It tends to have a +-10% hedger for our games no matter what, so it gets less accurate the more users you have.
Ok good point. Do you know a ballpark figure for what it would cost dev's to obrain these market figures? I may not be looking at it hard enough but wouldn't spending it on research that adds more than steam data be more useful? I have played many indie games and feel like market research was not that important to many of them but I am sure some do use it.
Also I hear steamspy is accurate. Do we know how accurate and if things like refunds are accounted for? I don't use steam a lot so not sure if it removes the game from your profile after a refund.
Regardless I like seeing the numbers , however I would think it was unfortunate if employees were ever affected by it.
I think it's because GOG and Xbox don't use Steam as their social/community service.
It tends to have a +-10% hedger for our games no matter what, so it gets less accurate the more users you have.
IIRC it also doesn't include GoG or (in our case) Xbox One, so it's not very accurate at all in terms of total through sales. At the moment for our game I think it only shows about half of total sales.
+/- 10% wouldn't be less accurate at 100 million sales than it is at 100 sales. In fact, the numbers should be more accurate with more users.
Also, from what I have seen from dozens of devs over the past year or so, actual sales typically fall in or close to the margin of error stated, so being off by 10% on your games seems pretty high (unless they were small sellers. I know Steamspy isn't very good at the low end of things).
Steamspy works on player profiles that are accessible via internet, non-Steam game shortcuts that you manually add to Steam are local only. You can see if people are currently in non-Steam game, but it doesn't save on your profile in any way.Naturally. Although I'm not sure if GoG games which are added as 'Non-Steam products' to Steam count. Surely they still pick up the game's exact title.
That's interesting, and mirrors something I noticed on some games I was watching closely -- when they are included in bundles their owners stat often doesn't go up nearly as much as the bundle sales would indicate.Yeap, same here with our only game on Steam.
Specially it seems that a lot of sales from our Humble Bundle sales last year are keys not activated yet.
That's interesting, and mirrors something I noticed on some games I was watching closely -- when they are included in bundles their owners stat often doesn't go up nearly as much as the bundle sales would indicate.
That's interesting, and mirrors something I noticed on some games I was watching closely -- when they are included in bundles their owners stat often doesn't go up nearly as much as the bundle sales would indicate.
Steamspy works on player profiles that are accessible via internet, non-Steam game shortcuts that you manually add to Steam are local only. You can see if people are currently in non-Steam game, but it doesn't save on your profile in any way.
I think something like 80-90% of my Humble Bundle keys are still unredeemed because it's kind of a hassle, so I only put them in when I intend to use them in the not too distant future.
I didn't study informatics so I have no idea how this works in any great detail. Surely +/- 10% on 100 users would be a range of 90-110, while +/- 10% on 100 million would be a range of 90-110 million, which is way less accurate?
Also my +/- judgement is just based on looking at our game, it does vary. For instance when our game was at 40,000 users according to SteamSpy it said +/- 4,000 users. When we were at 60,000 users it said +/-5,850 users. Etc.
Agreed on everything else.
Nit: SteamSpy generally IS accurate. It's just not PRECISE.[1]
The numbers it quotes, in almost all cases, are within its stated margin of error (+ or - so many units) for the numbers it actually claims to measure (# of OWNERS -- not purchasers -- of a given game on steam).
The confusion comes from several places :
1) It needs 3 days of data before it has any validity and some people want to know on day 1, which is impossible
2) People want to know the exact # but all it has is an estimate with a margin of error
3) People don't understand the difference between owners and purchasers
4) People mistakenly assume all purchases -- itself an unknown number -- were made at full price.
See also: http://steamspy.com/about
[1] https://en.wikipedia.org/wiki/Accuracy_and_precision
EDIT: And to be sure, the above is kind of a lot to task of the general public. It's really easy to jump to wrong conclusions without reading his methodology on the about page in detail.
As a follow-up, here's a concrete example where people can totally misunderstand the data, which shows why a dev/pub might be nervous about SteamSpy, despite it being accurate (but not precise).
Two games on SteamSpy show these stats:
Game A: 10K owners
Game B: 100K owners
Behind the scenes:
Game A: 10K copies sold @ $30 (full price): $300K
Game B: 100K copies from bundles @ $0.30 each: $30K
Compare:
Perception: B = 10XA
Truth: A = 10XB
The difference between Truth and Perception here is a full 100X!
Here the confusion has nothing to do whether SteamSpy is accurate/precise, but in properly interpreting what X number of owners actually means.
