You’re not getting it then. Or I’m failing to communicate it properly. Or you’re purposely being obtuse which unfortunately happens on this forum a lot. Impossible to know. Good luck, have fun!
2 Likes
Nope. I get it. You don’t understand independent events. It will always be the same drop rate no matter how many times you run it. One person could do it a million times and never ever see it happen and it wouldn’t even be all that surprising. If a million people did it a million times and saw nothing, we would have a problem.
100 times is an infinitesimally small sample size. Not surprised in the slightest someone could go 100 runs without seeing it. Why does that shock you?
1 Like
100+ arena no SGC.
If by “realtively attainable odds” you mean you expect to win one in 3 hands, losing 100 in a row is pretty unlikely.
Lets say you play some imaginary game with exactly 100 hands and 3 players with equal probabilities of winning any given hand. Furthermore, each hand is completely independent of all of the other hands.
No we ask the question:
“How often would you expect to lose all 100 hands in the same game?”
With the Bernoulli trial calculation, you’re looking at (1 – 0.333) ^ 100 = 2.46 ×10^-18… or one in 4.07 ×10^17 games.
So not very often. In fact, we wouldn’t expect to see that outcome even once in the history of that game, even if millions of people played that game regularly.
If you disagree, then please describe how often you would expect to see such a streak.
You have never played a long poker or blackjack session then. Watch for the variance man. It is deadly.
You haven’t answered my question.
What do you think the odds are of such a streak?
If you’re open to learning about the topic, I suggest you start here: https://en.wikipedia.org/wiki/Bernoulli_trial.
Not only am I saying it is possible to lose every hand with a 33% chance of success, but it is possible to lose every hand with an 80% chance of success. Can and does happen and is why many player go flat broke. The variance kills them. They lose bankroll before the statistics even out.
I didn’t ask you if it was possible.
I asked you what the probability was.
Do you have an answer to this question?
Sure. With an infinite bankroll and an infinite number of hands, an 80% chance of winning is a profit. People don’t take into account the 20% chance of failure over the short term and believe me, 100 events is extremely short term.
The chance of losing an 80% hand is 20% no matter what you do or how many you play. Ergo, it is always just that. It doesn’t even matter how many you play.
I missed a single BWL raid in Vanilla and it was the one that UTB dropped; it went to our worst warrior. Our only Bonereaver’s Edge went to our MT. A single of Eye of Sulfuras and set of bindings dropped in all of Vanilla.
Our guild has seen Mageblade once, and the very next day after giving it to a mage, they quit and haven’t been heard from since. We’ve seen Vis’kag once, and again said person quit and hasn’t been heard from again. But we’re on our 4th Deathbringer…
No bindings, but 2 Eyes of Sulfuras. Also 2 BREs and 1 Spinal Reaper. One Might chest, 1 piece of Wrath. 0 Might belts, but like 8 GS belts.
Classic has been the polar opposite experience as Vanilla was as far as loot goes.
You’re replying to yourself and STILL not answering the question.
How likely is it that you see a streak of 100 losses when your odds of winning are 33%?
So you’re saying that the probability of the event is the same as the probability of the streak?
You’re actually saying that if the probability of losing the hand is 20%, the probability of losing 100 hands in a row is still 20%?
Are you… smoking something? Please tell me you don’t actually think this.
Here’s a nifty tool you can use to help calculate what you’re looking for:
https://www.wolframalpha.com/examples/mathematics/probability/bernoulli-trials/
I am saying each event is independent of every other even that occurs before and after. Yes. An 80% chance of success still has a 20% chance of failure no matter how many times you repeat the action.
For an independent event, you can’t ever take history or future into account. Hence the term independent. The odds never ever change with the number of times you perform an independent action. 20% chance of failure on the first try. 20% chance of failure after a million years.
Prove me wrong. An independent event with a 33% chance of success will still be the same 33% on the first try or millionth try so it doesn’t matter how many times it was performed by you past or present. Your history or future doesn’t matter in this case. This means you as an individual can absolutely fail every single time you perform said task.
I still haven’t see a Ban’Thok Sash drop yet but I’ve seen two Savage Gladiator Chains in BRD Arena.
In the original run of vanilla on my paladin, I was up to 40 UBRS runs without seeing the Lightforge Breastplate or a Blackhand Doomsaw, yet saw the Dal’Rend main hand 7 times.
RNG can be absolutely brutal sometimes.
Yes. Independent repeated trials are required in order to calculate the probabilities of sequences the way I did.
See: https://en.wikipedia.org/wiki/Bernoulli_trial
Which is why I clarified to you that we’re not talking about an event. We’re talking about a particular sequence of events.
I did that already when I gave you the abridged version of one of the proofs for the Bernoulli Process.
Yes, and we can calculate the probability of a sequence of failures.
You take the probability of failure ‘P’ and raise it to the number of times you performed the task ‘N’.
P ^ N
Which is what I did.
There are people who regularly play blackjack and have busted every single time they have hit on 16. They only have a 62% chance of failure, but after a few hundred hits on 16, they bust. You are never guaranteed a result based on number of attempts.
Nope. A 66% chance of failure is always a 66% chance regardless of the number of times an action is performed. Independent events guarantee at least one person will never ever win.