Player stats, sales, and reception speculation thread

[IntelligentDisk]

"Better" here depends on your goals. If you want to track overall impression among players, data from more sources is better. But if you look at it as a commercial goal (reach "mostly positive" on recent reviews to improve further sales), looking at numbers shown by Steam could make sense.

I think most of the debate I've been involved in has been if user reviews have gotten better, so it makes sense to include all and not omit any if we're talking about the sentiment.

 

[bitula]

They're referring to SteamDB. SteamDB counts all reviews on Steam (from Steam purchasers and non-Steam purchasers). The graph, the recent reviews and overall figure on Steam only shows reviews from Steam purchasers and omits reviews from non-Steam purchasers, so it's better to use SteamDB. SteamDB shows 47.42% overall (up from 47.13% or whatever it was). The recent review figure from SteamDB is currently peaking at 58.7%.

Ahaa, good to know, thanks!

 

[Lazy sweeper]

Indeed, we were already writing here about the positive trends in player numbers and reviews well before the "play as one civ" announcement was made. And people who aren't buying the game until that feature is added also aren't leaving reviews. So, there's no reason to think that the announcement has anything to do with it.

The day people will be able to play as FRANCE, US, etc with (my opinion) to NOT have ANY Leader to choose from... from the get go, is where and WHEN things will turns.

 

[Core Imposter]

Part of the story of this game is misallocation of resources. Sid was proud, maybe rightfully so, that they could launch over all platforms and include depth in detail about the civlets, down to language and building styles and the inclusion of almost a glut of graphical art. But somehow, they didn't expend resources to ensure that the gameplay itself would be palatable. One would think that would be first order. And then the rush at the end and the UI problem.

The game can never be made good, but it will be an achievement if they can get it to an acceptable level for a small minority of the community. Futurewise, one hopes they test the gameplay out before they do a full development cycle.

I don't expect to live until another iteration comes out. Given the length of past development cycles and the historical survival rates of people my age, I have about a 1% chance to see it. If so, I hope above all things that they go back in the other direction and fix the scale of the game.

 

[Gori the Grey]

I have about a 1% chance to see it.

Citing your own actuarial figures gives a whole new meaning to "player stats."

May you live long and prosper, Core.

 

[Gorbles]

The game can never be made good

Your personal dislike of every Civ. past IV remains valid, but at the same time it makes proclamations like these seem a tad unnecessarily dramatic.

 

[Core Imposter]

Your personal dislike of every Civ. past IV remains valid, but at the same time it makes proclamations like these seem a tad unnecessarily dramatic.

Ok then, let me edit. It will be an achievement if they can get it to an acceptable level for a significant percentage of the community. My emphasis was not on the first part of the sentence. It is no small potatoes to make a save from where they started.

 

[Skojardu]

Ok then, let me edit. It will be an achievement if they can get it to an acceptable level for a significant percentage of the community. My emphasis was not on the first part of the sentence. It is no small potatoes to make a save from where they started.

Let's hope they get enough run support. I finally bought it and yup, tastes (metaphorically) like New Coke

 

[IntelligentDisk]

For the first time, Civ VII has averaged over 60% positive for a 4-week period. 1161 total reviews, 700 positive, 461 negative, 60.3% positive.

 

[PlatinumPoster]

For the first time, Civ VII has averaged over 60% positive for a 4-week period. 1161 total reviews, 700 positive, 461 negative, 60.3% positive.

View attachment 748673

Are the recent days/weeks in thus 4 week period also close to the 60% number?

 

[Smegger213]

Are the recent days/weeks in thus 4 week period also close to the 60% number?

Starting counting from a Thursday we get the following:
Week starting Thursday 30th October = 181 positive, 92 negative = 66.3% positive.
Week starting Thursday 6th November = 214 positive, 140 negative = 60.45% positive.
Week starting Thursday 13th November = 155 positive, 131 negative = 54.2% positive.
This makes the average for those 3 weeks as 60.3%.
No doubt you can get different numbers by starting counting on a different day.
EG the latest full week we have starts on Tuesday 18th November, and gets 144 positive, 113 negative = 56.03%.

I have also noted that the peak player count at weekends has dropped back down to around the 11,300 mark. After that high of 14,019 on 9th November.
Does this mean that its trending back down again in player numbers? Well, we will have to keep checking.

 

[IntelligentDisk]

Are the recent days/weeks in thus 4 week period also close to the 60% number?

As of today, the past 4 weeks are 67.4%, 61.4%, 55.6% & 61.7% from oldest to newest. The 4-week period is up slightly to 61.3% as of today's numbers currently.

