MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
lordrichter wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Common trap. It is not the sample size that matters. It is whether the sample accurately reflects the larger population. Until someone can show me non-Steam numbers so we can reasonably correlate, we don't know to what degree those numbers represent more than just Steam players.
MLGProPlayer wrote: »nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
We're talking about Steam Charts. The sample is tens of thousands of active players. If I had to guess, Steam probably replresents 20-30% of the game's total population. It's a massive sample.
MLGProPlayer wrote: »lordrichter wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Common trap. It is not the sample size that matters. It is whether the sample accurately reflects the larger population. Until someone can show me non-Steam numbers so we can reasonably correlate, we don't know to what degree those numbers represent more than just Steam players.
Steam users aren't behaviouraly any different from non-Steam users so the sample size is the only thing that matters here.
lordrichter wrote: »MLGProPlayer wrote: »lordrichter wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Common trap. It is not the sample size that matters. It is whether the sample accurately reflects the larger population. Until someone can show me non-Steam numbers so we can reasonably correlate, we don't know to what degree those numbers represent more than just Steam players.
Steam users aren't behaviouraly any different from non-Steam users so the sample size is the only thing that matters here.
You can't make that determination. Sure, if we were talking about people who visit Vvardenfell, or how many people have DKs, Steam Charts would probably be fine. Logging into the game? There are different factors on each of the platforms that can determine usage of the game. For us, whatever we get from Steam Charts is anecdotal across the larger player population.
MLGProPlayer wrote: »nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
We're talking about Steam Charts. The sample is tens of thousands of active players. If I had to guess, Steam probably replresents 20-30% of the game's total population. It's a massive sample.
Do you have any reason to think that the original statement by ZOS that the three platforms are broadly equal in population was either inaccurate or has changed?If so, what are your sources? If not, then you are suggesting that virtually the whole of the PC playerbase plays the game through Steam, an assertion which I imagine you would be hard pressed to substantiate!
MLGProPlayer wrote: »lordrichter wrote: »MLGProPlayer wrote: »lordrichter wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Common trap. It is not the sample size that matters. It is whether the sample accurately reflects the larger population. Until someone can show me non-Steam numbers so we can reasonably correlate, we don't know to what degree those numbers represent more than just Steam players.
Steam users aren't behaviouraly any different from non-Steam users so the sample size is the only thing that matters here.
You can't make that determination. Sure, if we were talking about people who visit Vvardenfell, or how many people have DKs, Steam Charts would probably be fine. Logging into the game? There are different factors on each of the platforms that can determine usage of the game. For us, whatever we get from Steam Charts is anecdotal across the larger player population.
You can absolutely make that determination. There is absolutely nothing empirical or theoretical you can draw on to assume that Steam users are behaviourally different from non-Steam users.
MLGProPlayer wrote: »MLGProPlayer wrote: »nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
We're talking about Steam Charts. The sample is tens of thousands of active players. If I had to guess, Steam probably replresents 20-30% of the game's total population. It's a massive sample.
Do you have any reason to think that the original statement by ZOS that the three platforms are broadly equal in population was either inaccurate or has changed?If so, what are your sources? If not, then you are suggesting that virtually the whole of the PC playerbase plays the game through Steam, an assertion which I imagine you would be hard pressed to substantiate!
I meant that Steam likely represents 20-30% of the total PC population.
Even realising that those millions of players are unlikely to play all at once
MLGProPlayer wrote: »I meant that Steam likely represents 20-30% of the total PC population.
Does it? It shows 25K concurrent players in the graph posted earlier. If that is 1/3 of PC population, and PC population is 1/3 of total population, that would put the max concurrent players just under 230K. Meanwhile, google says:
"How many players does ESO have 2019?
We've just seen the launch of a new Elder Scrolls Online expansion, and the game's had a big year. ESO has picked up 2.5 million new players in the past 12 months, bringing the total player count up to 13.5 million players over the game's lifetime just after the release of Elsweyr."
Even realising that those millions of players are unlikely to play all at once, based on these numbers I would still say that steam represents way less than 30% of total PC population.
MLGProPlayer wrote: »nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
We're talking about Steam Charts. The sample is tens of thousands of active players. If I had to guess, Steam probably replresents 20-30% of the game's total population. It's a massive sample.
MLGProPlayer wrote: »MLGProPlayer wrote: »nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
We're talking about Steam Charts. The sample is tens of thousands of active players. If I had to guess, Steam probably replresents 20-30% of the game's total population. It's a massive sample.
