Unless it is intentional to force rarity in gambling specifically.
That is not true. For each try the chance is reset, so it’s always 10% chance. You can do one try, ten tries or a hundred tries, you have 10% chance to get the item.
The probability of getting 1 “win” at x% in y tries = 1-(1-x)^y. So if x is 10% and y is 10 tries, the probability of getting 1 win in 10 tries = 1-(1-0.1)^10 = 1-0.9^10 = 1-0.349= 65%.
I’d say it’s 10% each try, with no change from one try to another. Each try is 10% chance, so overall it’s 10% chance. Otherwise, this would mean that after a certain amount of tries, we would be guaranteed to get the item.
It’s not though, go and Google how to work out the probability of any event in a number of tries, you’ll get the formula I gave above.
Llama is correct. You’re right that each individual roll has a 10%, but there’s a difference in a guaranteed chance and probability. If you try this 1000 times, there’s a very good chance that 100 out of those would roll as expected. But you wouldn’t know beforehand which of those 100 would be the ones you expected. The larger sample size the closer the successrate becomes to the expected chance.
2 Likes
Well, that’s something I’ll have to teach to the mathematicians I have at home.
Thanks for the clarification and explanations, guys! 
1 Like
Not to me.
Play few hours in Poe, then you will know hat gambling is.
Do I need to explain probability with an example you would understand? You keep saying that you may “never” get a success, but even the most basic grasp of possibility would tell you that after infinite tries the chance of success is literally 100%.
I will go with a coin toss, you are expected to get a head within the first 10 flips 1023 times out of 1024. This doesn’t mean that it is guarenteed, but the sheer amount of tries means that only a fool DOESN’T expect to get heads(assuming the coin isn’t rigged). As a large number of tries occurs the odds of not winning a game of chance at least once approaches the chance getting hit by a meteor next tuesday.
Now apply this to LE gambling, where the average is somewhere between 1 million and 20 million gold(you would need a large sample to test this, and that is hard with the second of wait between gambles, I am guessing values to make a POINT). If I assume 20 mil at 5000 per try(ignoring refresh), that means the number of tries is 4000.
if average tries is 4000 that means that you have 50% at that, you can therefore calculate the odds per try((1-x)^4000=.5 so 4000th root .5 =1-x, or .5^(1/4000)=1-x) so we get x equal to about 0.00017, I will use this value from now on to make a point(because calculating it more accurately is silly as I am working off a lot of guesses to begin with).
Taking 0.00017 and applying it to just 10,000 tries (1-.00017)^10,000= 0.1827 or only an 18% chance of failure, it would be reasonable to expect success within the first 10,000 tries. After 20,000 tries you have only a 3% chance of failure. at 30,000 it is .6% chance, at 40,000 it is .1%
Notice how over multiple tries the chance of success gets INCREDIBLY close to 100%?
Basically you have been talking out of your ass when it comes to probability, about us not being able to set an expected payout cost and how we “may never even get it”.
The ONLY thing that is unclear is the exact odds of success per roll and of getting bases on refresh rolls.(well that and the possibility that they are using pseudorandoms, which changes the way it works entirely and instead makes the exact outcomes predictable)
1 Like
To augment your example: Last Epoch Info feel free to check the gambling simulator. It should have accounted for gambling cost of the base but not for rerolls, AFAIK.
Maths is hard, yo!
Also I think some people get hung up on the common English usage of “expectation/expected” (aka guaranteed) compared to the maths/stats usage (likely outcome).
I personally do not see the difference between the common English usage and the maths usage, you expect not to get hit by lightning next monday for example. This doesn’t mean you know the exact chance, just that you believe it to be significantly low.
Probably an unpopular opinion, but I don’t see the point of having shopkeepers, gambler or gold in this game at all. It is simply not necessary. I just bought, installed, tried and played LE since last week. I only play SSF-HC. My stash is empty on all characters. 3 characters 1 death only.
I’ve never gambled, bought anything from a shop, or used gold for anything other than respec. I honestly don’t see the reason for why the developers even feel these “features” has to be in the game. It feels arbitrary.
Stash we need - no argument there. It’s just my opinion that the shops, gold and gambler is completely and utterly unnecessary.
I gamble a lot, to find good craft bases.
I understand your point and it’s good that gambling/purchasing is not mandatory and is totally avoidable, though.
1 Like
My only complaint is that it’s too slow. I don’t mind the gamble chances at all.
Re-read what I’ve written in this thread and you’ll note that I never said anything about probability, because I don’t consider it relevant. So please, take the condescending nerdrage out back.
If you engage with a random system expecting a result you want after X rolls because you’re treating mathematical probability as a progress bar that you just have to fill up, the only guarantee is that you are going to end up mad/frustrated/sad/disappointed. When that happens, it will only be your fault for setting an inappropriate expectation. That is what I said in my very first reply. Anyone who has played games with random systems for any amount of time should understand that no matter what probability says, you’re gonna have bad luck.
But if anybody wants to keep pretending it’s a smart idea to work yourself up in your leisure activity because you did math and decided you’ll definitely gamble your unique after X million gold, keep on keeping on I guess? I’ll be over here not constantly getting mad and threatening to quit the game when reality doesn’t line up.
You are the one who is treating probabilistic near certainties as if they are not the expected result.
I was not expecting a result AFTER X rolls, I am expecting the result WITHIN X rolls, as in set a budget before gambling and then spend until you run out or get a hit. That can mean anything from 1 roll to X rolls. It is important when taking games of chance to set a budget. But the nonsense you are speaking about how someone may “never get it” is implying that an INFINITE budget is the only acceptable amount.
You say “no matter what probability says, you’re gonna have bad luck.” but that is just plain false - bad luck is a feature of a large number of individual random events and some of them turning out poorly. It does not apply to large sample sizes of hundreds or thousands of rolls trying to get one specific result where the effect of bad luck will be eliminated by sheer brute force.
You only expect bad luck because you ignore the instances of good or even average luck.
There is point when it comes to probability when expecting bad luck to occur is Equivelent to astronomical anomolies.
1 Like
And by doing so, I never find myself in a position in which I’m so upset that missing an expectation I’ve set of hitting a “probabilistic near certainty” makes me - checks notes - “feel like a big loser” or “not feel like playing the game at all”.
So, I mean, you can yell at me about how much I don’t understand probability, but it seems to me that I’m enjoying my leisure activity a lot more by ignoring it than other people are by obsessing over it. 
1 Like
I am not upset by failing a roll: that was the other guy way back who was only asking the very simple question of what amount of gold should be enough to expect a resut - I am upset by you disparaging statistics and game theory. Two things I consider to be serious business,.
All probability is about is maths. Probability is why you don’t play the lottery, probability is also why brute forcing gambling in LE tends to work out in the long run.
1 Like