207 réponses à ce sujet

[MOELANDER]

bigstig wrote...

While these polls show a high percentage of people unhappy with the ending it is only a small percentage of sales.

Rough estimates put sales at 3.5million in the first few days. Most polls I have seen(please show me examples otherwise) have barely touched the 50,000 votes mark. Even if you assume no cross over and pick the top 10 (each at around 50,000) highest polls it averages out at 500000.

500 000 is a far cry from 3 500 000 while it is still a percentage(round up to 15) it is hardly the 80-90% that "fans" keep shouting from the rooftops about.

Polls are also biased as detailed above as with the right question you can get the same person to give you completly contradictory answers.

People are entitled to their opinion but please don't state things as facts with meaningless statistics.

Yeah, sorry buddy, but that was 3,5 Mio SHIPPED, which means transprted to Retail. It's not the same as sold. Also you are looking at this from a perspective, that there would be a 100% voting participation and that my good sir is sooooooo ludicrously optimistic, that it reaches the level of Space Magic 

[dfstone]

I wonder what the Asian countries think of it? They get really serious about their gaming.

[Computron2000]

Ziggeh wrote...
It's selection bias if people are involving themselves in the poll based on their opinion, or to put it another way: people only vote in internet polls if they care.

This thinking on the "few people care" is flawed. There are a few situations that must be considered
i) The game is so good that people care enough to want other people to know about the game
ii) The game is forgettable so people don't care at all beyond a niche
iii) The game is so poor that people want other people to know how poor the game is

If ME3 is truly forgettable or terrible like a recent truck transportation simulator then what you say holds true yet at the same time all those who say it is good for 95% of the game are wrong, including those who liked the endings. By their actions of constantly being involved, everyone on this forum has chosen ii) to be false (and iii) as well as i have seen nothing of people saying how ME3 is as bad as the roadsweeping simulator game from last year). Game critics also cannot hold ii) and iii) as fact unless they are rating it average or below.

This leaves i). The idea of a self selection bias would then apply to all who find ME3 good enough to care.

[Ziggeh]

Computron2000 wrote...

This leaves i). The idea of a self selection bias would then apply to all who find ME3 good enough to care.

How does that change that fact that people involve themselves in a poll based on opinion rather than being selected at random?

Modifié par Ziggeh, 26 mars 2012 - 04:39 .

[MOELANDER]

Ziggeh wrote...

Computron2000 wrote...

This leaves i). The idea of a self selection bias would then apply to all who find ME3 good enough to care.

How does that change that fact that people involve themselves in a poll based on opinion rather than being selected at random?

Well then explain top me why almost none of the supposedly much bigger Pro-Ending crowd voted in those?

[Ziggeh]

MOELANDER wrote...

Well then explain top me why almost none of the supposedly much bigger Pro-Ending crowd voted in those?

You've misunderstood my point, but don't worry you're in disturbingly large company.

I am saying that if there were such a group of non voting "pro enders", internet polls would give you no way of knowing. The means of selection, having people choose to vote, would have filtered them out, creating bias. This means such a poll cannot be used to represent a larger population.

Well, that's not true. People use them that way all the time. Several times just in this thread. But this is what we call "inaccurate".

[Ieldra]

Ziggeh wrote...

MOELANDER wrote...

Well then explain top me why almost none of the supposedly much bigger Pro-Ending crowd voted in those?

You've misunderstood my point, but don't worry you're in disturbingly large company.

I am saying that if there were such a group of non voting "pro enders", internet polls would give you no way of knowing. The means of selection, having people choose to vote, would have filtered them out, creating bias. This means such a poll cannot be used to represent a larger population.

Well, that's not true. People use them that way all the time. Several times just in this thread. But this is what we call "inaccurate".

He still has a point. While it is trúe that the sample is not representative, it's not necessarily biased in favor of anti-ending people, because all kinds of people read that site. You can expect people who care about the ending one way or the other to vote. Just as much as anti-ending people tell their friends to vote there, so would people who have an interest in the endings not being changed. 

So while the percentages may not be accurate, I say the result is indicative of a widespread dissatisfaction beyond the minority who actually voted. You can argue it's not representative, but in the end it's just as likely to be biased in favor of pro-ending people. There's no way to tell, and given the probabilities I would be confident to say that among those people who care about the endings in the first place, the people dissatisifed with them are likely a majority.

Modifié par Ieldra2, 26 mars 2012 - 05:28 .

[MOELANDER]

Ziggeh wrote...

Well, that's not true. People use them that way all the time. Several times just in this thread. But this is what we call "inaccurate".

