The reality of probability...don't let your brain warp it

[Xenosaur]

Let's make it simple. Throw a coin in the air 2 times. What if comes up heads both times, are ya going to say the coin is fake and has no tails? Of course not.

How about throwing it up 5 times, and get 4 heads in row? Is that going to keep that bee in your bonnet about the coin being rigged and now incorrect probability of 50/50? Of course not again.

Ever seen a run of 18 Reds in a row on a legitimate roulette table? Sure. We have all seen long duration Red/Black runs. Makes you want to bet on black next time, right?

Well, maybe not. Red has the same chance to come again after 18 reds...as a black... hope you know that. Past results have nothing to do with future events when it comes to probability in cases like this. They're independent events and not related. But your brain can get the better of you if you let it.

Time and tries evens out probability to it's true value. In these cases, the coin will eventually flip tails maybe directly after 2 or 4 heads in a row. Another red or even a black will come up in roulette after 18 reds in a row.

But over time, LOTS of TIME - the universe demonstrates the true probability of an event. Same with Inno's buildings. You just have to believe that. If you don't any game is a very slippery slope. Even real card playing.

Capture some data (tally your daily outcomes...) on what you're getting and you'll see that. Just give it some time to form for you.

PS: Probably not best to hear about a coin flip landing on the coin edge. It's entirely possible, though. A computational model suggests that the chance of a coin landing on its edge and staying there is about 1 in 6000 for an American nickel.

 

[Ebeondi Asi]

The mention of cards. nearly everyone has played cards.and it is clear that variability s huge regardless of statistics.
For most Foe probabilities this is all true.
However there are clear problems and have been problems in the past.
One in GbG with some areas of the map where three SC should be working it was 'as if' only one was working. That seems to have been fixed recently.
The other biggie is in some Events. and not all events / not all Daily rewards in events. But for Halloween in particular, the AA and House of horror Daily was clearly not being rewarded at the claimed rate and a lot of players complained.
There are two separate issues.. One. the true random nature of percentages. Some do not understand it. Second, and much more of a problem, since it undermines trust, is the apparent deliberate holding back of certain Daily prizes. And finally actual messed up stuff that happens now and then (like the as if one SC instead of 3in GbG)
So yes there are variabilities, but problems too.
The real proble here is those discussing variability refusing to acknowledge that some stuff is actually messed up. As if Inno is incapable of scerwing someting up and the claims it is 100% just probablilty reading like covering their behind

 

[Sharmon the Impaler]

I have been tracking 4 SCs in GbG over the past 4 months and have never gotten less than 9 attrition in 150 fights (121 sector levels to 150) in that time. While still close to the advertised 4% chance it has always been over by 1-2 %. The highest was 17 and the lowest was 9 attrition.

 

[Johnny B. Goode]

But for Halloween in particular, the AA and House of horror Daily was clearly not being rewarded at the claimed rate and a lot of players complained.

And those of us who got it/them at the advertised rate (or above) didn't complain. What's the percentage of players that complained compared to those who didn't? That's what I thought. So "clearly not being rewarded at the claimed rate" is a perception issue and not a proven fault with the game as you imply.

 

[Ebeondi Asi]

Your comment did not prove your claim either. Saying it's a perception is as much as error of presuppositions as mine might be,
You write the game cannot make errors. i say it can. You cannot prove you're correct, yet you claim that anyway.

 

[Sharmon the Impaler]

A thread was put up for general feedback where 68% of respondents complained about the drop rate. That would warrant at least a cursory look at the problem in my opinion. Your game seems to be going wonderfully and you always get more than expected so I don’t know why you feel the need to bother even writing in here. My drop rate sucked and well as well over half in my guild (sucked enough to stop doing the event not just low).

 

[My Push Account]

Let's make it simple. Throw a coin in the air 2 times. What if comes up heads both times, are ya going to say the coin is fake and has no tails? Of course not.

How about throwing it up 5 times, and get 4 heads in row? Is that going to keep that bee in your bonnet about the coin being rigged and now incorrect probability of 50/50? Of course not again.

Ever seen a run of 18 Reds in a row on a legitimate roulette table? Sure. We have all seen long duration Red/Black runs. Makes you want to bet on black next time, right?

Well, maybe not. Red has the same chance to come again after 18 reds...as a black... hope you know that. Past results have nothing to do with future events when it comes to probability in cases like this. They're independent events and not related. But your brain can get the better of you if you let it.