I'm all for this knowledge to be public.
Nobody is going to miss Bad Rats on Steam Spy
I think its a neat data resource, but it also inspires a lot of dangerously misinformed analysis from people who don't understand the data, and are unable to place the data into a historical context. From the dude who runs it as much as anyone else.
There was the whole "Indiepocalypse" nonsense, which was largely unfounded scaremongering, and people not really understanding how the long tail works with critically acclaimed indie games, especially in relation to bundles and discounts.
It's cool to be able to see this data, although obviously everyone should be mindful of the significant margin for error, but I wouldn't personally feel I was missing out massively if it went away.
I imagine Valve will add a button for devs to toggle opt in/opt out, and say that they are not going to have anything to do with moderating this stuff in a hands on manner. Will require a bit of work but it's probably the easiest solution for them.
Edit: There are countless business reasons for not wanting sales data to be public. You can debate the merit of these reasons, but I doubt it's something Valve would want to take a hard line against publishers who make them lots of money just because of a not especially convincing public interest argument.
Hi
Steam Spy guy here
Here is what I wrote to Polygon. Will be happy to answer your questions during the weekend - it's a bit busy right now in the office.
I have a huge respect for Techland, and I love Dying Light. They're doing amazing work in game development, proving that a small company from Eastern Europe can become a big and respected game developer through hard work and dedication.
I want other developers to be able to learn from Techland and follow in their steps. Removing Paradox was already a hard decision for me for the same reasons. Nobody is going to miss Bad Rats on Steam Spy - there is nothing useful to learn from it. But both Paradox and Techland are doing a spectacular job with their games, and I'd hate people to lose this valuable lesson.
The point of Steam Spy is to be a helpful tool for game developers. Removing several important independent games from the service will hurt everyone else while not necessarily benefitting the publishers of the removed games.
I've talked several times about Steam Spy's limitations. For once, it's not a "sales tracking service" it's merely a polling service, that estimates the number of owners by polling user profiles. It's a bit like an election survey.
Imagine if Trump would ask everyone to stop publishing polls because they make him look bad and diminish his negotiating power. And honestly, if you believe your bargaining power lies in the ability to lie about your games data, you don't understand negotiations.
So, while I would like to avoid being sued by publishers for estimating the number of their games owners, I don't believe they have a case here.
There was some post back during the Summer Sale here that pointed out how much of the digital data was vague enough pre-Spy to float bullshit like "it was grandparent's computers downloaded once and set to be start up!" "Everyone was on the PSN/XBox sale!" and other uncallable claptrap.
I'm not quite sure I understand. The site is collecting the info from Steam Users' Profiles. It's basically public information to begin with. I don't think devs/publishers have any say.
Hi
Steam Spy guy here
Here is what I wrote to Polygon. Will be happy to answer your questions during the weekend - it's a bit busy right now in the office.
I have a huge respect for Techland, and I love Dying Light. They're doing amazing work in game development, proving that a small company from Eastern Europe can become a big and respected game developer through hard work and dedication.
I want other developers to be able to learn from Techland and follow in their steps. Removing Paradox was already a hard decision for me for the same reasons. Nobody is going to miss Bad Rats on Steam Spy - there is nothing useful to learn from it. But both Paradox and Techland are doing a spectacular job with their games, and I'd hate people to lose this valuable lesson.
The point of Steam Spy is to be a helpful tool for game developers. Removing several important independent games from the service will hurt everyone else while not necessarily benefitting the publishers of the removed games.
I've talked several times about Steam Spy's limitations. For once, it's not a "sales tracking service" it's merely a polling service, that estimates the number of owners by polling user profiles. It's a bit like an election survey.
Imagine if Trump would ask everyone to stop publishing polls because they make him look bad and diminish his negotiating power. And honestly, if you believe your bargaining power lies in the ability to lie about your games data, you don't understand negotiations.
So, while I would like to avoid being sued by publishers for estimating the number of their games owners, I don't believe they have a case here.
Correct, however up until today he obliged takedown requests as this dissuaded the publishers from taking their grievances to Valve in an effort to cut the website off from its source entirely.
Someone else would just pick up where Sergey left off, even if it's not an online database that anyone is free to query at any time. Ars Technica was already doing this a year before Steamspy went live.
I don't see any value in Valve trying to block Steamspy, and it doesn't seem like any of the major publishers that could actually push that decision care. Valve themselves doesn't care that their own numbers are out there either.
