Valorant Stats Glossary

This glossary defines every stat Riptide surfaces, from the basics every tracker shows (ACS, ADR, K/D) to the context metrics that explain why rounds are won — entry, trades, clutches, consistency, and a composite Impact rating. Each definition is one or two tight sentences, with the corpus average where one exists so you can read your own numbers against the field.

Stats are grouped by what they measure: fragging, entry and trading, clutch and conversion, consistency and utility, economy, and the composite. Two metrics — Impact and KAST — have their own deep guides; everything else is defined in full below.

Fragging

Raw output: how much you kill and how much damage you deal per round. These are role-blind and rank-scaling — use them as a baseline, not a verdict.

Entry & trading

Who takes the first duel, and whether your deaths get avenged. This is the team-play layer that ACS misses entirely.

• KAST — share of rounds where you got a Kill, Assist, Survived, or were Traded. Role-dependent; ~71% is average. See What KAST means.
• Opening duels — your first kills vs first deaths. ~50% win is average, and the team that wins more openers wins ~74% of matches.
• Deaths traded — share of your deaths a teammate avenged inside the 4-second trade window. High = good team trading (the 'T' in KAST).
• Trade kills — share of your kills that avenged a fallen teammate. Low can mean baiting; read it together with K/D.
• Fast trades — the slice of your trade kills landed within 2 seconds (inside the 4s window) — a refrag-speed signal.

Clutch & conversion

Winning rounds you were supposed to lose. Raw clutch win rate is misleading because a 1v1 and a 1v4 are wildly different bets, so Riptide grades clutches against their measured difficulty.

• Clutches — 1vX rounds you won as the last player alive, broken out by how many enemies remained. Pooled win rate is only ~15%.
• CLOE (clutches over expected) — your clutch wins minus the sum of expected wins (CLOE = wins − Σ xClutch). Above 0 means you beat the odds.
• xClutch — the expected win probability of a single clutch, from a model over manpower (1vX), side, and plant state, fit to the live corpus.
• DOE (disadvantage over expected) — the same idea for multi-player deficits (2vX, 3v5): conversions minus the empirical base rate of each state.
• Multikills — rounds with a 2k / 3k / 4k / 5k (ace).

Consistency & utility

How steady your output is game to game, and how much you contribute beyond gunfights.

• Consistency — steadiness on a 0–100 scale, realized ~5–95 (higher = steadier), built from your game-to-game ACS coefficient of variation (spread relative to your average, so big scorers aren't punished for swinging in absolute points) plus how often you have a cold game. Needs at least 5 games.
• Cold game — a game whose ACS falls below 75% of your own average; a self-relative tilt signal, immune to rank baselines.
• Utility — ability casts per round (grenade + both abilities) plus ultimates per round — a read on the initiator/controller dimension. Shown once ability data is ingested.

Economy

Round-win rate split by how much you spent, so you can see whether you over-force or close out full buys. Each round is bucketed by credits spent:

Each bucket reports its round-win rate. A strong eco win rate signals pistol skill and anti-eco discipline; a weak full-buy rate points at execution or mid-round decisions, not aim.

Composite: Impact

Impact is your percentile vs the field on a 0–100 scale (50 = average). It is competitive-only — casual modes never move it — and collapses to three decorrelated axes so one fragging factor can't dominate:

The result is graded into five bands. Full method in the Impact rating explained.

Every raw stat here is free. Riptide Pro unlocks the transformative layer — CLOE/DOE trends, consistency tracking, side splits, and the AI coach that turns these numbers into what to fix next. See pricing.