or is it speculation that the virtual reel is one byte? The game runs on an S+ machine which was also used for Megabucks and clearly that has a virtual reel bigger than one byte.
Axiom is more confident about the 256 virtual stops than I am. I have a box full of par sheets and the total number of stops seems pretty arbitrary. Then again, most are by IGT, so I'm mostly getting a sense of their style. Lion's Share is my Mikohn, who I didn't even know ever made slots, until I saw this game.
In Axiom's defense, choosing random numbers from an even power of 2 was more of a programming style back when that game was made in 95 than it is now.
While I respect Axiom's opinion a great deal, and I'm not saying he is wrong, but I'm not so convinced Lion's Share is on a 256-stop reel map.
I hope ME stumbles upon this post as he knows slot programming style really well.
Quote: WizardAxiom is more confident about the 256 virtual stops than I am. I have a box full of par sheets and the total number of stops seems pretty arbitrary. Then again, most are by IGT, so I'm mostly getting a sense of their style. Lion's Share is my Mikohn, who I didn't even know ever made slots, until I saw this game.
In Axiom's defense, choosing random numbers from an even power of 2 was more of a programming style back when that game was made in 95 than it is now.
While I respect Axiom's opinion a great deal, and I'm not saying he is wrong, but I'm not so convinced Lion's Share is on a 256-stop reel map.
I hope ME stumbles upon this post as he knows slot programming style really well.
Are you sure the game is a Mikohn? In the 1990's they were the major progressive controller manufacturer and from what I remember the game looks like an IGT S+. I am speculating they bought the machine from IGT and sold it with their controller for the progressive.
Also, yes, I assumed that the number of stops would be a power of 2, and 256 jumped out of my data as the number that seemed to fit. If that assumption was wrong then I am definitely wrong about this.
However, either way, I think that we agree that we need more data about the probability of a lion on the payline, since being slightly off on the probability leads to a large difference in payout.
Ever since the discussion of the safety bet, and now this, I have been studying this -- how many trials are required to be fairly sure that your observed probabilities match the actual probability? Perhaps I should have taken more stats courses in school.... anyway, the Wilson score interval keeps coming up, but, honestly, I haven't had the time to really read it carefully so I don't understand it yet. (I've been spending a lot of time writing blackjack simulations -- there is just something addictive about the math of that game)
In particular, it says that the interval has "good properties" even for a small number of trials or extreme probabilities (which is exactly what our two problems are) so I'm hoping that this might give a good indication of how sure we can be that the theoretical return from the data collected is reasonably close to the actual return.
Quote: WizardAxiom is more confident about the 256 virtual stops than I am.
I probably am too. But a lot of this is based on some random internet guy saying he counted the lions on the line each reel for over 10k spins and his last 10k spins look like this:
http://www.vegasmessageboard.com/forums/showpost.php?p=615042&postcount=296
Reel 1: 33 lions on line
Reel 2: 42 lions on line
Reel 3: 34 lions on line
If one assumes a 1 in 256 probability for each reel then the reels fall in the following percentiles:
Reel 1: 18.8th
Reel 2: 71.5th
Reel 3: 23.6th
Wiz's chance of 5 or more lions in 332 spins under 1 in 256 hypothesis: 1.04% (unlikely but still possible)
All these results are definitely acceptable with a 1 in 256 hypothesis. Of course, you have to trust the data of a random internet poster.
Quote: AxiomOfChoice
Ever since the discussion of the safety bet, and now this, I have been studying this -- how many trials are required to be fairly sure that your observed probabilities match the actual probability? Perhaps I should have taken more stats courses in school.... anyway, the Wilson score interval keeps coming up, but, honestly, I haven't had the time to really read it carefully so I don't understand it yet. (I've been spending a lot of time writing blackjack simulations -- there is just something addictive about the math of that game)
In particular, it says that the interval has "good properties" even for a small number of trials or extreme probabilities (which is exactly what our two problems are) so I'm hoping that this might give a good indication of how sure we can be that the theoretical return from the data collected is reasonably close to the actual return.