Well of course they are inaccurate! All Statistics are, or do you really still believe, that spinach has lots of Iron in it?
Who was it who said: "Don't believe any statistic you haven't faked yourself!" ? Was it Winston Churchill?

The point just is, for the limited Time these polls ran, the participation is most likely quite representative. The poll that ran first even had a section for those who didn't finished the game, yet. And those were quite a lot.

So take it easy and immer mit der Ruhe steigt der Opa in die Schuhe.

[Ziggeh]

Ieldra2 wrote...
So while the percentages may not be accurate, I say the result is indicative of a widespread dissatisfaction beyond the minority who actually voted.

Ieldra2 wrote...
There's no way to tell, and given the probabilities I would be confident to say that among those people who care about the endings in the first place, the people dissatisifed with them are likely a majority.

I'm perfectly happy, and indeed even agree with both of those statements. My issue is statements made without caveat. It's probable, it's likely. It's not definite and it's not proof.

Ieldra2 wrote...
You can argue it's not representative, but in the end it's just as
likely to be biased in favor of pro-ending people.

This bit is a little dependant. There are more reasons than this, but these are probably the big two that are an issue here:

Question framing can bias response. The gamefaq one posted somewhere in here is fairly horrible. It makes the "for" answers sound defensive and the "against" empowering. The ones in the OP are slightly better but would almost certainly instill bias. People do this because you get high response rates, but it invalidates your data.

And you need to look at context. Self selection polls are a feedback loop. The more people agree with your perspective, the more confident you will feel about it and the more likely you are to vote. A small but very vocal minority can radically alter such polls (I could find you an example, but I feel like just typing "Tea Party" here). A lot of press has focused on the negative feedback, which again lets people that that opinion is the one largely being discussed which will lead more "against" people to forums and thus to polls, all while driving out people who might like it but feel drowned out.

Again, this isn't to say that I believe this is a minority opinion, but rather to say that it is very much a belief.

Modifié par Ziggeh, 26 mars 2012 - 05:57 .

[MOELANDER]

You know Ziggeh I really want to call out your calmness and civilness in which you state these things. You haven't called anyone of us whiners or anything, like a certain Dridengx. I have to say, that it is a pleasure to argue with you! If this forum had a Karmasystem I would bolster it immensely!

[Ziggeh]

MOELANDER wrote...

Well of course they are inaccurate! All Statistics are, or do you really still believe, that spinach has lots of Iron in it?

Firstly, statistics tend to be (you'd be suprised how often they're not mind), it's their usage that is problematic. Which is exactly what I said.

Secondly, what? Spinach is high in iron. What has that got to do with statistics? That's chemistry and demonstrably true.

MOELANDER wrote...
The point just is, for the limited Time these polls ran, the participation is most likely quite representative. The poll that ran first even had a section for those who didn't finished
the game, yet. And those were quite a lot.

Representative means that you can accurately apply it to a group large than the one present. It does not mean it involves lots of people

In polling lots of data points reduces your margin of error. What it does not do is eliminate bias. The data is bias.

[MOELANDER]

I see.

With Spinach and Iron, don't you know the famous story how spinach was believed to be extremely high in Iron but in reality has only about like 3,5mg per 100g instead of the 35mg people once thought? It happened because of a faulty statistic.

Or maybe that's an urban legend.

Isee your point now and I agree. Of course it's biased. Only one question I have left for you: The first poll was made on 3.14. ME3 was released in Germany on 3.8. The poll ran for 2 days. What do you make from this data?

I am genuinely interested. And again, it's a pleasure to not be called a whiner!

[kyban]

Well done!! I still would like to see how ME3 is doing in Asia/Japan.

[Computron2000]

Ziggeh wrote...

Computron2000 wrote...

This leaves i). The idea of a self selection bias would then apply to all who find ME3 good enough to care.

How does that change that fact that people involve themselves in a poll based on opinion rather than being selected at random?

Its easy. Unless the people who love the endings found ME3 to be forgettable, they would vote in the poll. Like i said, regardless of how people view the ending, as long as they found the game great (and in ME3's case, great for 95% of the game) they would vote.

You can only say that ME3 is forgettable to a very large porportion of the population, in which case, the game critics are wrong. Now even if ME3 is forgettable for many many people, it does not change the result that among those who find the game great, the ratio as expressed by the sites noted by the OP hold true.

[ZLurps]

dfstone wrote...

I wonder what the Asian countries think of it? They get really serious about their gaming.

There was a topic about how Taiwanese players felt about ending, basically they thought it sucked.

Someone from South-Korea posted earlier in this very thread and wrote that gamers in SK thought it sucked... At this rate we are going achieve the impossible, world peace, if BioWare keeps things running.

