-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
SteamSpy - Approximate LTD sales for every game on Steam (Updated Daily)
- Thread starter MUnited83
- Start date
I've added recommendation engine. It's pretty simple right now - it uses your most played games to find people like you, then it filters out people based on their similarity to your playing habits and creates a list of games that they play a lot but you don't,
I'm still playing with parameters because it's a resource hog and I'm trying to optimize it.
Could you guys please help me test it? Just go here and paste your steam profile link. You need to be logged in, though, as this thing is pretty demanding on my resources.
Here is how it looks.
Note, if you have Russian or German versions of certain games (Fallout, Wolfenstein) you might see their international version in the list.
I'm still playing with parameters because it's a resource hog and I'm trying to optimize it.
Could you guys please help me test it? Just go here and paste your steam profile link. You need to be logged in, though, as this thing is pretty demanding on my resources.
Here is how it looks.
Note, if you have Russian or German versions of certain games (Fallout, Wolfenstein) you might see their international version in the list.
alexandros
Banned
Amazing work, bravo!
For some reason, I have no data on Magicka 2 at all. That's why it's not in the search results. But it is in the database, all right - http://steamspy.com//app/238370
Very interesting
https://twitter.com/Steam_Spy/status/606764052220657664/photo/1
Should dispel the misconception that most people on Steam only buy games when they're 5$ or less.
https://twitter.com/Steam_Spy/status/606764052220657664/photo/1
Should dispel the misconception that most people on Steam only buy games when they're 5$ or less.
TC McQueen
Member
I think it's because a lot of those $60-$70 dollar games are highly anticipated titles and/or some kind of multiplayer component that incentivizes jumping in at the start.
Oh wow, I didn't know Survivarim came out. Also weird that TF2's numbers drop after the first ten minutes. Either people are checking out Source performance or some of those are bot accounts?
galyonkin said:
Oh neat. Gonna give that a try, but you need the profile to be public like that Word Cloud SteamStat thing that was featured on RockPaperShotgun?
If it's tracking changes in ownership you would see the change reflected in the numbers I'd assume.
I can't track refunds, but I can track changes in owners. So far it doesn't seem so bad
Turin Turambar
Member
I can't track refunds, but I can track changes in owners. So far it doesn't seem so bad
What's the scale in that graph?
Just a question, galyonkin
Is this based just on launch prices or does it count sales and any kind of fluctuation (i.e. the 20 to 29.99 doesn't consider only titles which were 20 to 29.99 at launch, but also those that maybe were 49.99 to 59.99 at launch and then got a price cut to 20 to 29.99)?
Very interesting
https://twitter.com/Steam_Spy/status/606764052220657664/photo/1
Should dispel the misconception that most people on Steam only buy games when they're 5$ or less.
Is this based just on launch prices or does it count sales and any kind of fluctuation (i.e. the 20 to 29.99 doesn't consider only titles which were 20 to 29.99 at launch, but also those that maybe were 49.99 to 59.99 at launch and then got a price cut to 20 to 29.99)?
I haven't saved original XLS file, but it's from 900M to 1.1B if I'm not mistaken
It only accounts for the current price of the title. So if a game was 59,99 and now is 29.99, it shows up in 20.00 - 29.99 section on this graph.
I don't have enough historical data to draw this graph based on the original price.
They also have cdp.pl where game costs around 30 euro and activates on GOG.
maniac-kun
Member
******** is banned. Why is this allowed?
Bullhit, made-up numbers vs pulling actual data from accounts which several devs have confirmed are accurate enough.
charlequin
Banned
"Seizure gifs are banned. Why is someone allowed to post this static JPG of a sunset?"
Chartz literally makes up bullshit out of thin air without any sources to back them up and with countless developers publicly admonishing them for presenting outright lies and totally inaccurate sales figures as fact.
SteamSpy does a thorough analysis of publicly available records to create a statistically significant survey that has been verified as accurate by multiple developers.
Welfare
Member
Because chartz is crap and uses completely made up numbers. This isn't.
Turin Turambar
Member
From the latest tweet: http://t.co/EvDZxfz6nc
Dat cluster 31.
But really, it's a rather unsurprising cluster. A few of the others are more unexpected.
Dat cluster 31.
But really, it's a rather unsurprising cluster. A few of the others are more unexpected.
Cluster 48 is scary. It's reading my mind, I tell you!
Yeah, I'm thinking to use this clustering to add new tab "More games similar to this one",
The good thing about this algorithm is that it should properly categorize new games in about three days and at least 2 or 3 thousand copies sold after release. Previous one could only recommend you games that are already popular.
The good thing about this algorithm is that it should properly categorize new games in about three days and at least 2 or 3 thousand copies sold after release. Previous one could only recommend you games that are already popular.
Fragamemnon
Member
Cluster 16, the god-tier.
Hmmm
Completely fake made up stats vs very reliable stats based on actual ownership...
Duh
The number of Witcher 1 and Witcher 2 on Steam is amazing despite both game not being released on Steam from the get go. And also as expected, the Polish audience on Steam for both of this game is small (though most of them actually played the game). I guess it'll be very hard for us to gauge accurate digital numbers on Witcher games due to the fact that Steam copy is also redemable on GoG. I wonder how CDPR measure the number with that overlap.
Yeah, we need GoGSpy. Hopefully CDP introduce a social API with Galaxy at some point
I've talked to them already, they don't want to show their dataYeah, we need GoGSpy. Hopefully CDP introduce a social API with Galaxy at some point
charlequin
Banned
I've talked to them already, they don't want to show their data
I imagine that unlike Valve, CD Projekt actually have a decent incentive to not give people an idea of their total sales volume.
Well, according to my friends, it's around 10% of Steam sales for their indie titles.
But "around" is a pretty broad word, right? It could vastly differ based on the game's exposure on the main page, it's appeal to GOG audience and so on.
If the game is redeemed with Steam key (or Steam photo), they don't count it.
Yeah, I'm definitely keeping on eye on some games numbers.
alexandros
Banned
Indeed. I'd also love to see the chart of overall game ownership after the Steam Sale is concluded, it would be fun to see how many game copies Valve sells in a typical sale period.
So because there is a lag of a couple days in these sales reports, we are now starting starting to see the effect of the Steam sale.
Valkyria Chronicles increased from 300-310k presale to 355k as of June 12th.
The first day of sales wasn't all that great, so most boosts weren't as high. It seems like 25% off newer titles isn't enough to drive huge sales.
Valkyria Chronicles increased from 300-310k presale to 355k as of June 12th.
The first day of sales wasn't all that great, so most boosts weren't as high. It seems like 25% off newer titles isn't enough to drive huge sales.
I did fast calculations based on 1-day sample from today as opposed to three days sample and it's even more impressive: 451,861 +/- 37,367
He, Valkyria Chronicles is also my go-to case on SteamSpy. It's very interesting because we have data from the start, then 2 sales at the same level (where you can see the first one causing a bigger bump than the second), and now we'll see the impact of a subsequent sale at a lower price.
alexandros
Banned
Oh wow, that's great!
ashecitism
Member
These are long term (almost 9 years) sales results though. At the time of its launch it probably didn't do good numbers.