Looked this Wilson guy up, and he was Josiah Willard Gibbs main grad student. Interesting!
Quote: tringlomaneI probably am too. But a lot of this is based on some random internet guy saying he counted the lions on the line each reel for over 10k spins and his last 10k spins look like this:
http://www.vegasmessageboard.com/forums/showpost.php?p=615042&postcount=296
The ratio of spins to lions for SmokaLot is 90.9 and for me is 65.6. If his data is accurate then it certainly makes it look like I got more than my fair share.
If I combined my lion data only with his, and then assume the frequency of every other symbol based on my data only, then the return table looks like this.
| Win | Pays | Probability | Return |
|---|---|---|---|
| Three Lions -- On Payline | 2,342,270 | 0.0000000568 | 0.044315 |
| Three Lions -- Off Payline | 10,000 | 0.0000030655 | 0.010218 |
| Three sevens | 300 | 0.0001037921 | 0.010379 |
| Three 3-bar | 150 | 0.0005523135 | 0.027616 |
| Three 2-bar | 60 | 0.0031613656 | 0.063227 |
| Three 1-bar | 30 | 0.0164971560 | 0.164972 |
| Any three bar | 15 | 0.1223141342 | 0.611571 |
| Any two lions | 15 | 0.0002963758 | 0.001482 |
| Any one lion | 6 | 0.0319746131 | 0.063949 |
| Total | 0.1749028725 | 0.997728 |
In other words a 99.77% return.
Do we think that the game had a base payout (when the jackpot was initially $1M) was greater than 97%?
That's what you currently have calculated. And this wouldn't fly in today's gaming climate. But if they haven't changed the paybacks since 1995...maybe...just maybe.
The overall win percentage of dollar slots/video poker was 4.87% (95.13% return) for >$72M gross revenue Strip properties in 1995.
http://gaming.nv.gov/modules/showdocument.aspx?documentid=3751
I am assuming most dollar video poker was close to 99% max return back then...I hope. So if Lion's Share was always above >97% return, this is definitely a "special" machine.
So how much variance do we think is left in the lower level payouts? Obviously it takes a lot of spins to try to predict the jackpot probability. Your total lions would be in the 91.7th percentile if all reels were a 1 in 256 to hit one. Is it possible you were hotter on the lower symbols too? Considering your data, I am hoping your actual results correlated with it! :)
Quote: tringlomaneI guess one lingering question I have is...
Do we think that the game had a base payout (when the jackpot was initially $1M) was greater than 97%?
That's what you currently have calculated. And this wouldn't fly in today's gaming climate. But if they haven't changed the paybacks since 1995...maybe...just maybe.
The overall win percentage of dollar slots/video poker was 4.87% (95.13% return) for >$72M gross revenue Strip properties in 1995.
http://gaming.nv.gov/modules/showdocument.aspx?documentid=3751
I am assuming most dollar video poker was close to 99% max return back then...I hope. So if Lion's Share was always above >97% return, this is definitely a "special" machine.
So how much variance do we think is left in the lower level payouts? Obviously it takes a lot of spins to try to predict the jackpot probability. Your total lions would be in the 91.7th percentile if all reels were a 1 in 256 to hit one. Is it possible you were hotter on the lower symbols too? Considering your data, I am hoping your actual results correlated with it! :)
Assuming a $1,000,000 initial jackpot (and using the last table that the Wizard posted, which includes smokealot's numbers, and giving the 100.1% return), the return from the jackpot is reduced from 4.65% to 1.98% when it was reset, which would put the total return (for 3-coin play) at 97.43%.
Furthermore, 0.75% goes to the meter (not to the casino) meaning that the house edge would be 1.82% (ie, return + meter = 98.18%). Was that considered reasonable for 3-coin dollar slots in 1995?
She sees me coming so she spins one more time then hits cash out button. Man on next machine says he is so glad she finally stopped. Josie gives him a funny look, What the hell, it's only $20 or $30. Well , not exactly $1,700 +. I laughed all the way to the airport.