[Ziggeh]

MOELANDER wrote...
but in reality has only about like 3,5mg per 100g instead of the 35mg people once thought? It happened because of a faulty statistic.

That is high! Not as high as say, a lump of iron, but as minerals in vegetation goes that's quite a bit.

MOELANDER wrote...
Only one question I have left for you: The first poll was made on 3.14. ME3 was released in Germany on 3.8. The poll ran for 2 days. What do you make from this data?

The data itself probably shows that a lot of people didn't like the ending, but displaying it as a percentage and as a pie chart (awesome though they are) is misleading.

[Ziggeh]

Computron2000 wrote...

Its easy. Unless the people who love the endings found ME3 to be forgettable, they would vote in the poll. Like i said, regardless of how people view the ending, as long as they found the game great (and in ME3's case, great for 95% of the game) they would vote.

Nope, there are more confounding factors. Like being the kind of person who likes to discuss gaming, who votes on things and the additional biases I mention above.

Computron2000 wrote...
You can only say that ME3 is forgettable to a very large porportion of the population, in which case, the game critics are wrong.

Can you say that? I wouldn't. I'd have no proof. Maybe if someone did a poll. And as we're talking statistics, I'd put good money that on average more game critics are wrong on average than right, so I'm not sure they're a useful metric.

Computron2000 wrote...
Now even if ME3 is forgettable for many many people, it does not change the result that among those who find the game great, the ratio as expressed by the sites noted by the OP hold true.

Again, further confoundation (that's probably not a word, but should be). In addition to above, this would be ignoring people who hated the whole thing.

You can say "of the people who voted in this poll", or by collating several "of people who voted in polls". Anything further is just inaccurate.

[MOELANDER]

Well thanks Ziggeh, that truly was enlightening. I am now off to play some Battlefield Heroes! Todeloo!

[Kilshrek]

Ziggeh, I understand your desire for an accurate poll, but a poll of such accuracy simply is not feasible.

We must first track down every person who has purchased ME 3, then ask them if they have finished the game. Then we should ask them if they have played previous ME games.
Then we must ask them for their opinion of the ending.

I understand that asking everyone is simply impossible, so the next best thing is to throw it out there and hope that those who have played ME 3 will respond. It does not necessarily mean that people who like the ending simply do not vote in the polls.

I admit that I am just as guilty of letting my emotions get the better of me when responding to posts about the ending, and in a battle of reason vs emotion, I'd put my money on emotions. I appreciate your efforts to bring reason to the discussion here, but too much reason won't go down well when people are still feeling rather raw about the whole thing,

[Megachaz]

blacqout wrote...

Megachaz wrote...

blacqout wrote...

There are a lot of unacounted for players in your chart.

What's funny is that you're more than willing to draw your own conclusions from limited data when deciding the size of your movement, but you can't do the same thing to make sense of a thought provoking ending.

Hilarious.

If you like drawing conclusions from limited data, why did you even need an ending at all?  The game could have cut to black the second we make our final choice.  Then we would have been able to speculate even more!  How thought provoking!

No.

The genophage was thought provoking.  Is biological warfare and genocide ever justified?  The Geth origin story was thought provoking.  If an artificial construct gains self-awareness and true intelligence, is it equal to organic life?  Cerberus and The Illusive Man.  Do the ends justify the means?  What if the means are mass murder?  At what point do you become the enemy you are fighting?

Having to try to guess at what really happened because the narrative was so unclear is frustrating, not thought provoking.

I didn't say that i liked anything.

It's hypocritical to declare your movement a majority, considering the very limited data, while decrying the Mass Effect 3 ending for not providing enough information.

My personal preferences don't come into it.

However, it is nice to know that the cliched big questions resonated with you... and if you think about it a little harder, you'll see that the prevalent themes of which were present within the finale.

At least BioWare know what they need to do to satisfy people like you in the future. Rehash the same material science ficton writers have been contemplating for decades.

It's actually not hypocritical at all.  It's not intelligent to declare a movement a majority based on biased polls, but it has nothing to do with the poorly written ending.  I never mentioned the themes present in the ending.  I was talking about how poorly written it was, considering it was full of plot holes and essentially required the player to make up their own ending because the one Bioware provided lacked necessary information to understand what happens after the credits roll.  It's incoherent, disjointed, and completely disregards the established personalities of many of the main characters.  That has nothing to do with a lack of knowledge of statistics.  I don't think you know the meaning of the word hypocrisy.

And if we're going to talk about cliched themes in sci-fi... ummm... technological singularity?

