• Dear forum visitor,

    It looks as though you have not registered for a forum account, or are not signed in. In order to participate in current discussions or create new threads, you will need to register for a forum account by clicking on the link below.

    Click here to register for a forum account!

    If you already have a forum account, you can simply click on the 'Log in' button at the top right of your forum screen.

    Your Elvenar Team

Bogus percentages in events

SunsetDanar

Well-Known Member
I see an algorithmic table of somewhat level percentages for success in event play. Certainly NOT the success percentages shown in actual game play. Some folks suggest that success may even be tied to the time of day. Success or failure regarding a specific prize one may be trying for is really very ordinary. You get a lot of very ordinary instants in supplies and coins and how many of those do we have that are useless? Personally, I've used more than 2K of the moon thingies trying for the trader and another wishing well and have gotten neither one. The percentage for success on both in the large chest is 25%. I've had nothing close to that...In fact, my success rate for those prizes is 0%
 

samidodamage

Buddy Fan Club member
They are exactly the same on each roll.
Same logic applies to all probabilities. Just because there was a 25% chance of getting the prize on the first roll, doesn't change the fact that there's still a 25% chance of getting it on the 1000th roll.
The number of rolls in the sample just smooths out the extremes (rolled a '1' 75% of the time in 4 rolls; in 3000 rolls it would come much closer to 17 times and the 16.66...%.)
 

SunsetDanar

Well-Known Member
I agree. I associate this as a matter of conjecture and more related to the rules of logic rather than the rules of probability. In other words, folks whine about this often but they may not really know what it truly is that they are whining about. I'll suggest that we may both be able to write a true mathematical statement that would support the percentages that INNO publishes as well as a logistical one that denies them. The rules of probability include so many variables that it's possible to write them in such a way as to support your own argument (Such as perceived probability) so tables of probability are pretty much out the window. I've run algorithmic tables (Generic ones, though...I didn't write them) both with and without percentages and, with a very small sample size, the tables sans percentages come much closer to real time results than do the ones with percentages plugged in...Hence, my original comment.
 

SoggyShorts

Mathematician par Excellence
If you look at rolls one at a time, thinking about a 75% of not winning creates fewer surprises than thinking "I'll win 1 in 4".
I think the way one looks at the percentages is why in every event some sort of conspiracy theory emerges.

Having the "real" odds be different than the posted odds is just more work for inno with no gain. I see no motivation for them to lie.
For example, if the odds of winning a daily building was 20% and not 25%, what does inno gain by posting the wrong number?
Or looking at it another way: How would posting the truth hurt them?
 

DeletedUser3507

Guest
If you look at rolls one at a time, thinking about a 75% of not winning creates fewer surprises than thinking "I'll win 1 in 4".
I think the way one looks at the percentages is why in every event some sort of conspiracy theory emerges.

Having the "real" odds be different than the posted odds is just more work for inno with no gain. I see no motivation for them to lie.
For example, if the odds of winning a daily building was 20% and not 25%, what does inno gain by posting the wrong number?
Or looking at it another way: How would posting the truth hurt them?

Maybe people would quit buying shards or whatever...:);):) If they used the real numbers...:p
 

samidodamage

Buddy Fan Club member
Maybe people would quit buying shards or whatever...:);):) If they used the real numbers...:p
Back to giggling here, jp! Have folks stopped going to Vegas? Real numbers get interpreted in peoples' minds like *there's a 1 in 4 chance I'll win...whatever* instead of how the house looks at it *there's a 3 out of 4 chance you're going to lose...*
Having the "real" odds be different than the posted odds is just more work for inno with no gain. I see no motivation for them to lie.
Agreed. It's why casinos don't have to finagle anything to come out ahead with using straightforward probabilities. Way too much work and risk for way too little gain.
 

DeletedUser17455

Guest
People's intuitions about statistics are not great. Soggy has a good point, think about the odds of not getting a building.

Also, when analyzing your luck, remember to calculate the odds of the event you experienced. If the odds of getting no building after 1 roll are 0.75 then the odds of "not getting a building on your fourth roll" are 0.42, which means that almost half of people will not see a building even after four rolls. A third of people will not see the building on their fifth roll. This is why it's possible to have bad luck, because 1/4 is not a guarantee.

In passing this is also why game designers often skew the odds in favour of the players, because they have learned that human's have bad statistical intuitions and get pretty emotional if they don't hit the curve.
 

The Unbeliever

Well-Known Member
Quick question for you.
You are rolling a normal die. You roll 3 times in a row, and roll a '1' every time. What are the chances you roll a '1' on the 4th try?
...well, if it's ME, then it's definitely 100% I'll roll another 1, especially when it's for a Space Marine's armour save!:p
(hint: I'm awful at rolling dice - hence why I'm terribad at table top miniature games!lol.)
 