So if DUMB luck count's, that Jackpot is Josie's !
Quote: AxiomOfChoiceFurthermore, 0.75% goes to the meter (not to the casino) meaning that the house edge would be 1.82% (ie, return + meter = 98.18%). Was that considered reasonable for 3-coin dollar slots in 1995?
I say no, but it doesn't mean it didn't happen either.
However, if I was running a casino, I wouldn't be setting this machine above the average payout for the denomination, which included video poker. Look how many people play Megabucks and don't care about the payout. A base payout of 95% would have been more likely back then I would think. But again, this is just speculation. Maybe for awhile they promoted the units to be 98%+ too?
Quote: NareedHow long has this machine not been paying the progressive? I get the impression it's been a very long time. One cannot help but think perhaps three's a glitch or bug (or feature) keeping it from paying up.
Not at all.
If the meter started at $1,000,000, that means that $1,342,270 has been added to the meter.
It's a 0.75% meter, so that means that $178,969,333 has been played through the machine.
If we ignore short-coin play, that is 59,656,444 spins.
If the probability of hitting the jackpot is, indeed, 1/256^3, that is 1 in 16,777,216.
The probability of going that many spins without hitting the jackpot is about 2.8%.
Realistically, there is a lot of short-coin play, so there have been fewer full-coin spins than that, so that probability is significantly higher.
Quote: AxiomOfChoiceNot at all.
Snip Progressive Math
Yep. Also, I would assume that the Gaming Commission has tested the EPROM chip(s) at least once, I would hope. So the winning combination should be in play.
Quote: djatcMgm should advertise the machine more. It'll get people excited to try and hit a progressive that's been building for such a long time
I agree. I couldn't even find the damn thing last time. :( But from estimates earlier in this thread, it looks like the machine gets much more coin-in than the average machine already.
What puzzles me is even if so, it hasn't been changed like the RWB Alan mentioned earlier. Maybe it's grandfathered in being unable for the gaming chairman to approve changes and common sense would say yes. All the regs seem corrupt anyways regarding progressive changes midstream and wish they would repeal allowing shenanigans with progressives. Knowing this one can't be changed once articles get written would be a good point for or against it.
Quote: tringlomaneI say no, but it doesn't mean it didn't happen either.
However, if I was running a casino, I wouldn't be setting this machine above the average payout for the denomination, which included video poker. Look how many people play Megabucks and don't care about the payout. A base payout of 95% would have been more likely back then I would think. But again, this is just speculation. Maybe for awhile they promoted the units to be 98%+ too?
I wouldn't be surprised at all if it was around 96%. That would have been a reasonable payback percent for a $1 game back then.
Quote: tringlomaneYep. Also, I would assume that the Gaming Commission has tested the EPROM chip(s) at least once, I would hope. So the winning combination should be in play.
You assume or you hope? ;)
Still, the thing about glitches is they can hide anywhere. Say the EPROM is ok, but the RNG isn't. Or something else, even. Remember sometime in the 90s, I think that AT&T's LD network went down all over the US due to three faulty lines of code? I figure that code was checked extensively before, during and after testing and implementation.
Oh, this reminds me of a bug in Win95. If the PC was on when daylight savings started, the system would ask permission to change the clock back one hour. If you were awake and clicked "yes," it went from 2 am to 1 am. Then when it hit 2 am again, it tried to change the clock once more. If you waited until the following morning instead, the system informed you the time had changed during the night.
Stuff like that happens.
Quote: Nareed
Stuff like that happens.
Oh definitely. It could be malfunctioning, but there is at least about a 3% chance a perfectly functioning machine would get to this state today after ~18 years. So obviously hard to say without various levels of testing.
And I hope they've checked it by now...that was redundant. Assume makes "an ass out of u and me".
Quote: DRichI wouldn't be surprised at all if it was around 96%. That would have been a reasonable payback percent for a $1 game back then.
Yeah, but >97%? That would be pushing it for a progressive, no?
