Stats show that every Valorant map favors the defense

A breakdown of Valorant's kill-to-death ratios and side win rates show that the game's defensive players kill more often and win more rounds than attackers, often by large margins.

Players have thought that Valorant's attacking side is the game's more difficult side to play, and now map statistics show that there is a lot of truth behind that feeling. Statistics sites all show that on each of Valorant's four maps, the defense is heavily favored. Defenders have a better overall K/DA than the attacking team by a large margin, and defensive round win rates outperform the same statistic on the offensive side.

This is at least somewhat unusual. Most maps found in competing games such as Counter-Strike: Global Offensive are usually split in whether they favor attacking or defending teams. But in Valorant, the defensive side has a clear advantage across the board.

Defense is better and easier than offense in Valorant

  •  Ascent:
    • Attackers
      • 46.3% round win rate
      • 1.01 total K/D
    • Defenders
      • 53.3% round win rate
      • 1.12 average K/D

  •  Bind:
    • Attackers
      • 46.7% round win rate
      • 1.01 total K/D
    • Defenders
      • 53.3% round win rate
      • 1.12 average K/D

  • Haven:
    • Attackers
      • 48% round win rate
      • 1.01 total K/D
    • Defenders
      • 53.3% round win rate
      • 1.11 average K/D

  • Split:
    • Attackers
      • 46% round win rate
      • 1.00 total K/D
    • Defenders
      • 54% round win rate
      • 1.12 average K/D

Objective-based games such as Valorant and CSGO often feature maps that favor one side over the other, but Riot has created maps that all favor the defensive side. This isn't necessarily a problem, and is a trend that tactical shooter games often show early on after release. While CSGO's Nuke has always been a map that heavily favors defending teams, the longer a map is in a game's rotation the closer round win rates come together. It takes time for players to fully understand and exploit a bombsite's weaknesses, and that experience also comes with a more evolved gameplan for the attacking side.

The more interesting information that players can take from these statistics is that Valorant developer Riot Games has a history of balancing anything that falls outside of 50% win rates. Riot is known for buffing or nerfing champions in League of Legends when that champion's win rate strays from the mean by even a small amount. So Valorant's clear favor of the defensive side seems strange given its developer's history, but the game's characters might be part of the reason for the difference between sides.

Valorant's abilities could explain defense win rates

Agents including Cypher and Sova bring abilities that can reveal enemy locations just seconds into a round, allowing defensive players the ability to make early rotations or leans and quicken the defense's reaction times to attacking plans. A well-timed Sova arrow, a cooldown ability that recharges during the round, can be instrumental at any point during the match. Cypher's Neural Theft reveals an enemy team entirely.

When used tactically, the defense gains massive advantages at minimal cost. While players in CSGO have to risk their lives for such information on an enemy's whereabouts, Valorant's abilities offer that same information almost for free.

While this isn't necessarily a good or bad thing, players switching from Valve's shooter to Valorant's ability-based game should use those abilities to their full effect. Informational abilities are king in Valorant, with Cypher or Sova picked in 100% of matches at Valorant's recent Pop Flash tournament.

Valorant allows players information for free to a player who knows what they're doing. And even though Valorant only uses damage inflicted as a player performance metric when deciding to increase a player's rank, match wins are still the best way to ensure players can hit Radiant. Information equals wins in all shooters, and information is easy to lockdown in Riot's new shooter.