If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Local6)   Woman pulls off trillion-to-one shot, hits second $1 million jackpot on same machine (with pic)   (local6.com) divider line 137
    More: Unlikely  
•       •       •

28801 clicks; posted to Main » on 10 May 2005 at 5:01 PM (9 years ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



137 Comments   (+0 »)
   

Archived thread

First | « | 1 | 2 | 3 | » | Last | Show all
 
2005-05-10 07:16:02 PM
This "penny" machine requires I think a bet of 200 credits per spin. So you're actually spending a pretty steep two bucks a spin. I'd guess this woman spent a LOT of her million winning the second one.

In general, slots can be set anywhere from 85% to over 100%. The ones with the big prixzes are all progressive slots and work like a lottery, with some of the money bet going into the jackpot pool. On these, payout percentage is fixed, they can't just decide to change the odds on a whim.
 
2005-05-10 07:17:20 PM
FlyingLizardOfDoom: The NGC requires a rough 98% payoff on all slots


That is incorrect. Nevada Law requires at least 75% payout. Anything over that is at the discretion of the casino.
 
2005-05-10 07:17:37 PM
I heard about this. The guy who wrote the article is a dumbass. She did not hit a jackpot on the same "machine", she hit a jackpot on the same "game". She hit the jackpot at 2 different casinos. Millionisers are all over the place, each casino in NV probably has at least 4 of them. Besides even if she did hit it on the same machine her odds would not go up to 1 in a trillion they would stay exactly the same. Your odds on a machine do not go down once you hit a jackpot, you can hit them back to back. The random number used to determine a jackpot is not removed from the main number database if it gets used, it remains in the number set. Temperature and time of day does not affect how a machine hits either, the random number set is always the same size.
 
2005-05-10 07:21:32 PM
Deveyn:

That is incorrect. Nevada Law requires at least 75% payout. Anything over that is at the discretion of the casino.


They do have "certified 98% payout" machines. For some reason that section always seemed to be empty. That and the $100 slot section (yikes).
 
2005-05-10 07:22:06 PM
The Lottery is math tax. Playing slots can be a cheap fix, although I personally, abhorr...

ooo. /**opening wallet*/ shiny.
 
2005-05-10 07:22:36 PM
2005-05-10 07:17:20 PM Deveyn


FlyingLizardOfDoom: The NGC requires a rough 98% payoff on all slots


That is incorrect. Nevada Law requires at least 75% payout. Anything over that is at the discretion of the casino.


Read last sentence. I was making a mockery of statistics, not quoting lagit ones.
 
2005-05-10 07:29:08 PM
IBelieveYouHaveMyStapler:

Besides even if she did hit it on the same machine her odds would not go up to 1 in a trillion they would stay exactly the same. Your odds on a machine do not go down once you hit a jackpot, you can hit them back to back


The odds of hitting it don't go up but the odds of hitting it twice are much higher than hitting it once as you have to multiple the probability of each event. Two specific 1% events happening would be .01 x .01 or 1:10000. I have to deal with this on a regular basis when calculating the failure probability of a redundant system given the failure probability of each redundant component.
 
2005-05-10 07:33:10 PM
IBelieveYouHaveMyStapler:

Your odds on a machine do not go down once you hit a jackpot, you can hit them back to back.

Your odds of hitting a second jackpot given that you won the first are the same as your odds of hitting the first, but the odds of winning 2 jackpots will certainly be on the order of a trillion-to-one shot.

Temperature and time of day does not affect how a machine hits either, the random number set is always the same size.

Actually, there probably are machines out there that use each of those parameters you mentioned in an attempt to come up with actually random (i.e. not pseudorandom) numbers.
 
2005-05-10 07:39:34 PM
FlyingLizardOfDoom

The NGC requires a rough 98% payoff on all slots...

Somehow I doubt that.
 
2005-05-10 07:46:00 PM
LineNoise

Best i ever did in one bet was a royal flush on a dollar video poker machine.

Damn thing paid me in dollar coins. After lugging them all to the window for cash, I headed straight to the bar.


No you didn't fnorgby.
 
2005-05-10 08:17:49 PM
LineNoise

Best i ever did in one bet was a royal flush on a dollar video poker machine.

Your internet life is awesome.
 
