Player stats, sales, and reception speculation thread

[Xur]

What are you saying is "normal"? It has been "normal" to see reviews well below 40%. The past 2 weeks with reviews being 50% is not normal. They're the best 2 weeks since February. It hasn't been on a constant downward trend since launch, it has been up and down. Take a look:

View attachment 745615

Civ VI went from from a low overall score of 66% to high overall score of 87%. If Civ VI can do that, I think Civ VII can reach at least 70%. It will take years, no doubt. Recent reviews consistently hitting 50% is the first step. The past 2 weeks are 50%, we will see if they hold, improve, or decline. As usual, I will report the worse numbers as well as the better numbers. Unfortunately some people are unable to acknowledge when things start to look better.


No, I can't see why. How exactly have I been silly or dishonest when I've been quoting factual data which gets ignored?

That's fair enough. I'm not saying the numbers are wrong - I'm saying you're using them in a misleading way. You are cherrypicking. What if I said, "The average human has one breast and one testicle"? It would be statistically true, but it wouldn't align with reality. You can use statistics to mislead - even if you don't mean to.

 

[SammyKhalifa]

Looking forward for the next few weeks, I'm going to guess that we don't see anytihng this month, but a pretty major patch Mid-Late November in preparation for the Holidays. So there will be another dropoff followed by a later spike--enough to satisfy all parties no matter what they want to be mad about!

Most of the other releases that I remember came out more this time of the year than when 7 came out: don't know if that plays into anything or not.

 

[Smegger213]

What are you saying is "normal"? It has been "normal" to see reviews well below 40%. The past 2 weeks with reviews being 50% is not normal. They're the best 2 weeks since February. It hasn't been on a constant downward trend since launch, it has been up and down. Take a look:

View attachment 745615

Civ VI went from from a low overall score of 66% to high overall score of 87%. If Civ VI can do that, I think Civ VII can reach at least 70%. It will take years, no doubt. Recent reviews consistently hitting 50% is the first step. The past 2 weeks are 50%, we will see if they hold, improve, or decline. As usual, I will report the worse numbers as well as the better numbers. Unfortunately some people are unable to acknowledge when things start to look better.


No, I can't see why. How exactly have I been silly or dishonest when I've been quoting factual data which gets ignored?

OK, over the last 14 days there were 203 positive reviews and 207 negative ones.
That is NOT 50% positive.Its 49.5% actually.
The game is STILL getting more negative than positive reviews.
I don't see the point in saying "but the positive is growing" when the overall number has not changed in months.

 

[SammyKhalifa]

OK, over the last 14 days there were 203 positive reviews and 207 negative ones.
That is NOT 50% positive.Its 49.5% actually.
The game is STILL getting more negative than positive reviews.
I don't see the point in saying "but the positive is growing" when the overall number has not changed in months.

You don't see the point in looking at trends?

 

[IntelligentDisk]

That's fair enough. I'm not saying the numbers are wrong - I'm saying you're using them in a misleading way. You are cherrypicking. What if I said, "The average human has one breast and one testicle"? It would be statistically true, but it wouldn't align with reality. You can use statistics to mislead - even if you don't mean to.

How am I cherrypicking or using them in a misleading way? They're the most recent reviews and recent player counts - they're the most relevant. I'm not going back and saying "well August had some better reviews!".

I've said several things which are factually true:

• Best reviewed and first mostly positive reviewed day (9th October)
• Best reviewed month since the first month*
• 4th best reviewed week since launch (week 36, 9th October - 15th October)

*1 month counted as 4 weeks, this is so you can directly compare to other games, it isn't cherrypicking. Tell me what you think is being cherrypicked or being used in a misleading way.

OK, over the last 14 days there were 203 positive reviews and 207 negative ones.
That is NOT 50% positive.Its 49.5% actually.
The game is STILL getting more negative than positive reviews.
I don't see the point in saying "but the positive is growing" when the overall number has not changed in months.

That's correct, it's 49.5% not 50%. See how I acknowledge you? You don't see the point in talking about recent reviews which are higher than the overall score? The best reviewed and first mostly positive reviewed day? The best reviewed month since the first month? The 4th best reviewed week since launch? Why are you coming to a "Player stats, sales & reception speculation" thread then?

 

[Xandinho]

No, there are certain people that keep on about the same thing time and time again.
But their so called facts just do not add up.
EG how does 472 positive reviews vs 577 negative reviews over the last 30 days indicate that positive reviews are improving?

In percentage terms, the numbers might still be improving, even if the overall reviews remain mostly negative. For example, let’s say it used to be 41% positive reviews, and now it’s 45%—isn’t that an improvement? Last week we had 51% positive reviews (the first time it has surpassed 50% since launch). This could be an isolated case or a slight upward trend, so we’ll have to wait and see. No one expects the reviews to suddenly jump to 70% positive, since the game still has a lot that needs improvement, and things don’t change that quickly. However, even small changes can gradually shift the overall perception of the game.

