Goblin Gamblin' Game - Mathematical Calculations

Cripes sake everyone. If you got an extra gold piece on you and you are waiting for everyone to assemble on TS docks for a raid to Cauldron Hollow in Nek, or a run through Zek or whatever. Whats the harm in pulling the old handle 10 times for a chance to win several plat. It is just a time and money sink in the game to give you something to do. I know for sure that if SOE audits the lotto and compares it to the active characters, they could quickly ascertain the plat farmer/botter from the real players.

It is a random game so your chances of winning are truly random. What the math states (as an anology) is that you have a 50% chance to get heads vs tails when flipping a coin. As chance would have it, does not work out that way.

I've noticed that the wheels change with every spin, not containing all numbers per wheel.
I've frap recorded about 30 spins. Slowed them down and each wheel definately doesn't contain every number.
So I'm guessing that you not only have to be lucky with the 36 per wheel, but the right wheels coming up good in the first place?
So how would that add to or take away from the odds in each spin?

I've been logging my plays with a home-brew program....

(note- i've carefully watched the jackpot, and determined that, on Butcherblock, exactly 3.7s per play gets added to the pool)

• Total plays recorded: 1481
• Dead losses: 1111
• 10s paybacks: 307
• 50s paybacks: 56
• 2g50s paybacks: 6
• 25g paybacks: 1
• Total money in: 14810 silver
• Total money out: 9870 silver
• Payback %: 66
• Contributed to jackpot: (3.7s per play) 5479.7 silver

Message Edited by linkedlinked on 10-20-2005 01:18 PM

You're right. The OP (and most people in this thread) have been calculating the odds incorrectly.

The odds of winning, say, 4, are *NOT*36/6 * 35/5 * 34/4 * 33/3 = 3,927 to 1

I've had a long, long line of math teachers from 8th grade through college that all ensured I knew how to *properly* calculate odds, and this is not the way.
(Pick 6 from 36) * (pick 5 from 35) * (pick 4 from 34) * (pick 3 from 33) = (odds of getting all 4 matches on four wheels, each with 36 numbers, and 6 "desired" numbers)

We don't have four wheels. We have six. It is necessary to divide all of the OPs answers by nCr( 6 , #{num matches} ).
Your web calculator does this, and gives much more reasonable odds- Odds that almost exactly match up with what my log file has shown so far; 2350 rolls and I have 9 match4's, 90 match3's, and 480 match2's.
Message Edited by linkedlinked on 10-21-2005 08:24 PM
Message Edited by linkedlinked on 10-21-2005 08:26 PM

As Rosanna-Rosannadanna says....

Nevermind.
Message Edited by Kitsilano on 10-21-2005 11:38 PM

Rosanna-Rosannadanna?
Man, you're old!
hehe...


oh wait.. it's Rosanne Rosannadanna...

That means I'm old... AHHHHHHHHHHH!!!!
Message Edited by BannocAB on 10-22-2005 09:18 AM

Hello all, all I have to say is you all are nuts,but you have got to be way smarter then I.:smileyvery-happy: I have read and re-read through each and every one of your math equations in this thread, and you know what? Hehe the more I read them the more confused I get. I come to one conclusion, I must have been sleeping during that day of math class in both High School as well as College. I only speak 1 language and math is definately not it. ROFL
If any of you figure out a guaranteed way to win the jackpot let me know, I'll even split it with you, 50/50. But that's as far as my math skills go. After all I passed Algebra 1 with a D - .:smileywink:

I won a 25gp prize after spending about 3gp. Looks like I beat the odds.

I love statistics, but enough people have beat it with a stick so I will keep this simple.

If you wager 10 plat in the Goblin Gamblin' Game, on average you will win back... drum roll... 10 plat!

This is true if and only if;

1. All money wagered stays in the pot, and the pot increases until won -- which is NOT true if pot resets after a server reset
2. The game doesn't skim off a portion of the money that is bet -- which we will not know unless SoE tells us
3. The game doesn't front a certain initial jackpot from Sony virtual coffers to get things going (e.g. a minimum 10 plat pot)
4. The gambling software works in a true random fashion

All the statistical 'factorial' calculations in previous posts are interesting, and some are even mathematically correct. Unfortunately, they don't predict the true average return on your wager.

Note: Anyone that truly believes that 10 plat in wagers will pay out an average of 10 plat in winnings is dreaming in Technicolor. Only Sony knows how much is actually paid out on average, and frankly I suspect even Sony doesn't know for sure. It certainly is less than 100% payout, and you can bet on that.

too.... much.... MATH! brain.... is ..... going..... to ..... EXPLODE! are you guys all math majors? maybe I'm just getting intimidated by the numbers but that looks confusing.

In reply to the above post, please note that a) the game skims off a large amt of the initial deposit, b) at server reset the pot is cleared.
Question: is the ncr formula used to calculate the odds in this case is ncr=n!/(n(n-r)!)? Probability has never been my strong suit and I always get the various formulae/permutations mixed up.