Time and tries evens out probability to it's true value. In these cases, the coin will eventually flip tails maybe directly after 2 or 4 heads in a row. Another red or even a black will come up in roulette after 18 reds in a row.

But over time, LOTS of TIME - the universe demonstrates the true probability of an event. Same with Inno's buildings. You just have to believe that. If you don't any game is a very slippery slope. Even real card playing.

Capture some data (tally your daily outcomes...) on what you're getting and you'll see that. Just give it some time to form for you.

PS: Probably not best to hear about a coin flip landing on the coin edge. It's entirely possible, though. A computational model suggests that the chance of a coin landing on its edge and staying there is about 1 in 6000 for an American nickel.

I generally get where you are going with this post, but I think there are a few caveats that you don’t address.

1.) The game mechanics are more complicated than flipping a coin, spinning a wheel, or rolling dice. I can’t imagine how many tables/arrays they rely on to go look-up the applicable percentage, send it to the RNG, then relay that back to the player as a result. There are many opportunities for the process to break down without even considering the actual RNG. And each of those opportunities is programmed by a human. The same group of humans that publish bug fixes every other week for other things that didn’t operate as expected. It’s not a stretch of the imagination that they, as all humans tend to do, make mistakes in all areas of the game and not just the cosmetic ones they share.

2.) Your assertion that Time and Samples will eventually show the true mean value, while true, is only part of the consideration. We know that the journey to infinity creates a predictable pattern of how quickly we will start approaching the mean. We aren’t going to have a fair data set that runs at 10% for 15,000 trials only to find out the true mean is 85% after 100,000 trials. There is an entire branch of mathematics dedicated to analyzing and describing these patterns. Without it, there would be no practical way to assess anything in this world, as infinite data sets are few and far between.

3.) In your coin-tossing examples, you show two examples that have a 25% and 15.6% chance of occurring, respectively. I agree, no one should be shocked by those results. But at what point should they be? 10 straight heads? 20? 100? How many straight Reds before the casino starts to wonder if something is up with the wheel or marker? There has to be a point where the odds are so overwhelming that no other conclusion can be drawn. We can disagree on what that % is, but we can’t just pick a randomly big number for each scenario and say that the result must just be good/bad luck. Again, we have to rely on the fundamental mathematical principles that describe probabilistic distributions.

To me, the biggest issue with most people that post about their “bad luck” isn’t that their sample size is too small, rather they don’t understand how likely their “bad luck” really is. The biggest issue with those that respond to those posts is that they only want to cite a single statistical principle, Law of Large Numbers, without any regard for the surrounding mathematics that would do a more apt job at describing expected outcomes. I don’t believe Inno is out to deliberately hold back prizes or rig the game, as that would be fraud and frankly because they don’t have to do that to be successful (what do they care if I get another Asylum or whatever building, I grow stronger and more invested in my city, probably spending more diamonds). I do believe that it is run by and programmed by people that are prone to make mistakes and that given their rapid growth, their underlying code is probably a nightmare to keep current and integrated with new features. And it is because of that belief that I think it is still worthwhile to be attentive and try to drive attention to things that might not work as expected, without an assumption of nefarious intent.

 

[Agent327]

I do believe that it is run by and programmed by people that are prone to make mistakes and that given their rapid growth, their underlying code is probably a nightmare to keep current and integrated with new features.

Question. What are those new features?

 

[Johnny B. Goode]

Your comment did not prove your claim either.

What claim did I make? Merely that your inference from a few complaints is not proof of anything. And it isn't.

Saying it's a perception is as much as error of presuppositions as mine might be,

Nope. It's not an error to say that merely stating that "a lot of players complained" is not proof of anything. Because it isn't proof of anything. And to jump from "a lot of players complained" (which is easily explained as a lot of players merely having bad luck...or the perception of bad luck) to "clearly not being awarded at the claimed rate" is just faulty logic.

You write the game cannot make errors.

Please direct me to where I said that. You can't. Because I didn't. Ever.

You cannot prove you're correct, yet you claim that anyway.

