![]() ![]() ![]() Here’s what the streak distribution looked like as we released more maps (with no major changes to our map selection methodology): 10% in the “pseudo-random” ticket method).Īs we introduced more maps to the pool, we saw significant reductions in how often players were encountering streaks of the same map. Here’s what the new distribution of consecutive streaks looked like:Īs you can see, the number of players experiencing a streak of 3 maps or more went down significantly (26% in the “truly random” method vs. However, that 5% chance meant that some players would see the Split 4, or even 5+ times in a row. This could lead to Split having a mere 5% chance of being selected as a map. For instance, a game could have a handful of players who recently played games on Split. In other words, it was less likely for someone to get unlucky and experience a streak, but it was still very much possible to get unlucky. Long story short, we aimed to reduce the number of streaks encountered, but the selection itself was still a weighted “random” choice. It also heavily punished maps that any player had experienced in a streak recently. This method favored maps that were, on aggregate, not seen by the 10 players in a game. ![]() To combat this, we implemented a “pseudo-random” selection system, aimed at showing maps a player hadn’t seen in recent games. We quickly realized that “full random” selection was not ideal-no matter how much you love a map, 5 consecutive games on Ascent can get stale pretty fast. Around 1% of the most unlucky (or lucky?) of you saw the same map 5 times in a row out of 5 games. If you played 5 games during this time period, there was a 26% chance that you’d get the same map 3 or more times in a row. Over the course of 5 games, 26% of players saw the same map 3 or more times. Here’s what the distribution of consecutive map streaks looked like during this time period: Obviously, with only 4 maps available to choose from, players were experiencing consecutive maps quite a bit. All maps had an identical chance (25%) chance of being chosen, regardless of whether a player had played that map recently. When 10 players entered a match, our matchmaker would randomly choose any of the available maps (back then, we had a total of 4 maps). When the game initially launched, map selection was truly as random as it could be. To explain what changes we made and the results of those changes, let’s first take a look at what the old system looked like. As a result, we wanted to make sure that we could improve the diversity of maps played without compromising the health of matchmaking (by influencing things like queue times or match balance). Playing the same map gets stale quickly, and limits the type of challenges that you face in the game. In a recent survey, over a third of VALORANT players responded that it is “Extremely Frustrating” to encounter the same map multiple times in a row. If you’ve ever wondered how it was possible for you to play Icebox for 4 games in a row, you’ve come to the right place!Ī common sentiment that we’ve seen in the past is frustration when you encounter the same map multiple games in a row. This article is going to be a bit different than previous ones in the Game Health series, focused on a very specific topic: map randomness. Hello again! I’m Brian Chang, a member of the Competitive team on VALORANT. You can also read our previous articles on AFKs, and voice and chat toxicity. Learn more about the series in our intro. EDITOR’S NOTE: This article is a part of a series of deep dives on topics in Gameplay Systems, specifically the Competitive/Social & Player Dynamics space of VALORANT. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |