What Kind of Chance?

[DeletedUser8152]

Why aren't players occasionally winning 2 pots with the one Double Chance shot?

Why isn't the pot increased from the extra missed shot?

Fair questions, but obviously you'd have to code it smart enough to distinguish a double shot from two single shots. Doesn't seem like it would be too hard.

Cruddy programming or design? Would INNO intentionally screw players this way?

What makes you feel like this would be screwing the players?

 

[Algona]

Again, stressing this post is strictly in regards to the speculation that the Double Chance is actually two shots.

What makes you feel like this would be screwing the players?

Folk are taking two shots, one may hit but one of the two is always missing: The pot ain't increasing from the missed shot.

 

[DeletedUser20367]

That still does not explain the remarkable number of times that a player new to the bidding wins the prize with only one shot.

Repeating what I wrote above ... I have observed many times that Player 1, failing to win a prize despite dozens of attempts, is followed immediately by Player 2 winning the prize with one attempt.

The more I play this contest, the more it seems that the deck is stacked, so to speak. The speculation about what is "behind the curtain" is interesting, but I still don't have a definitive answer to my question. An answer from one of the "wizards" would seem most appropriate.

InnoGames is being far too casual in its use of the term "chance" here. If the "double chance" is actually two chances with the same odds — rather than the common perception of doubling the probability of a single chance (since only a single shot is taken) — then I would say that players are being deceived, even if the deception were not intentional.

I've been Player 2 a few times (3 I think). I've also had several prizes snagged by others. Once or twice they only put in a shot or 3. You can kind of tell by watching the prize counter go up how many shots other people are taking. So far my 6 worlds are averaging about the same. Two days ago one was seriously behind but has caught up. I've seen no clear evidence of anything but random luck.

 

[DeletedUser14197]

Like Jenny, I don't see how you can determine when this happens. I can't say I've seen any evidence of it.



Perhaps I am not understanding, but if I do understand, it happened to me. I sunk a bunch of balls into the 5% pot and while I was yet shooting someone came on and shot once, twice at most and won the pot. Why can't one determine this. You see how much the pot goes up with every shot and you see it go up when you don't shoot. Am I not understanding what you are saying cause it just seems easy to determine if you are paying attention to how much the pot is going up.

 

[DeletedUser26154]

Once you have been designated a "loser" by the program, it doesn't matter how many shots you take

Having it bad: Being called a loser in life.
Having it real bad: A computer verifies it.

 

[DeletedUser8152]

Perhaps I am not understanding, but if I do understand, it happened to me. I sunk a bunch of balls into the 5% pot and while I was yet shooting someone came on and shot once, twice at most and won the pot. Why can't one determine this. You see how much the pot goes up with every shot and you see it go up when you don't shoot. Am I not understanding what you are saying cause it just seems easy to determine if you are paying attention to how much the pot is going up.

I suppose it is true that if you start a pot from the beginning and pay attention, then you can be sure that only you have taken shots on it. And then if you stop but keep watching and the next shot wins, then you do know. But are you really in that situation often enough to make statistical claims about it?

 

[lemur]

I bet then that the 2x button just makes your next shot count as two attempts at once.

I did something tonight that I should have done a few days ago. I read the Announcement of the Soccer Cup again. Here is a quote from that announcement:

I've done the 20% shot 5 times, but did not win Cups. Are the chances wrong?
The chances shown for any shot are not cumulative, but individual. Meaning you have a 20% chance to win every time you take that shot.
[green emphasis is mine]​

That looks to me like classic random trial probability — which further invites skepticism about the programming, given the profusion of high scores that I described above.

Another quotation from the Announcement by DarkArtist69 ...

... which building you receive is purely random based on the displayed odds. But even with randomness, the same scenario can repeat several times.​

Yes, indeed, the word "random" was used.