You can calculate all the odds you want and make as many prediction about how much you have to play/pay before you can expect a profitable return. You can't, however, get around the simple fact that each roll of the dice is an independent event and is not affected by anything that happened before and has no affect on what will happen in the future. That's the reason why lotteries and slot machines make money. The motivation to play is in the notion that any one pull of the lever will be THE pull that hits the jackpot. Your calculations try to predict that if you pay X amount then you should expect to see Y in return. In reality you can ay twice what you predict and never see a single copper in return.
One person had it right earlier in the thread. The only fool-proof way to predict the outcome is to not play. Every lottery ticket you buy has an equal chance of winning as well as an equal chance of losing. The only reason I play (with a preset limit) is because at that one moment that one ticket might be the one.

does anyone know what the cut is SOE holds back from the jackpot return? I know 100% of the money put in isnt going to jackpot. This game is simply made to pull money from the economy. I must admit it would be nice winning but I have already gave this game over 2 plat. Needless to say Im done but I woulkdnt of complained winning a 200 plat jackpot. Game is worthless and nothing but a sink if ya ask me. Not even fun considering you can get up on this thing. Ive tested this game with extensive plays and know by experience you cant walk away richer on this game. Unless you win jackpot . Or match 5 but I played over 2 plat and never matched 5. This is a one spin game if ya dont win leave. Its too bad SOE is always focused on taking away from the player community. For petes sake Sony when are you gonna turn that around.

you are all assuming that soe lotto code is actually 100% random...
anyone who uses lotto for loot distribution know that its not truely random... the other day in a full group i won the enire contents of 9 chests in a row.... other groupmates were mightily pissed off.

if they cant make it truely random for loot what make you think its random for lottery.... my guess is that its time based...

There is something to be understood here.

If the assumption is correct, that it is completely random. then there is absolutley no number that can define a chance of win. In randomicity there is no value for anything over 1. If there as 1 number, and only one slot to fill, i would have a 100% chance. anything over that has no number. so,

when: 1 play = 1 number and one slot = 100%

when: 1 play = (2+ number) and one(+) slot = undefined and undefineable percentage.

What you are asking is to put math into everything of our lives. You do not put a percentage on what a childs income will be in 20 years from birth, because factors are always unseen and any factor that is seen is always random as is the unseen. Therefore, when there is a completely random outcome no percentage can be attached. period.

All you people are calculating the odds of getting a jackpot and the chances of this and that and that you lose 94% of what you spend.
My thought: You cannot accuately calculate the REAL odds of winning. The jackpot is this: RANDOM. To closly define random: an event that does not follow a certain or set pattern and may happen at any time.
Random ia a thing that may not be set in math or science, but in living. You cannot win this jackpot without playing first. Yes, it is true that you may lose more money than win, but it is random so you may also win after playing one time. The numbers in the jackpot are random so they cannot be predicted but only waited on. I may play one time and win. I may also play until I die and NOT win, it is random and therefore unknown.
P.S. considering this thing can be calculated (which i believe to not be true) many, many people are playing at the same time, so your actual odds may be a little bit more in favor of winning.
Note: this post is not meant to be a flame, just a thought as to what i think and to add to this thread so please do not take offense as to what I type.
Another note: I think it would be cool if we could pick out our own numbers for the ticket. It makes me frustrated to constantly lose by the same numbers on my ticket.
--- notice a motif in my post: random ... my post revolves around the unique value of random. random is almost like the word infinate (sorry for spelling). Infinaty can be ANY number and technically random can also be anything, so to a certain extent they are the same word (mind you that i know the differances of these two words also).

In truth, you CAN calculate the "odds" of winning... The odds of winning don't necessarily mean that you will actually win that much, it just means that if you played an infinite number of games, your results would tend towards the "odds". In actual game play, with small sample sizes, your results will probably fall towards one side or another initially. Many people will play 10 games and win 7, and think the game is great... While many others will play 10 games and win 2, and think it's rigged. This is the nature of randomness in small sample sizes. For most purposes, the randomness in a computer system is good enough for this type of game...

Even if it's based on clock events on the server, there is no way to accurately predict the exact millisecond that a calculation will be done, using the time seeded random calculator, due to network latency, process latency, and various other factors. Just because you click once per second, on the exact second (Even if you use a program to time it exactly right), the calculations on the server will not happen at evenly spaced intervals due to these factors. For even better randomness, number generation can be accomplished with truly random events from nature...

Check out www.random.org for a random number generator that uses atmospheric radio noise to generate random numbers... They even have a nice description about how randomness works. We use their generator for picking winners on our web site for periodic freebie giveaways.

J

Edit: I would like to mention, we don't know what the "intentions" of the programmers were when designing this game... They may have designed it to be more or less "evil" by adjusting it's "odds" (rigging) based on the amount of coin in the economy.

Edit #2: For the person who won 9 loot lotteries in a row, Congrats! Lucky break That's exactly what can happen with small sample sizes! Too bad for the other sore losers though


Message Edited by Jhaerel on 11-09-2005 06:58 AM

Yes, I agree with you to an extent. What i really meant was that real randomness is unpredictable and yet people are saying that it is predictable (in a game nothing is truely random).
And thank you for telling me your thoughts with out it being a flame (many people have a hard time doing this).

I agree completely, real randomness is unpredictable. I however think that even randomness in a game is too unpredictable for anyone to come up with an "equation" to solve it all. I think what some of the OP's were saying was that if you invested "x" amount of coin in gambling in the game, you should expect a return of "y". It's never exactly that, but over time it will tend towards that if your sample size gets large enough. In short runs, the actual expected return of "y" is also random, you would have to invest a lot of time and platinum in finding any sort of true return, accounting for what the programmers have intended the game to do (Probably just intend for it to suck the excess coin out of the system).

As for flaming you, that's farthest from my mind. I'm not into fighting flame battles and calling names on forums. I'd rather have an intelligent discussion than resort to childishness!

Later,

J!