Someone else would just pick up where Sergey left off, even if it's not an online database that anyone is free to query at any time. Ars Technica was already doing this a year before Steamspy went live.
I don't see any value in Valve trying to block Steamspy, and it doesn't seem like any of the major publishers that could actually push that decision care. Valve themselves doesn't care that their own numbers are out there either.
Read his about page: http://steamspy.com/about
Regarding refunds:
https://twitter.com/Grief_exe/status/766663346204442625
Yeah. If they removed their games from the service Valve might but good that big puplishers don't care.
This may have already been said but, Techland are a polish company and CDPR recently got into legal hot water by speaking about their own sales figures.
Is it not unreasonable to think that Techland could be trying to protect themselves?
Hi
Steam Spy guy here
Here is what I wrote to Polygon. Will be happy to answer your questions during the weekend - it's a bit busy right now in the office.
I have a huge respect for Techland, and I love Dying Light. They're doing amazing work in game development, proving that a small company from Eastern Europe can become a big and respected game developer through hard work and dedication.
I want other developers to be able to learn from Techland and follow in their steps. Removing Paradox was already a hard decision for me for the same reasons. Nobody is going to miss Bad Rats on Steam Spy - there is nothing useful to learn from it. But both Paradox and Techland are doing a spectacular job with their games, and I'd hate people to lose this valuable lesson.
The point of Steam Spy is to be a helpful tool for game developers. Removing several important independent games from the service will hurt everyone else while not necessarily benefitting the publishers of the removed games.
I've talked several times about Steam Spy's limitations. For once, it's not a "sales tracking service" it's merely a polling service, that estimates the number of owners by polling user profiles. It's a bit like an election survey.
Imagine if Trump would ask everyone to stop publishing polls because they make him look bad and diminish his negotiating power. And honestly, if you believe your bargaining power lies in the ability to lie about your games data, you don't understand negotiations.
So, while I would like to avoid being sued by publishers for estimating the number of their games owners, I don't believe they have a case here.
Sounds cool, but I don't think that developers need Steamspy to know what games sold good on Steam or what games are popular.
And the Trump analogy is also very weird. I have given the example about Party Hard and it jumping from +- 50.000 to +- 150.000 because of a very cheap bundle. So, how could someone use such a number in negotiations or as a measurement for popularity? Being in a few very cheap bundles is probably enough for most games to have a high owner rate.
Also you suggest that publishers want their games taken of your site because they then can 'lie' about the sales data for a takeover or whatever. Maybe there is a case of that known, but to me it seems very far fetched and also very illegal.
Sounds cool, but I don't think that developers need Steamspy to know what games sold good on Steam or what games are popular.
Sounds cool, but I don't think that developers need Steamspy to know what games sold good on Steam or what games are popular.
Sounds cool, but I don't think that developers need Steamspy to know what games sold good on Steam or what games are popular.
You actually could not be more wrong here, sorry to say.
Benchmarking is critical in developing revenue forecasts, and, as an extension, development budgets.
The lack of information regarding the digital space has a huge negative impact on these essential business processes.
Is no one else going to call him out on how this site in general is obviously not for the developers? There are legitimate reasons to hide this info. You're hurting dev bargaining positions by posting it. He's a scumbag.
But you only quoted my first sentence. I explained why in the next ones, with an example and numbers.
Is no one else going to call him out on how this site in general is obviously not for the developers? There are legitimate reasons to hide this info. You're hurting dev bargaining positions by posting it. He's a scumbag.
But you only quoted my first sentence. I explained why in the next ones, with an example and numbers.
Sounds cool, but I don't think that developers need Steamspy to know what games sold good on Steam or what games are popular.
Squad (Kerbal Space Program): Paid their staff nigh zero wages, frequently fired them, game had sold 1.5 million copies at a high ASP. Their request was done under the cover of asserting that Mexican drug cartels would murder all their staff if their data was on SteamSpy because a drug cartel that was sophisticated enough to use SteamSpy wouldn't be able to figure out Kerbal Space Program was a success otherwise.
Can someone explain this one to me?
Fuck the publishers, but if a Dev asked, I would consider it... but I really don't understand the above for some reason.
Imagine if Trump would ask everyone to stop publishing polls because they make him look bad and diminish his negotiating power. And honestly, if you believe your bargaining power lies in the ability to lie about your games data, you don't understand negotiations.
A developer usually presents their internal sales figures for games it has developed in negotiations. The only way this information would hurt developer bargaining positions would be if the developer could not back up the information presented or the developer was trying to manipulate or hide data inappropriately.