I used the 'Double Chance' item and won cups. But the amount of cups I won was not doubled?
The 'Double Chance', as the name says, doubles the chance for you to get a winning shot. If the original chance for that shot is 20%, a double chance will increase it to 40%. However, it does not affect the amount of cups in any way, so you do not receive double cups.​

 

[DeletedUser8152]

Yeah, but it doesn't really mean anything to apply a statement like "a 20% chance to win every time" to a single trial. A single trial you either win or you lose. Me personally, I'd be comfortable using that language for any situation where, after a large number of trials, you can expect to win 20% of them, and you have no way ahead of time to predict which 20% they will be.

Just for an example, the weatherman says there's a 20% chance of rain tomorrow. That doesn't mean that the probability of rain has to follow a Poissonian (single-shot random) distribution, I know for a fact that it doesn't.

 

[lemur]

... a Poissonian (single-shot random) distribution.

I don't understand what you're trying to say. How could a Poisson Distribution be applicable? What would you label the x-axis and the y-axis?

 

[DeletedUser8152]

Collect many samples of say 100 days, all with 20% chance of rain predicted. X is the number of days with rain, Y is the number of samples that had that many days of rain.

 

[wolfhoundtoo]

You should note that long odds do occur. I don't see how you have enough data from just your hoods to say that large jackpots happen to often. Simply the fact that it's 1 in 6500 does not mean that event could not of happened. You sample isn't sufficiently large enough to draw conclusions and your 'someone new' wins it pretty much means you didn't win and whomever won it must of won in 1 shot which is what you think because you didn't win. yes i've had that though a couple of times when i ran thru shot packs and didn't win but I thought better of it.

One more thing to remember about the large jackpots: they tend to occur when more people are on as more take shots. Then it gets larger and more players become aware and join the 'rush'.

 

[DeletedUser13838]

As has been pointed out before, large jackpots should occur very very rarely assuming that each trial is independent, it doesn't matter when people are online. With the assumption of independence, the probability of n consecutive misses followed by a success is p(1-p)^n. Being very conservative and assuming no one uses the double chance, the probability of 100 misses followed by a success (which should result in a jackput of nearly 1600 cups) is .05*(.95)^100 = 0.03%. In reality the probability would be much lower. Yet this does not seems to be uncommon in my hood (I had the weekly high score of 1555 cups until someone beat it with ~1990 cups).

But instead of arguing against the coded probabilities, I'm suggesting that the assumption of independence is what's wrong.

 

[DeletedUser25273]

Konrad, that low odds is the odds of getting EXACTLY 100 misses followed by a goal. If you change it to at least 100 misses, the value goes up to (0.95)^100 which is about 0.59%, With a full neighborhood of 80, there are 1920 shots earned a day (80*24), or 13440 a week. Higher when you include the shot packages earned from quests. While not every shot is going to be done at the 5% goal, once the value builds up, a lot of them are, so you expect perhaps several of these fraction of a percent events a week. (and with all the neighborhoods and worlds, we are going to see some much less likely events somewhere). This doesn't prove that the the chance of a shot is independent, but does say that casual evidence is going to be unable to disprove it. Perhaps if someone took careful nots of all winnings on the scroll bar, and a rough idea of the lower odds winning, we could do an analysis on the upper tail of the 5% to see if it is far enough from the expected exponential decay to make it unlikely that it is a pure 5% independent result. I doubt that anything short of that would be able to make a strong argument against the presumptive method.

 

[DeletedUser8152]

I can say that the weekly prize for the the most cups usually seems to go in the 1100 to 1600 cup range, in my experience.

But surely once the prize is above several hundred, nearly everyone is use the 2x button.

 

[Algona]

I can say that the weekly prize for the the most cups usually seems to go in the 1100 to 1600 cup range, in my experience.

But surely once the prize is above several hundred, nearly everyone is use the 2x button.

First week in my hood was 1650, second was 1350, current is 1975.

I only use the 2x button on my last 5 shots, and stop calling me surely.