Do you have any reason to think that the original statement by ZOS that the three platforms are broadly equal in population was either inaccurate or has changed?If so, what are your sources? If not, then you are suggesting that virtually the whole of the PC playerbase plays the game through Steam, an assertion which I imagine you would be hard pressed to substantiate!
I meant that Steam likely represents 20-30% of the total PC population.
Does it? It shows 25K concurrent players in the graph posted earlier. If that is 1/3 of PC population, and PC population is 1/3 of total population, that would put the max concurrent players just under 230K. Meanwhile, google says:
"How many players does ESO have 2019?
We've just seen the launch of a new Elder Scrolls Online expansion, and the game's had a big year. ESO has picked up 2.5 million new players in the past 12 months, bringing the total player count up to 13.5 million players over the game's lifetime just after the release of Elsweyr."
Even realising that those millions of players are unlikely to play all at once, based on these numbers I would still say that steam represents way less than 30% of total PC population.
nafensoriel wrote: »MLGProPlayer wrote: »nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
We're talking about Steam Charts. The sample is tens of thousands of active players. If I had to guess, Steam probably replresents 20-30% of the game's total population. It's a massive sample.
We have actual monthly unique login numbers quoted from the president of ZOS themselves. They went up 500k in two years.
Steam has no bearing on reality. The number is not a ratio you can poll. By the listed numbers assuming all platforms are at 500k/uu/m the paultry 25k on steam isn't even a rounding error. Considering all MMOs have terrible steam population figures if they started outside of steam and then transitioned to steam you can safely assume they have no statistical value whatsoever.
nafensoriel wrote: »@Facefister Actually monthly unique logons is the de facto metric for the population of MMOs.
@MLGProPlayer
The daily average is month/30 for all practical purposes. It's the same principle and no it's not really significant. You are looking at too small a dataset in exclusion to the whole. At best you are looking at a fraction of 1/6th the total population of the game in those figures.
Tell me how you can extrapolate anything from such a small and incomparable subset of data, please. I'm curious about your line of thinking. As it stands all I can see in your data is cherrypicked figures which does not a good argument make.
Lord_Eomer wrote: »ZOS is again on a wrong track, nerfing everything and letting players leave ESO.
Lack of good contents are also to blame, players are asking for new Solo Arena or Duo but Q4 DLC becomes such a disappointment.
I have also taken a break from ESO, will get back once ZOS start putting things back on track.
Source: https://steamcharts.com/app/306130
MLGProPlayer wrote: »MLGProPlayer wrote: »MLGProPlayer wrote: »nafensoriel wrote: »MLGProPlayer wrote: »It's a massive sample size. You can very reliably extrapolate these trends to the rest of the player base.
Uh.. what? No, you cant. That is not how statistics work.
It is a sample size drawn from a very tiny portion of the game's population (forum users) in a section of the board where even fewer people visit. It is basically like looking at one square inch of a 100 story building and claiming you can tell jack and squat about its construction quality.
You cant use forum polls as reasonable reflections of a game where 95+% of the player base will never once log into them.
We're talking about Steam Charts. The sample is tens of thousands of active players. If I had to guess, Steam probably replresents 20-30% of the game's total population. It's a massive sample.
Do you have any reason to think that the original statement by ZOS that the three platforms are broadly equal in population was either inaccurate or has changed?If so, what are your sources? If not, then you are suggesting that virtually the whole of the PC playerbase plays the game through Steam, an assertion which I imagine you would be hard pressed to substantiate!
I meant that Steam likely represents 20-30% of the total PC population.
Does it? It shows 25K concurrent players in the graph posted earlier. If that is 1/3 of PC population, and PC population is 1/3 of total population, that would put the max concurrent players just under 230K. Meanwhile, google says:
"How many players does ESO have 2019?
We've just seen the launch of a new Elder Scrolls Online expansion, and the game's had a big year. ESO has picked up 2.5 million new players in the past 12 months, bringing the total player count up to 13.5 million players over the game's lifetime just after the release of Elsweyr."
Even realising that those millions of players are unlikely to play all at once, based on these numbers I would still say that steam represents way less than 30% of total PC population.
Concurrent = currently logged on
You're correct. Take surveys as an example. The most precise data would be asking the entire population but asking like 10k people out of 1 million is also enough because percentages don't fluctuate that much after asking additional 10k people. The Steam chart isn't maybe 100% precise but it tells us the direction.MLGProPlayer wrote: »As to your second point, a sample size >10,000 (the total Steam sample is likely in the 100,000s) is more than enough for any statistical testing.