2020 was a big year for VR on Steam with the release of new headsets, Valve’s flagship VR title Half-Life: Alyx, and the Coronavirus pandemic which drove large increases in gaming. These factors combined pushed monthly-connected VR headsets on Steam to pass the 2 million milestone for the first time.

Monthly-connected VR Headsets on Steam

Each month Valve collects info from Steam users to determine some baseline statistics about what kind of hardware and software is used by the platform’s population, and to see how things are changing over time, including the use of VR headsets.

The data shared in the survey represents the number of headsets connected to Steam over a given month, so we call the resulting figure ‘monthly-connected headsets’ for clarity; it’s the closest official figure there is to ‘monthly active VR users’ on Steam, with the caveat that it only tells us how many VR headsets were connected, not how many were actually used.

While Valve’s data is a useful way see which headsets are most popular on Steam, the trend of monthly-connected headsets has always been obfuscated because the data is given exclusively as percentages relative to Steam’s population—which itself is an unstated and constantly fluctuating figure.

To demystify the data Road to VR maintains a model, based on the historical survey data along with official data points directly from Valve and Steam, which aims to correct for Steam’s changing population to estimate the actual count—not the percent—of headsets being used on Steam.

Reviewing the data we can see that the number of monthly-connected headsets on Steam peaked above the 2 million milestone for the first time in April, no doubt thanks to the launch of Half-Life: Alyx, but also due to the onset of the Coronavirus pandemic which drove record numbers of gaming on Steam throughout the year.

While the threshold was first crossed in April from the Alyx spike, the last four months of 2020 were consistently above the 2 million mark, with the all-time high now set at 2.3 million in November.

To put the milestone into perspective: it took 40 months for Steam to reach 1 million monthly-connected headsets, but just 7 months after that to reach 2 million. Year-over-year, the number of headsets is up 94%.

The growth of monthly-connected VR headsets on Steam has stuck fairly close to an exponential curve thus far, with an R² value of 0.954. A naïve projection (simply drawing out the line), suggests the number of monthly-connected headsets could hit 4.5 million by the end of 2021.

Share of VR Headsets on Steam

As for the share of headsets on Steam, we can see competition heating up over the course of 2020. Whereas at the start of the year the share was largely dominated by three headsets—Oculus Rift S, Oculus Rift, HTC Vive—by the end of 2020 there’s five headsets vying for the bulk of VR users—the aforementioned three, plus Valve Index and Oculus Quest.

From a vendor standpoint, the growth of Oculus Quest alongside Rift S and Rift has allowed Facebook headsets to suck up just over half the overall share of headsets on Steam. HTC’s headsets still hold onto 20.8% of the market, but have continued to see a slow but steady decline.

Windows Mixed Reality, long past its peak share of 11% in mid-2019, now holds just just 5.5% of headsets on the platform. With the launch of HP Reverb G2 at the very end of 2020 however, we could see WMR’s share start to tick back upward.

– – — – –

Beyond these numbers, Valve also shared a handful of interesting statistics about VR usership on Steam in 2020:

  • 1.7 million first time VR users were added
  • 104 million VR sessions total
  • Average VR playtime: 32 minutes (+30% year-over-year)
  • VR game revenue increased by 71% (Half-Life: Alyx contributed 39% alone)
Newsletter graphic

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. More information.

Ben is the world's most senior professional analyst solely dedicated to the XR industry, having founded Road to VR in 2011—a year before the Oculus Kickstarter sparked a resurgence that led to the modern XR landscape. He has authored more than 3,000 articles chronicling the evolution of the XR industry over more than a decade. With that unique perspective, Ben has been consistently recognized as one of the most influential voices in XR, giving keynotes and joining panel and podcast discussions at key industry events. He is a self-described "journalist and analyst, not evangelist."
  • Christian Schildwaechter

    A naïve projection (simply drawing out the line)

    That is not naïve, it is wrong. You are assuming that the data follows a continuous growth trend, which clearly it doesn’t. Basically you have much slower increase up to Half-Life: Alyx, then a huge and sudden jump, then again a slower increase. If you include outlier events like HLA for your projection, you are implying that these events will occur regularly, instead of counting them as one time effects.