 

[DeletedUser13838]

Konrad, that low odds is the odds of getting EXACTLY 100 misses followed by a goal. If you change it to at least 100 misses, the value goes up to (0.95)^100 which is about 0.59%, With a full neighborhood of 80, there are 1920 shots earned a day (80*24), or 13440 a week. Higher when you include the shot packages earned from quests. While not every shot is going to be done at the 5% goal, once the value builds up, a lot of them are, so you expect perhaps several of these fraction of a percent events a week. (and with all the neighborhoods and worlds, we are going to see some much less likely events somewhere). This doesn't prove that the the chance of a shot is independent, but does say that casual evidence is going to be unable to disprove it. Perhaps if someone took careful nots of all winnings on the scroll bar, and a rough idea of the lower odds winning, we could do an analysis on the upper tail of the 5% to see if it is far enough from the expected exponential decay to make it unlikely that it is a pure 5% independent result. I doubt that anything short of that would be able to make a strong argument against the presumptive method.

That's a good point. However, we are looking for the likelihood of sequences with, say, 100+ misses following a success.

For |p|<1, Σp(1-p)^n = 1 and P(p=5% and n>100) = 0.56% <-- assuming no one uses double chance

After the number of cups reaches 400 (22 misses), the value is in paying 20 cups to double the chance. If even half of those excess shots are at double chance the probability of a sequence of 100 misses following a success is 0.07%. So if all 13440 shots are used on the 5% target, and 5% are successes (on average) that is 672 +/- 50 successes of which 0.07% (about a half) should be followed by a sequence of 100+ misses. Even if you assume that nobody uses double chance, you should only see about 3-4 such jackpots in a week and that requires a lot of really unlikely assumptions to be true.

 

[lemur]

After the number of cups reaches 400 (22 misses), the value is in paying 20 cups to double the chance. If even half of those excess shots are at double chance the probability of a sequence of 100 misses following a success is 0.07%.

I get the same result: 0.95^61 * 0.9^39 = 0.000719

Even if you assume that nobody uses double chance, you should only see about 3-4 such jackpots in a week and that requires a lot of really unlikely assumptions to be true.

One assumption would be that 13440 shots are actually collected. Given the amount of inactivity (especially during the summer), I would estimate the number collected to be only about half of the maximum — due to untimely collection, and due to players abandoning the event altogether. Every day, on multiple worlds, I also see a large fraction of the shots being taken at goals other than the 5% one. On worlds where I have been unlucky, I have switched to a strategy of simply converting all shots into 10 cups with the sure shot — so that I get at least one big prize out of this event. Many of my friends are doing the same thing.

But the parade of extreme jackpots continues. On Jaims this morning, I won a jackpot of 2395. That is the result of 153 misses.

I'm suggesting that the assumption of independence is what's wrong.

I have suspected this since a few days into the event. If that were true (lack of independent trials), then how could the stated probabilities be truthful?

Consider this extreme example. A Las Vegas casino advertises a "5% chance" to win a jackpot at one of their gambling tables. But rather than offering chances that are independent, the operators simply divide the number of attempts by 20 and award the jackpot that many times to one of their friends. How long would the Nevada Gaming Commission allow them to operate?

 

[DeletedUser8152]

I have suspected this since a few days into the event. If that were true (lack of independent trials), then how could the stated probabilities be truthful?

Because nothing in the concept of probability demands that individual events be independent.

Consider this extreme example. A Las Vegas casino advertises a "5% chance" to win a jackpot at one of their gambling tables. But rather than offering chances that are independent, the operators simply divide the number of attempts by 20 and award the jackpot that many times to one of their friends. How long would the Nevada Gaming Commission allow them to operate?

In that case the results are not random, in that they are predictable ahead of time. If you have figured out a way to predict the results of the soccer event, then you probably should stop posting and just revel in the wealth instead

 

[lemur]

Because nothing in the concept of probability demands that individual events be independent.

If "5% chance" does not refer to the probability of an individual event, then to what does it refer?

 

[DeletedUser13838]

Suppose there are 20 tickets in a hat, 1 of which says "winner". You and 19 other people each draw 1 ticket from the hat at random.

What is your chance of picking the winning ticket?

Suppose that before picking your ticket, you saw that the person picking before you did not pick the winning ticket - what is your chance of picking the winning ticket now? Does this in any way change your answer to the first question?