After some babies grew up My food supply was dwindling and noticeably in the red. I stacked the diner with the 6 highest agility people and needed to rush for emergencies so my vault wouldn't starve to death. It was a 25% rush at first, which failed. But regardless of the continuously spiraling upward fail rate (which is totally lame) I have to keep trying because of starvation. I'm about to loose many many hour into an other wise very good vault with Unique characters etc, because of stupid game luck. I'm now on 25 (and counting as I am typing this failures in a row)OK now 26 failures? I don't believe this is even mathematically probable at the % rates being shown. Is this a bug or a Sick joke or what? Why can you not just show us the correct % of success?

Loosing all of your work because of pseudo random luck makes for a really lame game. Otherwise I'm quite pleased with it. I just don't understand why they want me too loose all of my work because of a really horrific game mechanic.

Maybe next version put in some kind of fail safe?

Edited:

How many times can one fail in a row loosing 98% of the population in your vault, trying to rush food in the diner? The answer is 39. 39 Fails and I have 2 women and a little Girl left. Oh and Scribe Rothchild is about 4 hours out returning from the wastes with huge piles of loot... Yea that's great. Thanks. Chances are he returns to an empty vault.

If you can afford it just build some temporary tier 1 diners and throw all the unnecessary personnel in there even steal some ppl from water and power if possible. Pretty sure if you fail a rush the timer resets so you have to wait even longer for a normal production cycle. when you are in a crisis it seems better not to rush a room just build more and try to balance it that way. even if they are temporary. A lot of people ive talked to feel like the RNG wants you to fail. i never rush rooms anymore except to complete incedent objectives like "put out 20 fires"

Yah I've failed on a 2% with 6 MAX STAT dwellers rushing a Nuka Cola room. Sure you can say those are the random chances but once you buy Lunchboxes you see nothing is random chance in Fallout Shelter. $20 Lunchbox pack will guarantee you 17-20 of the Lunchboxes will contain a "rare" 500 caps prize. A joke. Going back to rushing, I've found 30% is more like 60% (even with MAX 10 Luck dwellers). This at least being the case in my 5+ weeks experience.

That's a pretty good idea about making temp smaller rooms!

I agree it does seem like the RNG is either weighted to fail or the % they display is incorrect by a large margin. Anything around 50% should succeed about 1/2 the time but in the game 50% fails a large majority of the time.

Wow! Couldn't agree more and yes, 50% is like a 80%-85%.

I agree the rush % is totally inaccurate and I noticed that once you fail once you will probably fail the second third and fourth times also depite the chances being 30%, 40%, and 50% respectively.


The game seemes out to screw you no matter what....I wish they would fix these things and nerf deathclaws and the other disasters...


Radroaches in a lv3 room are still radroaches. In the real Fallout games they still have 5 hit points and you can still kill them with a BB gun in one or two shots....


There are no behemoth radroaches...


Why are they so hard on us with this game?

$$$$$$$

Personally I think you people are suffering from Negativity Bias - Psychological phenomenon by which humans have a greater recall of unpleasant memories compared with positive memories. You remember all the failure more than the successes so the negative results seem to be more common because you remember then more often. Second some of you I think are experiencing cluster biased thinking a cluster of failures are a result of a bug, I mean it has to be a bug right? This many failures at once can't be random. yeah it is.

25% failure rate means you are going to fail 1 in 4 times it also means at times you will fail multiple times in a row because that's just random chance. And you pretty much ignore all the successes you get in a row.

39 failures back to back is not Negativity bias it's bad code. Many games counter cluster failures in poor RNG with a fail safe. Basically you are only allowed to fail/succeed X many times in a row to prevent a bummer experience when using RNG to determine the outcome of things.

We are all aware of Psychological reasons why X happens. Negativity bias, information bias, cluster bias blah blah blah. Someone always has to bring this up in game threads over and over to try and explain away bad RNG. Never fails. Hows that for Bias?

If it was life based I would agree with you but in programing there is no such thing as random, code can be made to seem random but it still runs in a sequence, Ive played many games with bad RNG failing 4 or more times in a row with 95% chance of success. I'm not saying this game has a bad RNG and I wouldn't say that until I've done testing which imo isnt worth the effort when its just simpler avoid the mechanic when possible since it isnt game breaking. but i will say this, remeber twice in this game ive failed rushes at 14% then immidately try again at 20% and fail. I can also tell you i have never succeded at a 60% or higher rush and ive done many to start fires on purpose to complete objectives. both scenarios are statistically improbable.

I think Goth should pur her money where her mouth is and tell us how many back to back failures she gets over 40%.


I have actually tried rushing multiple times like the majority of the posters in this thread and like them I agree the rush percentage is totally bogus and statistically broken.

Lol. Now that's funny! If you believe this after even playing 1 hour then math may not be the strongest of skills. This is not at all meant to be a negative remark or hostile post. I just believe in truth and in this case, all you have to do is play 1 hour of this game and you'll know the percentages are against you.