Quote: AxiomOfChoiceWhen you walk by a craps table and see someone roll a hard 8, do you assume that the dice are loaded? Same thing...
If I was on it, no it's not loaded, it's working as intended. If I wasn't on it, who knows. If I am playing the right side and I seven out quickly it's loaded.
Same for video poker. If I run good on a machine it's working as intended with no gaffs, but if I run bad and lose a lot it's obviously cheating.
Quote: MoosetonI might totally be wrong here and I don't want to spread misinformation but I thought I'd read somewhere that it started at $10,000. Anyone else heard that one too?
Considering you get $25k on one coin bet, and $50k on 2 coins bet, I am pretty sure that the jackpot for 3 coins did not start at $10k.
Quote: jimandjudiIn https://wizardofodds.com/games/slots/lions-share/ Michael speculates, "It is possible they changed the EPROM chip between my sessions." However, I was under the impression that to changing EPROMs requires "remove machine" and "add machine" paperwork, possibly changing the machine number. In general, can (Nevada) casinos change EPROMs without changing the machine number? What about Indian casinos? (Sorry if that is too much off-topic.)
It is my impression that Nevada casinos have more flexibility in changing game settings, and they are fined heavily if they are caught doing something wrong. In other jurisdictions, the gaming regulators want to know exactly what software is installed at all times. It used to be, in Colorado, that the Colorado Division of Gaming had to be present to change an EPROM.
In any event, I think that changing the EPROM would be unnecessary and an extra headache. It would involve clearing the RAM on the game and a lengthy setup process. On top of that, these machines are ancient, and I wouldn't want to touch anything. The electronics are protected by a layer of tar by now.
Speaking of gaming, Josie was working the door at Lady Luck one day checking ID's. She even ID'd several ladies in late 30's or 40's, just to see them blush. She stopped doing it after a lady from gaming got pissed. She thought Josie knew she was from gaming. LOL
Quote: WizardIt is my understanding that linked progressives like Megabucks have only one setting/EPROM.
This is likely true. At GLI, when there was a Megabucks submission, there would be either 1 or 2 EPROMs and they did not have configurable paytables. Also, they would have the same top award odds on both EPROMs, since linked jackpots must have identical odds. The difference would be only in the frequency of other symbols.
Quote: CrystalMathThis is likely true. At GLI, when there was a Megabucks submission, there would be either 1 or 2 EPROMs and they did not have configurable paytables. Also, they would have the same top award odds on both EPROMs, since linked jackpots must have identical odds. The difference would be only in the frequency of other symbols.
A little off topic, but is there a mandate that the odds of the Megabucks jackpot must be the same between the 3-reel game and 5-reel?
Quote: WizardA little off topic, but is there a mandate that the odds of the Megabucks jackpot must be the same between the 3-reel game and 5-reel?
Yes, and no. The odds must be the same for any machines on the same progressive relative to the bet. A 1 cent machine odds can be five times that of a nickel machine if the bet is the same number of credits.
Edit: Only the odds for the progressives must be the same. All other combination odds can be different and offer a different payback percentage.
If a casino forgets to submit the proper paper work whats the fine? I have a feeling its a slap on the wrist and the casino just gets a warning and given a chance to submit the new paperwork without incident.
I don't know why people think gaming is running around checking every chip. I have a felling there has to be extraordinary circumstances before a full investigation is done. Most likely some malfunction concerning a significant jackpot would be the only way they would make a full investigation or if someone was continully making money due to some gaff in a machine.
Quote: AxelWolfDose anyone know if once a chip is changed and the proper paperwork is sent in. Is this information privileged? If you called Gaming control do they have to disclose if the chip has been changed or do they just say the information is confidential.
It is privileged.
Quote:If a casino forgets to submit the proper paper work whats the fine? I have a feeling its a slap on the wrist and the casino just gets a warning and given a chance to submit the new paperwork without incident.
I'm not sure about that. IMO, Gaming does not appreciate the image of being a paper lion (sorry for the pun) and will not hesitate to throw out serious fines for rule violations.
Quote:I don't know why people think gaming is running around checking every chip.