I don't have to prove that I'm correct, because I'm not claiming anything except that your conclusions are faulty and based on a smattering of anecdotal evidence that is undoubtedly skewed by those players' perceptions. You are claiming that the game's stated percentages are incorrect, so it's up to you to prove that claim. So far no one has ever supplied actual data to prove that claim. Time after time complaining players are told to track the numbers and provide hard data in sufficient numbers to at least indicate that there is an issue. No one ever has successfully. To my knowledge (and I've been on the Forum for over 6 years) every time someone has posted their numbers on the Forum, it backs up Inno's stated percentages. So, once again, track the numbers and prove your claim.

 

[Johnny B. Goode]

The biggest issue with those that respond to those posts is that they only want to cite a single statistical principle, Law of Large Numbers,

Why would we cite anything else when that is the very statistical law that the complainers are ignoring? Do you disagree with the Law of Large Numbers? Seriously, you're not doing these players any favors by feeding into their delusions.

 

[Emberguard]

Interestingly when I spend in bulk and if I were to divide the currency into the amounts you expect on a per Free to Play player I get a full range of results ranging everything from "received 1 Daily Special" to "multiple Daily Special streak run". Both of the extremes happen maybe once or twice while everything else is inbetween those two extremes. I haven't seen anything to suggest the algorithm is faulty, just random.

 

[Pericles the Lion]

These debates about statistics remind me of "Groundhog Day". Same players, same arguments. There seems to be two sides. In the first corner are the players that "perceive" something is amiss with the algorithm and the results are not as advertised. In the other corner are the players that "perceive" that everything is hunky-dorry. The fact is that neither side has done any modelling to confirm their point of view. Neither side is right and neither side is wrong....their opinions are all based on supposition. The one thing that amuses me is that fact that the daily changelogs almost always report bug fixes. Why on earth would anyone think that INNO's RNG, and the algorithms driven by it, are immune from bugs?

 

[Agent327]

Why on earth would anyone think that INNO's RNG, and the algorithms driven by it, are immune from bugs?

Ever seen any bug on the RNG fixed?

 

[Pericles the Lion]

Ever seen any bug on the RNG fixed?

If anyone had seen a bug in the RNG get fixed we wouldn't be having this debate now would we?

 

[Emberguard]

If anyone had seen a bug in the RNG get fixed we wouldn't be having this debate now would we?

Somehow I think if we had seen a bugfix for RNG bugs we'd keep revisiting it given bugs tend to get reintroduced every so often whenever changes occur

 

[Agent327]

If anyone had seen a bug in the RNG get fixed we wouldn't be having this debate now would we?

You checked all bug fixes. Because there are bug fixes you conclude that the RNG can be bugged as well. If so, since bugs are fixed, so would that bug be. See any mention of it?

 

[Pericles the Lion]

You checked all bug fixes. Because there are bug fixes you conclude that the RNG can be bugged as well. If so, since bugs are fixed, so would that bug be. See any mention of it?

So, you are implying that INNO announces all changes to the game and all bugfixes? That they keep nothing from the players? Full transparency?

 

[Agent327]

So, you are implying that INNO announces all changes to the game and all bugfixes? That they keep nothing from the players? Full transparency?

Now why on earth did I expect that comment?

So when it suits you, Inno is transparant in the bugs they fix, but if it does not suit you, they might not be?

Nine years of RNG in this game. Nine years of players complaining about the RNG, but every time a player takes the time and effort to keep track, they confirm the numbers given by Inno are correct. It's the complainers that never come up with the proof

 

[RazorbackPirate]

So, you are implying that INNO announces all changes to the game and all bugfixes? That they keep nothing from the players? Full transparency?

If you think there's bugs in the RNG, prove there's bugs in the RNG. Otherwise you're just making stuff up. Proof. Something you seem unwilling and unable to provide. Afraid you'll prove it works correctly?

Nine years of RNG in this game. Nine years of players complaining about the RNG, but every time a player takes the time and effort to keep track, they confirm the numbers given by Inno are correct. It's the complainers that never come up with the proof

This.

 

[Pericles the Lion]

If you think there's bugs in the RNG, prove there's bugs in the RNG. Otherwise you're just making stuff up. Proof. Something you seem unwilling and unable to provide. Afraid you'll prove it works correctly?

This.

How about you answer the question that I asked in #12? Given all the bugs in the game WHY would you conclude there have never been bugs in the RNG? I've made no claims that there have been because no proof exists: however, you, Agent, and JBG all seem to believe that no bugs have ever existed and I'm wondering where your proof is.