Now fast forward to 5+weeks (in my experience) and you will know 100% that the percentages are off. Additionally, in my new vault I have a 5 Intelligence dweller by himself in a stimpack room maxed upgraded. I receive 24% Rush rate at it's lowest and he has yet to have 1 success from this out of 10 times. No negative psychological mishaps here just an understanding of seeing this game's cash grab angles.

BTW, I love this game but I am highly frustrated with losing 4 and a half weeks of time/money in a built/trained 50 MAX STAT, 200 dweller vault to this horrible load crash bug. Additionally, Bethesda has yet to address this to the thousands that have had this issue.

Do you how many games are claimed to have bad RNG? Everyone. Do you know why? It is because of cognitive bias. Algorithms can be made that take the sequence nature of the computer and have it give a random result because the seed imputed is taken from variables that are not constant like the time, making the net output random or near enough that there is no legal distinction which is why you can have electronic gaming machines, even though as you say computers are sequential.

The reason why every game has a bunch of people claiming the RNG is broken is because games are play in most case by large numbers of people so you are more likely to get outlier from the average result and some of those outliers are going to be negative. And time and time again especially with results that can be both positive or negative people will remember the negative more than the positive which automatically makes it appear that there are more negative results then positive. And the thing is you as an individual may have a statistical result that has more negative results than the stated number and it still doesn't actually mean the RNG is broken. Clusters exist in random numbers, this is why people who gamble at craps can have a "hot streak." The existence of a negative cluster is not actual evidence of a problem. You can't take an isolated set of results and use them in isolation to show cause of a problem you have to get set thousands, hundreds of thousands of results and see what the results are.

Negative Bias is one of the most common form of bias that gamers encounter and very rarely do the claims of bad RNG actually result in the RNG being bugged. The thing is it is really easy to have the games metrics tested on a mobile device and they can see the number of rushes the number of success for each % range and if there was a problem it would be a glaring as the numbers would clearly show if there is a problem it would be almost impossible for such an error to make it unnoticed through development hey almost exactly like I said about having a large set of thousands of results and comparing actual results to projected results. Imagine that. A few people upset on the forums isn't evidence, in fact this type of evidence for this type of issue has been PROVEN to be woefully inaccurate because of cognitive biases.

The fact the people refuse to even consider the bias is yet another form of bias called Bias blind spot, where people are able to identify bias but failing to see the impact of biases on one's own judgement.

I guess you ignored the part about how for quite some time now most games that use RNG based code often put in anti cluster code, to keep someone, for example, from rolling 39 negative or positive results in a row. To omit that is bad code, period. There is no psychological explanation needed.

As far as far more negative results then positive on rushes, this has been clearly documented all over the internet enough that it simply does exist. The irony behind your negativity bias theory is that it has it's own built in argument where by refuting the theory somehow proves it right. This is called silly stupid gibberish to the rest of the world. We understand what you think is going on but when there is a collective of documented results with real data to look at it totally destroy's the idea that it's just all in our heads. RNG is not actually random. No mater how many semantic arguments you make, machines can not do random.

Now, it's very likely the case that the % shown in this game is just a wrong representation of what it actually is. This has been common with every Fallout game(and many other titles) after Fo1 actually. The skill % in Fallout 2, for example is not the actual success rate but rather their own oddball calculation. That's why you could exceed 100% and still not guarantee success and it not be a bug.

Can machines do true random numbers? No, but they can simulate is so closely that there is no LEGAL distinction between the results given from an anologue device vs a digital device, which is why you can have electronic gaming machines which are all based on random chance.

What "collective of documented results?" Anecdotal evidence is not evidence, especially when it is shown to be biased and this unreliable. You seem to fail to understand what random means, any individual could get a cluster of results that far outweigh the percentages. This is especially true when dealing with larger number sets. This is why gambling works because there will always be someone that defies the odds and wins, which in turn gives hope to others and so they gamble too hoping to be a big winner. Well just as there can be outliers that are positive you can have outliers that are negative it is how casinos profit, because for every cluster of wins there is a cluster of losses. Casinos win because the percentage chance to lose is such that on average people will lose money not win it. While most people will win and lose close to the mean not all people will. So your individual results, your anecdotal evidence isn't substantive because of how random works you need large set sizes to see if there is actually a problem and not just a cluster.

Every game with RNG has a bunch of people QQing over the broken RNG because of Bias and a complete lack of understanding of statistics. But you say you have a "collective of documented results" so please how large is the set size thousands, hundreds or zero just a few people going the rush feature is broken because they failed? Given the fact that this claim is ALWAYS raised by gamers I think it is far more likely you are suffering from cognitive bias then there being an actual problem. One because its a case of the gamer crying wolf over and over again and second because there is no DATA just anecdote. Until there is actual evidence this is just the same old QQ over RNG. And the same Bias blind spot as that people always seem to think "I couldn't possibly be biased."

Hey i could be wrong but until I see actual numbers in the thousands showing a skew of the projected norms then its just gamer QQing as per usual.

Omg you are not worth any more replies as you are either trolling us or you are daft.


Have fun in fairy land....