Tinder Changes Its Ranking Algorithm, and The New Score System Is… Basically The Same?

As you may have heard, Tinder recently announced a change to their ranking algorithm, or the way it determines the attractiveness of its users to match them with people of a similar desirability. According to them, they have replaced their outdated [sic] elo score system – explained in detail in this SwipeHelper guide – with a new, improved algorithm. So, do we have to rethink how we use Tinder to make the best of this new system? Let’s start from the top and see what Tinder had to say:

Tinder’s explanations on their new ranking algorithm

What really matters, apparently.

activity-is-what-really-matters-misleading-tinder-algorithm-bs

They start us off with the mind-blowing revelation (according to them), that being active on the app is the most important factor in improving your chances of matching, as they put the profiles of active people front and center. While it is true that your placement in other users’ decks is pushed further and further back the longer you stay inactive (as explained in the algorithm guide), the only metric used for that is “time since last active”, and coming online after a two week hiatus will put you in the same position you’d been in if you’d been active all this time.

We used to be able to see how long ago someone was last active in older versions of the app. So yes, being active helps you get more matches (duh), but to say it’s the most important factor is a bit misleading, especially when it comes to the implication that it has anything to do with your desirability score or what kind of matches you can expect. Visibility is not the same thing as “quality” of suggested matches.

 

Away with the old…

elo-no-longer-relied-upon-tinder-algorithm

Okay, so the Elo score is old news. Just a quick clarification, the way it worked was similar to point systems in WoW, or Warcraft, among others: Your desirability score was adjusted based on how many likes you received and what the score of the person liking you was. If someone desirable (according to their score) liked you, that had a greater effect than an undesirable profile liking you. I think that already gives a better picture than the official explanation, but for the long version, see here.

If so, that means we’ll have to adjust our behavior strategies to appeal to the new algorithm, if we want to maximize our number and quality of matches. Or does it?

 

…in with the new

new-system-equals-old-system-tinder-algorithm-bs

So Tinder wrote a blog post to clear up any questions the community might have about their (old and new) ranking algorithm and this is the information we get on the new system. There you have it: The new ranking system adjusts the potential matches you see every time your profile is liked or noped, except any changes will only become visible withing 24 hours or so. Well that tells us… exactly nothing at all.

Seeing as this has already been effect for a while, and we haven’t really noticed a difference, it seems the new system is an elo score by any other name (Glicko and Trueskill come to mind). It would make sense that you are still rated by how many likes you receive by whom, and our experience using the app would confirm that. It’s not like we suddenly see all kinds of users regardless of our own attractiveness.

One of the possible changes to the matching algorithm that seems feasible would be the recent implementation of Amazon’s AWS image recognition software, which Tinder uses to create their Top Picks recommendations that are supposed to present you with both aesthetical types and characters you’d be interested in (and failing hilariously at it by the way). In conjunction, they seem to be using text recognition to analyze users’ bios.

By extension, it would make sense for them to implement this type of information in their matching algorithm as well, adding your personal preferences to your desirability score. This further zeroing in on a target audience to which you get shown could also be responsible for most male users getting even fewer matches than ever before in recent months.

So, what should you do with this information?

thats-all-folks-or-is-it-tinder-algorithm-bs

Basically, you can continue using the app as before. The same old algorithm rules appear to still be in effect. We might have to start calling the elo score something else, but for the time being, the change of ranking system seems to be a distinction without a difference.

One option, if you’re among those suffering from the ever decreasing number and quality of matches, and/or you’re sick of an over complicated algorithm determining whom you can see and with whom you even have a chance of matching, would be to give the competition a try: Both Bumble and Hinge make for good alternatives and may improve your dating life drastically depending on where you live.

And that’s all, folks. Any questions or remarks? Feel free to leave a comment below or visit us on the SwipeHelper subreddit. See you there 🙂

4 Comments

  1. One thing I have been meaning to test: I’ve been noticing that I get way more matches at exactly the time I am actively using Tinder and swiping on it. This matches up with what Tinder says in this blog post: “We prioritize potential matches who are active, and active at the same time. We don’t want to waste your time showing you profiles of inactive users. We want you chatting and meeting IRL.”

    • What it matches up with is the fact that tinder shows the profiles of more recently active peolpe first, all things being equal, but that’s nothing new.

  2. Recently I experience more shadow bans for, as it seems, not talking to new matches or being removed by one. Though probably it’s not only that.
    Do you know if it’s only men getting punished for not engaging in chat or only the person that made the match (liked as a second) and how to avoid getting shadow bans (they seem to last for few days, in which I see low ranking profiles and seldomly or not at all get liked)?
    Are there any more reasons algorythm would ban me?

    • I always suspected that, I only use tinder like Pokemon Go and after reaching 130 matches and +99 likes the matches dropped off a cliff (only matched with superlikes after that) and my likes are now down to 14.

Leave a Reply

Your email address will not be published.


*


This site uses Akismet to reduce spam. Learn how your comment data is processed.