I don't know how often they do surprise inspections, but I would guess it is rare. However, the casinos have no incentive to lie to them so I think the casinos are generally compliant with the rules.
Quote: AxelWolf
I don't know why people think gaming is running around checking every chip. I have a felling there has to be extraordinary circumstances before a full investigation is done. Most likely some malfunction concerning a significant jackpot would be the only way they would make a full investigation or if someone was continully making money due to some gaff in a machine.
They might not like having their OWN PEOPLE checking on those machines in fear of one of them rigging the machine while inside
http://www.youtube.com/watch?v=VtedwJqyg1Y
Examples? Other then well published incidents? I generally think casino are on the up and up, But i also think there are many violations going on that are unnoticed or are not penalized.Quote: Wizard
I'm not sure about that. IMO, Gaming does not appreciate the image of being a paper lion (sorry for the pun) and will not hesitate to throw out serious fines for rule violations.
.
Quote: AxelWolfI generally think casino are on the up and up, But i also think there are many violations going on that are unnoticed or are not penalized.
I tend to agree. People seem to assume that because cheating is not discovered, the regulators must be preventing it from happening. But that's not the only explanation.
Quote: AxelWolfExamples? Other then well published incidents? I generally think casino are on the up and up, But i also think there are many violations going on that are unnoticed or are not penalized.
I would be surprised if I could not find a gaming violation in at least 50% of Nevada casinos. 99% of them would be a warning or slap on the wrist if gaming found out. I have been personally involved in many over the last 20 years and most of them just result in Gaming asking for you to file the proper paperwork.
As long as Gaming doesn't believe a patron is being cheated they are lenient and just want the issue to be corrected.
This is the perception I got. Even in cases where the casino was clearly wrong its usually just a simple correct the situation. I bet if there was a gaff in a machine that affected the players (not done purposely). Gambling control would just say Fix it.Quote: DRichI would be surprised if I could not find a gaming violation in at least 50% of Nevada casinos. 99% of them would be a warning or slap on the wrist if gaming found out. I have been personally involved in many over the last 20 years and most of them just result in Gaming asking for you to file the proper paperwork.
As long as Gaming doesn't believe a patron is being cheated they are lenient and just want the issue to be corrected.
For instance, lets say a less known machine, but like Lion share was found to be broken somehow and the program was avoiding paying the jackpot. I doubt they would be forced to give back all the money they made over the years. If they were fined, it would be less then what the casino made on the machine. However, if the machine malfunctioned and a Jackpot was forced to hit, they may not have to pay it.
I'm not saying gaming is totally unfair, I have seen them rule for the players even, when I felt the players had no chance. I think Gaming just want everyone to be happy and not to create to much controversy. The higher profile the situation and the more people involved the more likely they are to rule for the players.
Quote: AxelWolfThis is the perception I got. Even in cases where the casino was clearly wrong its usually just a simple correct the situation. I bet if there was a gaff in a machine that affected the players (not done purposely). Gambling control would just say Fix it.
For instance, lets say a less known machine, but like Lion share was found to be broken somehow and the program was avoiding paying the jackpot. I doubt they would be forced to give back all the money they made over the years. If they were fined, it would be less then what the casino made on the machine. However, if the machine malfunctioned and a Jackpot was forced to hit, they may not have to pay it.
I'm not saying gaming is totally unfair, I have seen them rule for the players even, when I felt the players had no chance. I think Gaming just want everyone to be happy and not to create to much controversy. The higher profile the situation and the more people involved the more likely they are to rule for the players.
I agree with you on almost all of your points. In the industry, Gaff is generally used as a term for something intentionally put in to alter an outcome. If it was unintentional it would just be considered a bug or malfunction. I have dealt with Gaming control on six occasions. Three times I called them to dispute a pay on a machine and three times I have been called out as a manufacturer. All of the agents I dealt with were fair and trying to find an amicable solution for the player.
Quote: BuzzardGaming is nothing, if not anal. Colorado Gaming made the Isle tape a Discard Rack onto a Digital 21 game. Really.