At least Bioware know what they need to do to satisfy people like you in the future.  Rehash the same material science fiction writers have been contemplating for decades, while leaving out any real closure to characters and conclusion of major plots.

[Computron2000]

Ziggeh wrote...
Nope, there are more confounding factors. Like being the kind of person who likes to discuss gaming, who votes on things and the additional biases I mention above.

You do know that i was talking about your own view that the majority of people would not care enough to vote? This is what you said

Ziggeh wrote...
It's selection bias if people are involving themselves in the poll based on their opinion, or to put it another way: people only vote in internet polls if they care.

Your quoted reply here  is of no relation to the point of "people only vote in internet polls if they care".

Here is the short form. The bias of "they care" for the subset is irrelevant for the subset as the superset of data already requires the bias of "they care".

Ziggeh wrote...
Can you say that? I wouldn't. I'd have no proof. Maybe if someone did a poll. And as we're talking statistics, I'd put good money that on average more game critics are wrong on average than right, so I'm not sure they're a useful metric.

Well if the game if forgettable for a large section of a population, then do you think it should rate a 9.0 or 10.0 for a large number of game sites?

Lets hear your answer.

Ziggeh wrote...
Again, further confoundation (that's probably not a word, but should be). In addition to above, this would be ignoring people who hated the whole thing.

You can say "of the people who voted in this poll", or by collating several "of people who voted in polls". Anything further is just inaccurate.

As i already noted, people who hated "the whole thing" (meaning ME3 in its entirety) would be motivated to vote as well to express their hate, probably by going against the majority.

As the saying goes, "The opposite of love is not hate, it's indifference". Only those indifferent to the whole game would not be motivated to vote and your selection bias argument requires a large population of the people who own the game and is indifferent to it, meaning the game is forgettable, meaning the game critics are wrong.

Modifié par Computron2000, 26 mars 2012 - 07:42 .

[Ziggeh]

Kilshrek wrote...

Ziggeh, I understand your desire for an accurate poll, but a poll of such accuracy simply is not feasible.

I don't want an accurate poll. I want people to understand how polls work. These polls are accurate, just not at giving you a percentage of any population beyond their own.

And it is feasible, just not with an internet one.

Kilshrek wrote...
We must first track down every person who has purchased ME 3, then ask them if they have finished the game. Then we should ask them if they have played previous ME games.
Then we must ask them for their opinion of the ending.

I understand that asking everyone is simply impossible, so the next best
thing is to throw it out there and hope that those who have played ME 3
will respond.

Not so, you just pick a sample size and select them randomly (what consitutes random is important) from within the population and then ask the question (how it is worded is important). If you've eliminated bias and picked a reasonably sized sample it would be representative of the whole within a calculatable margin of error.

This is the next best thing, throwing a poll on the web is somewhere further down the line, somewhere closer to "asking some guy in the pub" or "reading it on a toilet door"

Kilshrek wrote...
I appreciate your efforts to bring reason to the discussion here, but too much reason won't go down well when people are still feeling rather raw about the whole thing,

Oh I'm fine with ranting, I do plenty myself. I just feel it's important to present things correctly.

[Motherlander]

It's too complex for me. Too many colours. If it can't have only red, green and blue, then it clearly doesn't make any sense and is not intellectual enough.

[Niemack Saarinen]

Deutschland!

[Ziggeh]

Computron2000 wrote...

Your quoted reply here  is of no relation to the point of "people only vote in internet polls if they care".

Here is the short form. The bias of "they care" for the subset is irrelevant for the subset as the superset of data already requires the bias of "they care".

No, it was a response to the point that if they care they will vote. That most people will not consider voting is just the tip of a large iceberg of confounding variables. If the subset you intend to talk about is "people who care" then you need to account for those, which you can't in an internet poll.

Computron2000 wrote...
Well if the game if forgettable for a large section of a population, then do you think it should rate a 9.0 or 10.0 for a large number of game sites?

Lets hear your answer.

You would like to hear my answer to a rhetorical question? (You don't have to answer that.)

Computron2000 wrote...
Only those indifferent to the whole game would not be motivated to vote and your selection bias argument requires a large population of the people who own the game and is indifferent to it, meaning the game is forgettable, meaning the game critics are wrong.

I think this is where you've gotten a bit confused: I raised the possibility of a large group of indifferent potential voters as an example to show the flaws in the posted polls. I'm not saying such a group exists, or that the game is largely disposable. Even a small group of indifferent gamers, or people who don't go to the internet to discuss their gaming, or understand the misuse of polls would make such polls inaccurate but presenting an extreme makes the issue clearer. Or rather should.