Pheryll

Set Designer
I see an algorithmic table of somewhat level percentages for success in event play. Certainly NOT the success percentages shown in actual game play. Some folks suggest that success may even be tied to the time of day. Success or failure regarding a specific prize one may be trying for is really very ordinary. You get a lot of very ordinary instants in supplies and coins and how many of those do we have that are useless? Personally, I've used more than 2K of the moon thingies trying for the trader and another wishing well and have gotten neither one. The percentage for success on both in the large chest is 25%. I've had nothing close to that...In fact, my success rate for those prizes is 0%

Let me add to that algorithmic table the following graph

UFQnycp.jpg


I usually give people the rough number 50 small chests is like a six sided die roll. I posted that chart in another thread which looked like a line versus a well centered and distributed curve. While a trained eye could see how close the two were by their alignment, I thought that in this case extending the data to 150 chests versus 3d6 would make it very obvious how the average and deviation of opening a large number of chests (a little less than you would acquire this event) ended up being very close to something that we expect to have different results when repeated, that is a die roll.

Many people see drastically different outcomes from one event to another and do not treat entire events like a small amount of data. Given the high standard deviation for prizes, one must expect the total from an entire event to be quite variable, and trying to get a single daily at any one slot should be accompanied by a decent amount of excess splinters just due to the deviation present.
 
Last edited:

DeletedUser12171

Guest
I, too, believe in magic.

Nah not really. The only "bogus" I believe are in the percentages, are rounding "errors" because decimal points are not represented. As someone mentioned elsewhere, one of the chests has percentages adding up to "101%"

Back in topic, it seems to me that @SunsetDanar is suggesting that there is some kind of system that doles out the prizes based on an unpublished set of rules. This means that even if the percentages are real as published, the dice roll is not random. Instead, the dice have memory of all previous rolls.

Now this is saying a bit more than simply that the percentages are not real. It implies that while the population data will match the published rates, individuals do not have equal chances at each prize, depending on uncertain and uncontrollable factors. The bottom line would be - prizes are determined every roll based on how many of each prize has already been won (hence, dice having memory). Other unpublished rules could be in play - i.e. rates change per those rules but ultimately all prizes won add up to the published percentage. Is that so? It is possible, but I really doubt it. How would such a system be better to Inno than a straightforward unbiased dice roll?
 
Last edited by a moderator:

DeletedUser4576

Guest
They are exactly the same on each roll. You have a 1 in 6 or a 16.666% chance of success each time. Not bad odds at all. In your scenario, the number of attempts is of no consequence, only the number of sides on the die.
You are confusing reality with software simulation. There is no actual die in this game. The payoff is whatever the programmer wants it to be. It is no closer to reality than the electronic slots in Las Vegas.
I see an algorithmic table of somewhat level percentages for success in event play. Certainly NOT the success percentages shown in actual game play. Some folks suggest that success may even be tied to the time of day. Success or failure regarding a specific prize one may be trying for is really very ordinary. You get a lot of very ordinary instants in supplies and coins and how many of those do we have that are useless? Personally, I've used more than 2K of the moon thingies trying for the trader and another wishing well and have gotten neither one. The percentage for success on both in the large chest is 25%. I've had nothing close to that...In fact, my success rate for those prizes is 0%
This is a virtual world, not reality. There's no actual die or spinning wheel that follow the laws of physics. The payoff is whatever the programmer wants it to be, just like the electronic slots in Las Vegas. From observation it seems to me this game is giving away more "consumable" items like instants and buildings that have limited life spans. I think it's fairly obvious items of value like sectors and excellent buildings are being rationed to prolong the money earning lifespan of Elvenar. Doesn't mean it's no fun to play, just that Events are less rewarding and if you want the good stuff, bring diamonds.
 

Pheryll

Set Designer
This is a virtual world, not reality. There's no actual die or spinning wheel that follow the laws of physics. The payoff is whatever the programmer wants it to be, just like the electronic slots in Las Vegas.
From observation it seems to me this game is giving away more "consumable" items like instants and buildings that have limited life spans. I think it's fairly obvious items of value like sectors and excellent buildings are being rationed to prolong the money earning lifespan of Elvenar. Doesn't mean it's no fun to play, just that Events are less rewarding and if you want the good stuff, bring diamonds.

I am not sure how the first quote relates to the second, even though they are from the same paragraph. First, yes, a simulation of randomness is being used, as true randomness is impossible to obtain, but that does not mean the simulation is any further off from a random outcome than a die roll.

For the second statement, yes, a rationing is going on by making less of the pure culture available and placing more consumables in the standard offering. This does not mean that it is less likely to get a permanent on its day than a consumable on its day. Rather, before we had an offering of nearly all pure culture and population and culture, and now we have many other new productions added on.

As anecdotal evidence seems to be what is in vogue for offering "verified" conclusions, let me enter my anecdotal evidence which runs against the grain of the random number being tinkered to ration. Going through this event without spending diamonds, I was able to obtain 6 shrines of Kirit and 10 shrines of Krarak, giving myself a total of 16 dailies in about 150 chest openings (1 in 17 chance of doing that well or better). When I spent over 500 splinters for the consumable trading outpost the next day, I got zero and used up all my splinters. The good anecdotes like this are often the ones not spoken about, and the bad luck ones (like jps' 2100 splinter investment that returned nothing, a 1 in 400 chance) are touted immensely which often causes the "verification" to run towards bad luck, as that is what is being discussed the most.
 
Last edited:
Top