    Why is this wrong? Imagine someone trying to determine in 2025 if people adopted healthier lifestyles by looking at the mortality for the last 10 years. In most countries life expectancy has been slowly increasing for decades, so you’d expect to see this again. Only this time you’d have a pandemic in the middle, which seriously screws up mortality numbers. If you simply integrate the outlier value, you might get a completely wrong picture, e.g. a falling life expectancy over ten years. Which is why extreme data points are often ignored in statistics, e.g. everything below 5% or above 95% of the average. Otherwise these statistical artefacts distort the results and may lead to a false result.

    Now this is nothing new. Road To VR has drawn overly optimistic trend lines into these graphs for years. And somehow it always ends in a massive increase in VR user numbers that is just around the corner. Maybe you should go back a few years and check whether your optimistic projections were even close to what actually happened. And if not, maybe a) stop publishing naïve projections based on curves that are not even based on the data points and b) talk to someone who understands statistics.

    • Matthew Lake

      Companies’ are always using these kinds of projections to determine what is projects are viable for the market in x number of years, not what is viable today. Based on the trend that has persisted for many years now, it’s not that unrealistic or even crazy to expect continued growth which will closely match the projection above.

      The tipping point where VR growth will really take off was expected to be around about 2021. A slide from Unity conference in 2016 about how the growth of the VR market will look like compared to the unrealistic expectations people had.

      Keep being pessimistic, VR is still going mainstream soon. And let’s see what happens in 3 months or 6 months! My prediction is that it will continue to follow this trend and we will hit 4+ million on steam by January next year.



      • Christian Schildwaechter

        Based on the trend that has persisted for many years now,

        Please draw a trend line through the data points that followed the Half-Life: Alyx peak. Does this line follow the projection? Does it indicate that there is exponential growth?

        it’s not that unrealistic or even crazy to expect continued growth which will closely match the projection above.

        Let’s say we are looking at a time of five years/60 months with a monthly growth of 2% (you should like this, because this would be exponential growth, adding more than 25% new users every year).

        Somewhere in the middle of these five years there are two months with a massive increase of 40% per month due to a special event. So now we have 58 months of 2% and two of 40%.

        The average growth per month over five years is 3.27% (47%/year).
        The median growth per month over five years is 2% (27%/year).

        Which of the two do you think is a better projection for future growth?

        I’m not pessimistic. I just do the math.

      • Blaexe

        At this point, we really shouldn’t throw “PCVR” and “VR” into the same basket. VR as a whole will see significantly higher adoption – but it will be driven by standalone VR.

        PCVR will likely see slower growth, which the data basically implies already.

        • Christian Schildwaechter

          Interestingly we can derive the growth of standalone VR (Quest) via PCVR, because the Steam usage data lists the Quest as a separate HMD. We don’t know how many Quest users actually stream from a PC, it could be 10% or 50%, but the percentage is most likely somewhat stable. Which means that the relative marketshare is an indicator for how many Quests are in use.

          Currently the Quest is at 16% on Steam. If every Quest owner used streaming via Steam, this would mean that 1 out of 6 headsets was already a Quest. If every second Quest owner uses Steam, 1 out of 4 HMDs is a Quest. And if only 10% of the Quest owners stream Steam, then today 2 out of 3 VR headsets are already a Quest. My guess is that currently 10% of streaming Quest users is more realistic than 50%.

          The number of Quest owners streaming from Steam obviously changes with a more/less technical audience, improved quality and ease of use leads to more streaming, while a larger number of high quality native Quest leads to less streaming etc. The method only compares relative development of PCVR and Quest and has many other limits.

          But it reveals trends in Quest sales and usage. It shows an steady increase for Quest 1 over five months, followed by five more of staying at about 11% until the Quest 2 was released. This lead to a huge jump to 16% in the first month, then a small decrease in the second. This may be due to e.g. the Quest 2 being sold out or most enthusiasts buying on day one. The numbers for January will be interesting, showing the impact of Quests given as presents.

          • Blaexe

            I’d say that even 10% is way too high. Rumor says that Facebook ordered 3m Quest 2 for 2021 alone. Given that it was basically sold out, they probably have sold more than 3m Quest 2 by now. Even 10% of Quest 2 users on Steam would have a stronger impact than what is currently reported.