 

[IntelligentDisk]

Looking forward for the next few weeks, I'm going to guess that we don't see anytihng this month, but a pretty major patch Mid-Late November in preparation for the Holidays. So there will be another dropoff followed by a later spike--enough to satisfy all parties no matter what they want to be mad about!

Most of the other releases that I remember came out more this time of the year than when 7 came out: don't know if that plays into anything or not.

You could be very right with this prediction. They've been giving us an update every month so far, so maybe they will still give a small update next Tuesday, with news about what they're working on in future updates, or maybe we will just get news with no update. The Community Manager in the Discord stated there would be news later this month, and that the devs are working on the next update, so that's all we have currently.

The timing matters in my opinion. A game being released in October in time for Winter & Christmas is going to greatly benefit from increased sales early on compared to being released in February leading into Spring.

I already took a look to see how the Christmas periods affected Civ V & Civ VI. Both of them were seeing gains of anywhere between 26% to 52% in their average weekly peak concurrent players during the Christmas period, with them all seeing permanent gains in the weeks after the Christmas period in comparison the weeks before the Christmas period, except for Civ VI after its first Christmas which saw a decrease of around 15%. Civ V instead saw an increase of around 16%. Two big differences though:

• Civ V was on sale for 25%, and released September (bigger sale and released further away from Christmas)
• Civ VI was on sale for 10%, and released October (smaller sale and released closer to Christmas)

With Civ VII releasing almost a year away from the Christmas period, and likely going to be on sale for 40% during the Winter sale, the game could see big permanent gains similar to that of which Civ V or VI saw after their first Christmas periods.

 

[itix]

I can illustrate the giant hole civ7 is in by a little thought experiment. What if we wanted civ7 to reach a 75% positive review score and the current reviews stay as they are. How many more reviews would we need? If we are generous and assume all new reviews have a 80% positive 20% negative split, then they would need 204.888 more reviews (I dont assume anyone care for the calculation?). Compare that number to their current 36.584 reviews. That would require anything from 6 million to 15 million more sales on Steam - depending on how review happy those buyers are. The 6 million is a rather agressive 1 review in 30 sales. I personally find this scenario almost impossible.

This is it exactly. The gap is too wide.

Which makes me think that Firaxis should just say "screw it, full speed ahead with the original vision". There is no chance they can recover it; they should just do it as planned and see where it goes. If not far, then so what. There is no other direction they could go.

 

[Smegger213]

You don't see the point in looking at trends?

No, I don't see the point in looking at trends.
What matters is:
1. The overall review count and the percentage difference between positive and negative reviews.
2. The average player count at the end of each month. Is that count rising or falling?

Look at the difference between the 30 day review graphs for Civ 5, 6 & 7.
Yes, Civ 5 & 6 have been out for years. But Civ 5 has a 95.6% positive and Civ 6 has 85.8% positive.

It will take some major changes and many years if Civ 7 has any hope of matching those positive percentages.

 

[Leucarum]

This is it exactly. The gap is too wide.

Which makes me think that Firaxis should just say "screw it, full speed ahead with the original vision". There is no chance they can recover it; they should just do it as planned and see where it goes. If not far, then so what. There is no other direction they could go.

The counter is that while it seems to be in a slightly deeper hole than Civ6, it's not THAT much deeper. It is still plausible to turn things around. The difference to me seems to be that there is a lack of uniformity in terms of what people dislike. There isn't a clear direction which would count as "giving the players what they want."

 

[GeneralZift]

I haven't studied statistics in a billion years, but I remember distinctly something.

Isn't there like a threshold for significant statistical change?

Like one month it's 48% and the next it's 49% this could be considered within a certain threshold where the change is just random noise and the reality is the same.

I remember this being a concept, being P < 0.5 or something. Anyone know what I'm talking about and can apply it to this situation?

 

[bladex1200]

I haven't studied statistics in a billion years, but I remember distinctly something.

Isn't there like a threshold for significant statistical change?

Like one month it's 48% and the next it's 49% this could be considered within a certain threshold where the change is just random noise and the reality is the same.

I remember this being a concept, being P < 0.5 or something. Anyone know what I'm talking about and can apply it to this situation?

You're thinking of p-value and statistical significance testing. Worth noting that a low p-value does not indicate that the outcome is or isn't statistically-significant, it just means that the odds of the null hypothesis (e.g. that what you're observing is unrelated to the variable you are studying, the variable you are studying doesn't actually exist or has no effect) being true is very unlikely.

A single study with a high p-value is not enough to conclude that an outcome is random or not statistically-significant. All this says is that the single study's outcome doesn't have strong rigor either way and further studies are needed.

 

[Siddorm]

I haven't studied statistics in a billion years, but I remember distinctly something.

