Note: this is my first Reddit post like this so sorry if its off
I started feeling like the numbers were off about two years ago. It seemed like I was seeing way more Linux users in games then should be possible based on the numbers, it felt like more then 3% where openly using Linux so the true number must be higher.
I took a closer look at the numbers on steam and I saw a pattern. Overall, 26% of steam users had Chinese set as their language. But, only 2% of Linux steam users had their language set to Chinese, thats a huge gap. so I ran the numbers and saw that the amount of English Linux users was at 6%! around 25% of that is from the steam deck, but I feel like around 25% of linux users dont have steam, so that 6% could be used in a more general desktop market share.
I did that around 2 years ago and redid it today, and the number is now 11%! so if my calculations are correct, about 11% of English steam users are on Linux, meaning we might be about to cross the threshold where companies start caring about us.
But who knows, I might be wrong about all of this. I just wanted to know what you guys thought.
Edit (an explanation of the methodology)
the data is from the public steam hardware and software survey https://store.steampowered.com/hwsurvey .
Normalized step 1 is the language% divided by the total % of users of that os. ie. 82.33% of linux users speak have their language set to English, 2% of steam users use linux, so the output is 2.37% (the exact command used is =(C2/100)*(C5/100)*100)
Normalized then changed the percents from % of users per os, to % of steam users total (gotten by dividing the % of language+os by the % of the language in total, so for the one in question its =(C8/100)/(E8/100)*100)
I thought I link I gave had the calculations I used because they are there in the main app, but I guess not.