 

[IntelligentDisk]

Do tell me what day you start counting from?

I believe it's self explanatory? Today until 4 weeks prior.

 

[IntelligentDisk]

I asked what day of the week you start counting from?
Simple question, how about a simple answer?

Oh and you cannot count todays reviews until tomorrow, because todays numbers will rise further.

Simple question which you should be able to simply work out yourself with 5 seconds of thought.

 

[Smegger213]

Funny that when I add up all the reviews for November so far, not counting today.
I get 625 positive, 432 negative.
That equates to 59.13%.

 

[uppi]

For those who genuinely don't know what a rolling average is and so that the others have no excuse, here is how to calculate a 4-week rolling average:

1. Take the date and time you want to calculate the average for, let's say 2025-11-25 12:00 UTC (pro tip: use UTC for this, if you don't want to worry about DST changes)
2. Subtract 28 days from it. You would get 2025-10-28 12:00 UTC
3. Take all reviews that fall within this time period
4. Assign a 0 to negative reviews and a 1 to positive reviews
5. Take the mean

You can do this as often as you want (though obviously, you might want to think about how often you want to). You could do this in six hours for the time period from 2025-10-28 18:00 UTC to 2025-11-25 18:00 UTC. This will add a few extra reviews and drop a few old reviews, but overall the result will not change by much. And that is the advantage of a rolling average. Note that there is no free parameter here: The rolling average for a point in time is well defined (no matter which day of the week it is and what time it is) and the question: "Which day did you start" does not make sense.

 

[Smegger213]

For those who genuinely don't know what a rolling average is and so that the others have no excuse, here is how to calculate a 4-week rolling average:

1. Take the date and time you want to calculate the average for, let's say 2025-11-25 12:00 UTC (pro tip: use UTC for this, if you don't want to worry about DST changes)
2. Subtract 28 days from it. You would get 2025-10-28 12:00 UTC
3. Take all reviews that fall within this time period
4. Assign a 0 to negative reviews and a 1 to positive reviews
5. Take the mean

You can do this as often as you want (though obviously, you might want to think about how often you want to). You could do this in six hours for the time period from 2025-10-28 18:00 UTC to 2025-11-25 18:00 UTC. This will add a few extra reviews and drop a few old reviews, but overall the result will not change by much. And that is the advantage of a rolling average. Note that there is no free parameter here: The rolling average for a point in time is well defined (no matter which day of the week it is and what time it is) and the question: "Which day did you start" does not make sense.

If someone is going to come out with statements like:
As an example: Week starting 13th November the positive review average was 54.2%, then you need to know on which day of the week you are going to start counting from.
EG many posters in this thread have posted things like "last week was 60% positive" Or "the last 2 weeks have averaged over 60% positive". In both examples you need to know what day of the week you started counting from.
After all you can get a totally different review count for a week, by EG starting counting from a Monday instead of a Thursday.

 

[uppi]

If someone is going to come out with statements like:
As an example: Week starting 13th November the positive review average was 54.2%, then you need to know on which day of the week you are going to start counting from.
EG many posters in this thread have posted things like "last week was 60% positive" Or "the last 2 weeks have averaged over 60% positive". In both examples you need to know what day of the week you started counting from.
After all you can get a totally different review count for a week, by EG starting counting from a Monday instead of a Thursday.

If that someone has been posting rolling averages for weeks and made plots with them, then the safe assumption is: Starting from now

 

[Smegger213]

If that someone has been posting rolling averages for weeks and made plots with them, then the safe assumption is: Starting from now

Here's an example. A few pages back, someone said this statement "63% over the past 2 weeks & 60.5% over the past week".
Now if you want to find the percentage of positive reviews, then you have to start counting from a given day of the week. EG count 7 days starting from a Monday, then add up the total positive & negative reviews to get the percentage.
Its just the same when you say "it was 63% positive over the past 2 weeks". You need to know exactly which day you started counting from. Then add the 2 weeks of reviews together to come out with a percentage.
That is the sort of thing I am getting at.

 

[uppi]

Here's an example. A few pages back, someone said this statement "63% over the past 2 weeks & 60.5% over the past week".
Now if you want to find the percentage of positive reviews, then you have to start counting from a given day of the week. EG count 7 days starting from a Monday, then add up the total positive & negative reviews to get the percentage.
Its just the same when you say "it was 63% positive over the past 2 weeks". You need to know exactly which day you started counting from. Then add the 2 weeks of reviews together to come out with a percentage.
That is the sort of thing I am getting at.

Again: follow my instructions, assume that "past two weeks" means exactly "past 14 days" and there is no need to know when to start counting from.