K/D, KDA and win rate: which one actually matters?
Three stats that measure different things, why League and Siege players argue past each other, and which number to trust for judging your own play.
4 min read · Updated 7 August 2026
These get used interchangeably and they should not be. Each answers a different question, and picking the wrong one is how people end up defending a statistic that was never measuring what they claimed.
| Stat | Formula | Answers |
|---|---|---|
| K/D | kills ÷ deaths | How often do I win the fights I take? |
| KDA | (kills + assists) ÷ deaths | How much do I contribute to kills overall? |
| Win rate | wins ÷ matches | Does my team win when I play? |
K/D suits games where a death costs a round
In Siege, Valorant or any round-based shooter, dying removes you until the round ends. Your death is a permanent disadvantage for the rest of that round, so a ratio of kills to deaths tracks something real about how much you cost your team.
That is also why K/D is nearly meaningless in games where you respawn constantly. Dying in a deathmatch costs you a few seconds; dying in a Siege round costs your team a player for the whole round.
KDA suits games with assists that matter
League counts assists because most kills are collaborative. A support who lands the crowd control that makes a kill possible gets nothing under K/D, which is why League tracks KDA and why a support with a 1.5 K/D and a 4.0 KDA is having a fine game.
The weakness is that assists are cheap. Being near a kill counts, so KDA rewards proximity as much as impact, and it can be inflated by hanging back in fights you contribute little to.
Win rate is the honest one, and the least useful
Winning is the actual goal, so win rate is the only stat that measures it directly. The problem is that it is mostly noise in the short term. In a five-player team you control roughly a fifth of the outcome, so a fifty-game sample tells you very little about yourself. Over several hundred games it starts to mean something, which is a long time to wait for feedback.
Use win rate to judge a season. Use K/D or KDA to judge a change you made this week. They operate on completely different timescales.
What to do with all three
- Rising K/D, flat win rate — you are fragging well but not converting rounds. Look at when you get your kills, not how many.
- Flat K/D, rising win rate — you are probably playing a support role well, or making better decisions with the same mechanics.
- Both falling — usually a rank change. You are playing harder opponents, and the numbers are recalibrating rather than getting worse.
- Both rising — the only combination that unambiguously means improvement.