Isn't there like a threshold for significant statistical change?

Like one month it's 48% and the next it's 49% this could be considered within a certain threshold where the change is just random noise and the reality is the same.

I remember this being a concept, being P < 0.5 or something. Anyone know what I'm talking about and can apply it to this situation?

I think that statistical hypothesis testing could extinguish this vibrant discussion.

 

[Kev1916]

That's fair enough. I'm not saying the numbers are wrong - I'm saying you're using them in a misleading way. You are cherrypicking. What if I said, "The average human has one breast and one testicle"? It would be statistically true, but it wouldn't align with reality. You can use statistics to mislead - even if you don't mean to.

At a glance Steam shows

53% of all reviews are negative, within last 30 days the reviews 54% are negative , yea things are getting better ...

If people get a warm glow when checking daily reviews and find a 6 to 4 positive score for a massive 10 people than bash on

 

[Siddorm]

Claude Sonnet 4.5 (LLM AI) can do statistical tests correctly.

 

[UWHabs]

The counter is that while it seems to be in a slightly deeper hole than Civ6, it's not THAT much deeper. It is still plausible to turn things around. The difference to me seems to be that there is a lack of uniformity in terms of what people dislike. There isn't a clear direction which would count as "giving the players what they want."

It's a tough decision on the whole. I mean, frankly, there's too much of a hole and too much from the initial impression that the game will have a very hard time making it out of the "Mixed" bucket on Steam ever. But you can still have a profitable and fine game with 60% or so reviews. The real questions for the internal team is basically what they have to do to sell enough DLC to be able to have the game at least as a commercial success, and then at some point in a few years, when do you stop putting work into this edition and pivot to the next edition. This is all assuming that there's not some magic expansion which corrects all the issues.

But frankly, the game will need to be able to sell a couple expansions before they could ever start on a civ 8, and if the reviews are still mixed or negative for too long, they might not be able to sell those expansions and DLC to the same level they would have initially hoped. And at that point you just have to hope the studio and publisher are willing to put in to invest in having a next edition.

 

[kotpeter]

At a glance Steam shows

53% of all reviews are negative, within last 30 days the reviews 54% are negative , yea things are getting better ...

If people get a warm glow when checking daily reviews and find a 6 to 4 positive score for a massive 10 people than bash on

All reviews also include the first week when the game was released. That's a massive number of reviews, and the game was different in many aspects back then. It had worse UI and more bugs. While % of positive reviews wasn't as bad that week, the decline in player count was massive. If the game had released in a 1.2.5 patch state, we might've had roughly the same % of positive reviews, but more concurrent players now. What I'm trying to say is that comparing latest 30 days to lifetime has its niche uses, but it's not indicative of small positive changes. Whoever brings up lifetime scores for comparison is, unintentionally or not, dismissive of the iterations of changes the game is going through and instead focuses on how far it still has to go.

What really matters, in my opinion, is the period N days before and N days after the latest patch. With N being at least a few full weeks. This comparison will indicate if the game is getting better in terms of presented metrics and will help us understand if the patch was warmly received by the playerbase. Surely it doesn't include people who didn't buy Civ 7 at all, but the those are hard to estimate. And no, you cannot use civ 6's player count to make this estimation.

 

[Evolena]

While I generally agree with this statement, if the total count is very different, percentages may actually lie or show incomplete picture.

E.g. if number of reviews dropped 5x compared to past month, we shouldn't look at percentages.

Yes and no. The real limitation of percentage is with low total count, or with one value that has a different seasonality than another (like a probe that generates regularly a value whatever the day while real data are affected by week-ends)

 

[IntelligentDisk]

At a glance Steam shows

53% of all reviews are negative, within last 30 days the reviews 54% are negative , yea things are getting better ...

If people get a warm glow when checking daily reviews and find a 6 to 4 positive score for a massive 10 people than bash on

As you want to use the last 30 days, how about we compare it to 30 day periods before then too?

• The last 30 days: 45% positive (1,032 reviews)
• 30 days before the last 30: 38.1% positive (886 reviews)
• 30 days before above 30: 37.1% positive (1,220 reviews)
• 30 days before above 30: 39.9% positive (1,205 reviews)
• 30 days before above 30: 27.3% positive (1,758 reviews) - experienced some review bombing
• 30 days before above 30: 32.4% positive (1,672 reviews)

A 5% to 7% increase is obviously significant. This isn't one day and it isn't 10 reviews. It's a 30 day period with 1,000+ reviews.

 

[Evolena]

At a glance Steam shows

53% of all reviews are negative, within last 30 days the reviews 54% are negative , yea things are getting better ...

Means are only means. If in the meantime it had reached 60%, then it has decreased since that point, but if it had reached 45%, then on the contrary it has increased.

No Man's Sky could do it, so never say never. (A lot of) (More) Time will tell.