I recall that. Apparently the Colorado Gaming Statutes said all Blackjack tables must have a discard rack. Clearly that statute was written before the electronic games.
bonus bet. Reduced HE for the player.
Quote: WizardAxiom is more confident about the 256 virtual stops than I am. I have a box full of par sheets and the total number of stops seems pretty arbitrary. Then again, most are by IGT, so I'm mostly getting a sense of their style. Lion's Share is my Mikohn, who I didn't even know ever made slots, until I saw this game.
In Axiom's defense, choosing random numbers from an even power of 2 was more of a programming style back when that game was made in 95 than it is now.
While I respect Axiom's opinion a great deal, and I'm not saying he is wrong, but I'm not so convinced Lion's Share is on a 256-stop reel map.
I hope ME stumbles upon this post as he knows slot programming style really well.
I have no idea how long the reels are in Lion's Share, though the MGM may be willing to part with the par sheet because the manufacturer is defunct and they'll take it away as soon as it hits. That might even be a good way to drive a bit of publicity, especially if the machine is currently sitting at, you know, 110% or something.
It may be possible to use math to ascertain the likely length of the reels. You can observe, presumably, the rate at which the progressive accrues relative to coin-in. Does anyone know what this is? Is it 1 penny per $3? Point is, you can take the progressive rate and the anecdote/assumption that it's never been hit since it started at $1M, and then determine how many pulls it took to get to the current jackpot. Based on that, you can then determine the likelihood of that happening under various reel lengths.
Quote: AxelWolfExamples? Other then well published incidents? I generally think casino are on the up and up, But i also think there are many violations going on that are unnoticed or are not penalized.
How about the million-dollar fine the Venetian got for rigging a drawing? They also gave the Golden Nugget a big fine (I forgot how much) with the owner deliberately let a player win in blackjack, as shown on their reality show.
Quote: MathExtremistIt may be possible to use math to ascertain the likely length of the reels. You can observe, presumably, the rate at which the progressive accrues relative to coin-in. Does anyone know what this is? Is it 1 penny per $3? Point is, you can take the progressive rate and the anecdote/assumption that it's never been hit since it started at $1M, and then determine how many pulls it took to get to the current jackpot. Based on that, you can then determine the likelihood of that happening under various reel lengths.
It goes up by 0.75% of money bet. Nobody seems to be sure where the meter started.
Quote: WizardIt goes up by 0.75% of money bet. Nobody seems to be sure where the meter started.
From posts like the one linked, I think it started at $1,000,000. If this is true and the progressive meter has remained fixed at 0.75%, the machine has been spun a minimum of 59.6M times. And if every bet was a max bet, it would lead to a likelihood of not being hit of about 2.864%.
http://www.vegasmessageboard.com/forums/showpost.php?p=785016&postcount=721
Quote: MathExtremistI have no idea how long the reels are in Lion's Share, though the MGM may be willing to part with the par sheet because the manufacturer is defunct and they'll take it away as soon as it hits. That might even be a good way to drive a bit of publicity, especially if the machine is currently sitting at, you know, 110% or something. It may be possible to use math to ascertain the likely length of the reels. You can observe, presumably, the rate at which the progressive accrues relative to coin-in. Does anyone know what this is? Is it 1 penny per $3? Point is, you can take the progressive rate and the anecdote/assumption that it's never been hit since it started at $1M, and then determine how many pulls it took to get to the current jackpot. Based on that, you can then determine the likelihood of that happening under various reel lengths.
ME, the meter has been reported to be 0.75%. That would be nine cents going into the meter per $12 wagered.
Quote: mickeycrimmME, the meter has been reported to be 0.75%. That would be nine cents going into the meter per $12 wagered.
The OP's picture shows the meter at 2.274 million. If the meter started at 1 million then 1.74 million has gone into the meter. With a 0.75% meter that would be about a $170,000,000 wager. At $3 a spin that would be about 57 million spins. But the problem is I don't think the amount of shortcoining can be quantified.
