That is absolutely not how polling works and that should be obvious to everyone that spends more than 30 seconds thinking about it. Polling agencies spend a tremendous amount of work in building representative sample groups to which they then reach out to. They don’t just ask whoever happens to come through the door.
A poll on the forum is the opposite of that. Only a small fraction of the userbase uses the forum, opinions on the forum are disproportionally negative as people usually come here to complain, polls on specific things will mostly attract people that are invested in them one way or the other. Trying to judge the sentiment of the WT playerbase as a whole from one isolated forum poll is like claiming you could tell the outcome of a national election from polling the people that went to some random townhall in a random district.
Which can only mean one thing Gaijin adding fake non exist weapons to handheld Russia.
800 people vs the whatever (even exaggerated) amount of people is very strong. i already said that with 150k people (even if that is exaggerated) the error is at 2.6%
don’t run in circles saying the same arguments as the other person that i already called out on his non sense
even if you want to exaggerate it just to be sure by assuming that some people are invalid then it will still be less than 10% and that does not make a difference in the poll result
No poll is needed for that as Gaijin already told they can add fake vehicles if they deem necessary.
Ensuring that the poll is valid is on the poll maker, I’m just here pointing out obvious flaws.
No, it’s not, because you have no way of knowing how votes of the playerbase as a whole, forum users and voters on this one topic are distributed.
You can’t determine a margin of error without a confidence level. You can’t assume you have a high confidence level unless you’re polling a randomized group. You’re not, you’re polling a self selected one without any comparison data.
It just not good for the long run create more problem and sometime lead to lawsuit
and I don’t wanna deal with those mess so I choose not to do it.
ofcourse people at gaijin would validate it however im pointing out we can also do it if needed as information is open source here
this is about weapons not vehicles
it is a huge difference and a false equivalency at that to assume that they bear the same weight.
they don’t.
ill still be waiting for gaijins reply on this thread. if they don’t reply then they don’t. it is really your and mine opinion whether a poll is required or not and there is no point arguing that, however i am saying that it would be statistically accurate when done the way i said.
95% is the default and you should probably know that as it is the general standard
do we have one person arguing about the user base having to be picked for validation and another user arguing that it has to be fully randomized? definetly closing each other’s arguments out
even to you each other, either one of you is wrong. is it you, @maguumo or you, @MotorolaCRO ?
ill expect an answer
I already explained how SM3 might not be able to get balanced better at other BRs and with other weapons, but in another thread that’s more appropriate.
SU-34 can’t even be pushed down that much due to ARH limitation.
That too, but FMs and A2A capabilities are also an important thing in GRB, as you’ll be meeting enemy aircraft there too.
AAs need help across the board, that’s pretty obvious.
95% is the “standard” if you have a randomized group. You do not have a randomized group. You are specifically asking people that are riled up about it and made the effort to go to the forum to complain about it. It’s like asking people that just got a ticket for speeding what they think of radar traps compared to asking every 1000th inhabitant of the city they live in.
That has nothing to do with each other. If you want a reliable poll you need to poll a reprenentative group. Excluding “bad faith” actors is one thing you can do to slightly improve your dataset, it’s still going to be garbage compared to a randomized group.
this isn’t a medical placebo study as you know. you also have a false equivalency similar to what the motorolaguy posted, this in your argument wouldn’t be a “people who got a speeding ticket” because in the sense of argument this isn’t a speeding ticket forum. this is the general war thunder userbase forum and while the post is “kh-38mt may not exist”, the thread is representative of people who both agree and disagree with it and is not locked out from anyone who does not agree.
this is more like going to a town general forum and asking what they think of speed cameras rather than going to ask only people who were fined.
your false equivalence shows that you’re arguing in bad faith just like the other person
i also mentioned that i support gaijin making the poll themselves, yet here you also stayed quiet.
If Gaijin was keen to remove it they’d have done it already, I don’t think they need our guidance. That being said, I’m definitely not the one who’ll validate that thing.
Weapons are just a giant paper weight without vehicles that are launching them, so I agree with you. Fake vehicles should bear much more weight than weapons.
Actually we’re both right.
You can’t accept votes from people outside the target range and you also need to guarantee the selected people have been chosen randomly.
So yeah, having 80 year olds voting on a poll for 18 - 25 people is wrong. Likewise, polling people on a platform that is mostly used to complain is wrong as it’s obviously not random.
sometimes gaijin only does things after user pushback. we know what happened earlier, however not sure if we are allowed to mention it as you know why.
except they haven’t been chosen randomly if they’re from a target group as all of them would have been selected
(thats what you 2 wanted?)
you can’t select everyone from a target group and say it was random as there was no randomization - and you can’t go and select some in your own arguments as you claim that is inaccurate and everyone needs to be picked
both of your arguments dissolve each other and itself
That pushback happened on another platform and it casted a bad light on the game as a whole. It didn’t happen on the forum.
Randomly from a target group.
In this case it would be from random players playing top tier GRB. We already established that this forum and many others alike often serve as an echo chamber for complaining, so it’s not randomized.
Possible replacements for Kh-38MT:
it was a prime example regardless of where it happened
however prior to that we had a lot of “discussing” about it on forums.
a single weapon has less pushback however it also has way less weight on it to be changed
yet you earlier implied that everyone needs to be picked or otherwise its not accurate?
sure, @MotorolaCRO tell me about your ideal poll and how it is done, including the amount of people you want to ask from and who you dont want to ask from and the criteria for who is asked from and who is excluded
It was you that claimed to know margin of error and confidence level based on bogus assumptions in a complete absence of data, not me. We wouldn’t be talking about this if you had an inkling of an idea about statistics.
Used by a self selected part of the userbase (pretty much the opposite of randomized) that disproportionately comes here to complain.
There is no such thing. I already mentioned that, political polling agencies go to great lengths to ensure a representative sample group. If they just went to some convention and asked everyone there their results would be garbage.
his ideal poll would only ask russian mains who regularly us Kh-38MT and have already talked about how it shouldnt be removed on the forums
Pushbacks don’t happen here as this place is pretty isolated and won’t really stop them from gathering new players.
The larger the size the more accurate it is.
Perfect would be an in-game poll that is shown to people that play in modes and areas affected by poll topic.
Asking Naval-only people what they think about 8.7 helicopters doesn’t make much sense, don’t you think ?
except there aren’t “2 groups” that need to be asked something thus comparative groups are useless
disproportionately? or just more people disagreeing with your opinion?
feel free to prove your claim
which is invalid because what we would have here would be considered an election type of poll and not a prepoll which people can change minds after or otherwise highly volatile
you also did not address the false equivalence which i told you about
even here we could expect a difference between prepoll and the actual vote
however we dont do prepolls because the opinion can indeed change afterwards by some %
Ideal poll (this applies to all pools) would be if it was asked directly in game so it can notice as many people as possible.