Is using Popcorn Time without a VPN safe?

For the past year, I have been working on a project that tries to find out if it’s safe to use Popcorn Time without VPN. It is based on the hypothesis that some ISPs are slowing down torrents, but not streaming traffic – this means they would treat traffic from Popcorn Time as “not-torrent” data and leave it unthrottled.

This is based on the findings of a paper by this name. While most ISPs don’t discriminate between traffic types, there are some that do – for instance in New Zealand ISP Slingshot was caught throttling BitTorrent but not HTTP. I’ve also mined user data from Popcorn Time’s official subreddit to see how many people are using VPN with Popcorn Time and to see if there’s a correlation between that and ISP throttling. For more details on this, you can read the technical discussion at the bottom of this post, feel free to skip it if you don’t care about the details.

The results I have collected so far come from both my own speed tests and the official subreddit for Popcorn Time. I started running my own speed tests in December, but due to changes made by ISPs over the past year, it was not possible to compare results. The data from this post is limited to April 2016 onwards, though it does include some of the older data collected using a different metric.

The first set of results used the number of peers and total download speed (in KB/s) to find out how many people can be downloaded from at what speed in a certain ISP. For comparison, I also ran this test with three different VPN providers: Private Internet Access, Kepard, and TorGuard. ISPs are split into seven groups depending on their download speeds.

The second set of results looked at how many people in a certain ISP are using VPN. While the number of users is around 10%, these proxies could be shared between multiple people which means that not all Popcorn Time users would necessarily be anonymized by a VPN. Keep this in mind when looking at the numbers.

Please note: These are the results of two separate tests. The first one includes all users, while the second uses only active users to get more accurate data.

Fantastic results – looks like Popcorn Time has a bright future ahead!

Unfortunately, this is not good news for most US citizens. This table shows that ISPs in the US do not discriminate between different types of traffic.

Since the number of users in certain ISPs is too low, I ran a correlation test to find out if there’s a relationship between how many people are using VPN and the download speed offered by an ISP. In this case, users that have been active for more than 1 month have been used, but no other restriction has been applied.

As expected there is a negative correlation between the number of VPN users and overall download speed. In other words, ISPs throttle Popcorn Time traffic when there are fewer users using VPNs. This doesn’t mean they will never throttle your connection – it only means that if you want to use Popcorn Time without VPN, choosing an ISP with high download speeds is not the way to go.

There was also a positive correlation between the download speed and VPN use, which means that ISPs that provide faster connections are more likely to favor Popcorn Time users over regular traffic. If you want an ISP that won’t throttle your connection while using Popcorn Time, consider getting one of those. There are lots of options in most countries, including the US.

While we don’t know for sure what causes these correlations and what is actually happening under the hood, I think it’s safe to say that throttling happens and that VPN use makes Popcorn Time traffic stand out less from others – at least when there are more people using them. If you’re an ISP, the numbers are clear – if you want to reduce Popcorn Time use your best option is not to throttle it but to make high-speed Internet available.

These tests don’t include UDP traffic because I didn’t have access to a server that allows UDP connections. This means that I won’t be able to provide estimated speeds for VPN users that use UDP – which is usually the best choice for streaming.

The next round of tests will include active users regardless of how long they have been active and will also include UDP data, so stay tuned.