2005-05-10 08:26:51 PM
A woman from Boulder City, Nev., pulled off what gambling experts said had to be a trillion-to-one shot: she hit a $1 million jackpot, twice, on the same slot machine

The experts are idiots. This is an independent event. The odds that you will win or lose are not influenced by the previous outcome. Consider this example: The odds that you will win the lottery is 1/5. You win on monday. You play again on Tuesday. The odds you win on Tuesday is 1/5, not 1/25. Each day, the odds are always 1/5. Just because you won on Monday does not mean your chances of winning from then on are lower than 1/5. How would they be? Every day the odds are the same. You just happen to be lucky if you beat the odds every day. Furthermore, think about this over the course of a year. With such low odds, you're bound to win multiple times. If you win two days in a row over the course of 10 days, that is no different than winning on day 1 and day 10.
 
2005-05-10 08:30:49 PM
evaned:

Your odds of hitting a second jackpot given that you won the first are the same as your odds of hitting the first, but the odds of winning 2 jackpots will certainly be on the order of a trillion-to-one shot.

No. If you play the slots long enough, you'll hit the jackpot twice. Hitting it twice in a row is the same as hitting it with 1000 no pay off pulls in between two wins. Think of it this way: If you flip a penny ten times, chances are it won't come out heads 50% of the time. But if you flip it 1000 times, it will be extremely close to 50% (in fact, you can easily calculate how many times you need to flip the penny before it stabilizes at 50% -- I just don't remember enough from Stat to give you the equation).
 
2005-05-10 08:35:30 PM
WOULD NOT HAPPEN! COULD NOT HAPPEN!
Wanted for questioning.
 
2005-05-10 08:38:59 PM
no matter how many times you flip the coin, (not factoring in chaos theory), its always 50% either heads or tails... even if you hit 10 straights heads, it'll still be 50% chance of heads or tails on the next flip...

i don't know how this applies to crooked old ladies on fixed slot machines.
 
2005-05-10 08:48:37 PM
T.rex:

no matter how many times you flip the coin, (not factoring in chaos theory), its always 50% either heads or tails... even if you hit 10 straights heads, it'll still be 50% chance of heads or tails on the next flip...

Exactly. So if the odds are 1 million to one, and you play 2 million times, the odds of back to back jackpots are the same as hitting it on the first pull and 2nd million pull.

I you people are really having trouble grasping this, take out a coin, flip it ten times, record the number of heads and tails, flip another ten, take down the data, then another ten, and so on. Notice how the longer you play, the number of heads and tails evens out to 50%.
 
2005-05-10 08:49:25 PM
thornhill

You win on monday. You play again on Tuesday. The odds you win on Tuesday is 1/5, not 1/25. Each day, the odds are always 1/5. Just because you won on Monday does not mean your chances of winning from then on are lower than 1/5.

But what are the odds of winning on BOTH Monday AND Tuesday? 1/5 * 1/5, or 1/25.

They are giving the odds that she won two rounds, not that she won the second.


No. If you play the slots long enough, you'll hit the jackpot twice.

With increasing probability, yes, but not with probability 1. Ever. No matter now much you play. (Assuming a perfect random number generator.)

Hitting it twice in a row is the same as hitting it with 1000 no pay off pulls in between two wins.

To clarify, hitting 1000 no payoff pulls then two wins (or two wins then 1000 losses, or...) is the same as getting a win, 1000 losses, and another win.

'Win, win' is more likely than 'Win, (1000 losses), win'. But anything with 2 wins and 1000 losses will have the same probability.

Think of it this way: If you flip a penny ten times, chances are it won't come out heads 50% of the time. But if you flip it 1000 times, it will be extremely close to 50% (in fact, you can easily calculate how many times you need to flip the penny before it stabilizes at 50% -- I just don't remember enough from Stat to give you the equation).

What does this have to do with anything?

One of us is misunderstanding the other I think.

All that is relevant is that if the probability of hitting the jackpot on any given pull is p, the probability of hitting it twice is p^2 if the pulls are independent. If the probability of a win on any given pull is a million to one against, the probability of getting two wins on two pulls will be about a trillion to one against. The article doesn't say how many games she played, so reduce that factor roughly by the number of games.
 
2005-05-10 08:56:33 PM
2005-05-10 07:33:10 PM evaned Actually, there probably are machines out there that use each of those parameters you mentioned in an attempt to come up with actually random (i.e. not pseudorandom) numbers.
I work on these machines every day, the random number set is not depedent on any outside forces. Gaming Control would freak out if there was a machine that would "get hot" if it actually got hot.
thornhill You understand it...almost too well.

Jackpots are just basically standard deviations in a number set, hitting one does not make the machine "cold" and someone pouring in hundred after hundred into it does not make it "bound to hit soon". Sorry.
 
2005-05-10 09:02:22 PM
Gaming Control would freak out if there was a machine that would "get hot" if it actually got hot.

That's why you'd want to use the low-order bits.


Anyway, I'll admit I'm wrong if I am, but perhaps you can give a couple examples of things that are used as seeds for the random number generator?
 
2005-05-10 09:04:08 PM
Or are they seeded once at the factory and not again?
 
2005-05-10 09:26:08 PM
 
2005-05-10 09:54:38 PM
evaned:

One of us is misunderstanding the other I think.

Yes, you :)

This is a conceptual issue that logically doesn't seem right. Going back to my lottery example, Monday and Tuesday are apples and oranges. You cannot compare them. They are to totally different events. You only do the math you did if you need to win the lottery two days in a row in order to receive the money.

The pennies come in because most people fail to grasp the idea because they do not think about it if a person went on playing the lottery, slots, infinitely.

You flip a Penny 10 times. The odds that it lands:

HHHHHHHHHH

are the same as:

HHTHTHTTTH

The second outcome looks more random, but it is not. The second outcome looks more random, but it is not. There are 2^10 different permutations; each occurs only once out of 2^10 films. But you're going to get more permutations that have a random appearance like the second than the first. Getting 5 H's does not make it more likely that the next flip will produce a T because so far you've gotten 100% H. Each time you flip, the out come is 50/50. If you play a million times, and total up the H and T's, it will be about 50/50 perfectly, and you'll have runs that look like the first example as well as the second. But if you look only at a ten penny run out of the million flips, then you don't have 50/50.

The fundamental issue is that back to back wins look less random than spread out wins, but they are not.
 
2005-05-10 10:06:49 PM
evaned, I'm not sure what you mean by seeds. I wasn't ignoring you I had to go somewhere. I'm back now for a little while.
 
2005-05-10 10:27:22 PM
Yes, you :)

Ah, I would say you. ;-)

They are to totally different events. You only do the math you did if you need to win the lottery two days in a row in order to receive the money.

And I'm finding the probability that both events occur. This is legal, and makes perfect sense.

You flip a Penny 10 times. The odds that it lands:

HHHHHHHHHH

are the same as:

HHTHTHTTTH


Only because the probabilities of getting a tail and a head are the same.

If the probability of getting a head on any given throw is p, then the first has probability p^10. The second has probability p^5*(1-p)^5.

The second outcome looks more random, but it is not.

Depends on how you define randomness. This is of acedemic interest here, because in this context, that's right.

Getting 5 H's does not make it more likely that the next flip will produce a T because so far you've gotten 100% H. Each time you flip, the out come is 50/50.

I know that. I'm very well aware of independence, believe me.

The fundamental issue is that back to back wins look less random than spread out wins, but they are not.

I still don't see what the order of the wins and losses have on anything. The only relevant info present in the article is that she won two games. The only other relevant information is the number of losses, which isn't in the article. Assuming the number of games is relatively small, the probability will only be increased by roughly the number of games.

For instance, the probability of getting two wins in a 365 game set (one a day for a year) where p is the probability any given game is a win is:
(365 choose 2)*p^2*(1-p)^363.
If p is 1 in a million, the above is about 1 in 15 million.
If p is 1 in 2 million, the above is about 1 in 60 million.

If you decrease it so that only two games -- the two wins -- are played, then the big formula simplifies to p^2. If if the odds of a single win are 1:1,000,000, then the odds of two wins is 1:1 trillion.
 
2005-05-10 10:40:39 PM
evaned, I'm not sure what you mean by seeds. I wasn't ignoring you I had to go somewhere. I'm back now for a little while.

The "random" number generators in (modern, electronic) machines are algorithmic and only pseudorandom. This means that when they are asked for a random number, they use an mathmatical formula to calculate it based off of the previous "random" number (or more than one back, but all I know only use the previous). This is because true random numbers require specialized equipment, such as the radioactive isotope and geiger counter someone mentioned.

But you have to start the algorithm somewhere. The first number you use is called the seed. Most computer games and whatnot will use the system clock as the seed, because it's random enough. However, the impression I have of casino slots is that they use very strong seeds. To generate these, information is culled form lots of places.

For instance, another place random numbers are critical is in cryptography. PGP makes the user type lots of random stuff or move the mouse around randomly when generating a new cryptographic key as a source.

This is the place where the temperature would be used. Take, say, just the part of the temperature to the right of the decimal place and work that in as part of the seed. If the random number generator is seeded when it is something.65 degrees the result will be vastly different than something.66 degrees.

The thing to note is that you need not seed the RNG every throw, only when it is first turned on. However, it might be possible to discern a pattern and thus figure out the algorithm being used if this was done, so I'm guessing that it's reseeded every few throws. (Otherwise, knowing the algorithm and the current state would allow you to figure out the next throws exactly.) This would make it much more difficult to find the algorithm.

However, the more you seed, the more important a strong seed becomes, and thus my lack of doubt at the use of temperature sensors.
 
2005-05-10 10:57:48 PM
Didn't read the entire thread, and was disgusted before I made it as far as I did.

Anyone who claims to use their last five dollars to gamble and wins big is a plant.

Quoting Pen and Teller "Gambling is a celebration of bad math".

Gambling lives off of those who can't afford it and deserve better.

It should be clear to everyone since ESPN started televising poker as a sport that casino's desire new (more rubes) to willingly and stupidly give them their money.

Unless it makes you happy, don't do it
 
2005-05-10 11:07:00 PM
evaned:

And I'm finding the probability that both events occur. This is legal, and makes perfect sense.

Ok, I'm going to give up after this.

What you're not understanding -- and I'm not trying to sound condescending -- is that there is a difference between consecutive and two-in-a-row.
 
2005-05-10 11:07:51 PM
I don't know what is used as a seed starting point for the number set and it varies by the machines manufacturer and possibly even the game theme. A lot of the higher level math the slot machine manufacturer does not want you to know because it is a trade secret and they are afraid of cheats building emulators, etc. etc.
 
2005-05-10 11:09:10 PM
bye thornhill.
 
2005-05-10 11:11:04 PM
forest25:

Gambling lives off of those who can't afford it and deserve better.

It should be clear to everyone since ESPN started televising poker as a sport that casino's desire new (more rubes) to willingly and stupidly give them their money.


Exactly. I have a friend who actually does make quite a bit of money from playing poker, but that's because he was an econ major, and understands a lot more about math and probability than most people playing. He says that 50% of the folk who play online or at casinos have no farking clue what they are doing. 25% are a little better, but still pretty clueless. It comes down to being able to identify good players and knowing to avoid playing against them.
 
2005-05-10 11:23:54 PM
I came to Vegas once with $70 in my pocket (plenty of more cash in the ATM) and went home $11K richer -- that was interesting.

A friend of mine won $120K (with about $30K of seed money) in Vegas and got the shiat comp'd out of him by New York New York for about two years.
 
cot
2005-05-10 11:44:13 PM
I don't like to gamble, but I also don't necessarily buy the "tax on those bad at math" idea. Sure, many people have problems with gambling, and some have unrealistic expectations of winning big, but if I really thought it was fun, I would have no problem saying "I'm going to take $50 to the casino and see where it gets me"

Hell, I like video games, and I can't tell you how many quarters I must have pumped into them in my lifetime, but I'm pretty sure that video poker pays off better than Tron.
 
2005-05-11 12:06:11 AM
What you're not understanding -- and I'm not trying to sound condescending -- is that there is a difference between consecutive and two-in-a-row.

Can you please say what that difference is?

Especially because no one used the word 'consecutive' before there?
 
2005-05-11 01:20:14 AM
BERNOULLI!
 
2005-05-11 09:59:09 AM
Even if you knew exactly how a machine generated the seed, you still wouldn't be able to predict it, it's way too complicated to emulate or predict.

As someone mentioned before, many machines use the moment the player presses the button as the random event. This IS sufficiently random, and the result is NOT dependent on the previous spin. For all intents and purposes, the result is the same as spinning a many sided die.

A second win is unrelated to the first, winning once doesn't help or hurt your chances of winning a second time. But the odds of winning twice ARE much higher than winning once, regardless of whether that's two in a row or two in a year.
 
2005-05-11 01:07:05 PM
evaned: Anyone care to explain why gambling is a tax on people who are bad at math but insurance isn't? Or does everyone who views casino gambling as a bad idea not have any homeowners', life, etc. insurance? (Auto is another thing because it's required)

Buying insurance is essentially paying a fee to be shielded from catastrophic loss or expense. For instance, when I buy medical insurance, I know that typically I will pay more in premiums than the company will pay out in benefits. However, if I should need treatment costing hundreds of thousands of dollars, I will be able to get the treatment without bankruptcy.
It's for this reason that dental and eyeglasses plans have never seemed to make much sense to me, unless subsidized by an employer, in which case they are not "insurance" but "a benefit".
 
2005-05-11 05:22:06 PM
This article is inacurate, as tv reports often are due to broadcast journalists being lazy and sloppy. She did not indeed hit the jackpot on the same machine. Same game, millioni$er, but not the same machine. Her first jackpot was at Sunset Station in Henderson, NV last June.

Just thought I'd clarify that for ya.
 
Displayed 37 of 137 comments

First | « | 1 | 2 | 3 | » | Last | Show all



This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report