            Quest 2 is absolutely marketed as standalone headset. I’d be surprised if more than 5% of Quest 2 owners would use it to play PCVR.

          • Rogue Transfer

            3m Quest 2 sold is impossible, due to the fact that the Google Play Store installs for all Go, Quest 1 & Quest 2 users still sits at 1m+ and thus hasn’t reached the next milestone of 5m+. From what estimates there are for these headsets: Go achieved ~2.5m; Quest ~1.3m; leaving only up to ~1.2m for Quest 2.

            Now of course, there is also the Apple store Oculus app, where Apple phones have around 25% of the market to Google’s 75%, so you’d need to adjust the figures above for that. But even still, you can see clearly there is no possibility of 3 million Quest 2 alone.

          • Blaexe

            2.5m Oculus Go? Sorry, that seems made up. Do you have a source?

            I also don’t think 75% android is accurate. In the US, it’s more like 50/50 and generally VR owners are probably not exactly the ones using low end Android smartphones which also drive market share.

            The 3m Quest 2 figure comes from internal sources that have been right about a lot of things prior to Quest 2 release and reveal.

          • Ad

            This is headsets in use, which you’re comparing to total Quests. The Quest is an impulse buy being sold to casual consumers. The number of active users will be between 50% and 75%, not to mention people who don’t buy anything when rec room, sidequest, and vrchat exist.

    • benz145

      I understand what you’re saying—the exponential fit may not continue to hold true into the future, and the projection has no modeling beyond ‘what would happen if the current trend continued?’, which is clearly stated.

      The first time we looked at the data, saw the fit of the curve, and drew it out was at the very start of 2019. At the time the curve would have predicted 1.6 million monthly-connected headsets by January 2020:


      With all the additional data points we have since that date has come to pass, our best estimate of the actual number of monthly-connected headsets as of January 2020 was 1.3 million. Considering all the variables, I’m pretty happy with that result.

      I think there’s a fairly intuitive reason that we might expect something like an exponential curve. VR adoption is not happening in a void. While the portion of VR users is growing, so too is the number of gamers in general.

      Steam itself grew 26% from December 2019 to December 2020. On top of that growth, the portion of Steam’s population that’s using VR headsets grew 56% from December 2019 to December 2020.

      Beyond both of those factors, there are companies which are actively working to accelerate the growth of VR, which has led to leaps in hardware and software. While we may not get an Alyx or an Index every year, there is much more room for innovation in VR than the overall gaming market which is far more mature.

      Thank you for the feedback and suggestions, I’m interested to know what the projection would show with some statistical modeling applied and will look into it.

    • Matthew Lake

      We’re at about 2.6 Million connected headsets now based off the 2.13% of steam using VR last month? Still following the trend…

  • Rudl Za Vedno

    Companies will have to show better VR tech in order to achieve mass adoption. As VR enthusiasts we tend to lack objectivity. I have Q2, Odyssey+ and Reverb G1 in my collection, but my kids still prefer playing Fortnite and pancake social media platforms over using VR. Kids are the litmus test VR needs to pass first in order to see mass adoption imo.

    • Ad

      We should put the whole “mass adoption” meme to rest. It has never been helpful.

      • Sandeman21

        Denial is worse. ;)

    • Ray Pope

      I think a lot of the problem is only one person can play at a time. Consoles are way easier to market as a family choice.Consoles are able to entertain more than one kid at a time without the others crying for their turn. Also kids need supervision with VR so they don’t break anything. I think when they get to the size of a pair of swimming goggles where they can sell two pairs in the same bundle at an affordable price a lot more people will buy into VR.

  • Love my G2. Incredible.

  • Ajedi32

    Did anyone ever figure out why there’s suddenly so much oscillation in the “Steam users with VR Headsets” stat in the Steam Hardware Survey? It definitely seems to be related to the counting methodology change back in February 2019, but not sure exactly how, or why the oscillation is so predictable (it’s been consistently down one month, up the next for the entire year now).

    • benz145

      Yeah it’s almost certainly related to the collection methodology, but I’m not sure exactly why. I’ll ask Valve at the next opportunity.

  • Sandeman21

    One thing is certain, scaremongering and pseudo-privacy bullshit is not going to make other headsets sell more. It only makes the most desireable headsets sell less, and that is a bad thing for VR: