Start with the sample
Check the source, number of matches and date range before comparing two players. A few exceptional games can move a small sample sharply. Steam lifetime totals and a recent match sample describe different periods; they are not interchangeable.
Use the recent matches table to put an aggregate in context. Map, result, opponents, role and game mode can all affect performance. Missing data means a source did not provide that metric, not that a player scored zero.
K/D, ADR and KAST
K/D compares kills with deaths. ADR means average damage per round. They describe different contributions: damage can help a team win even when another player gets the final kill. Read them together rather than treating either as a complete skill rating.
KAST measures the share of rounds in which a player recorded a kill, assist, survival or trade, under the source’s definition. Ratings from Leetify and other providers use different formulas and scales. A rating of 1.0 on one service cannot be assumed to mean the same thing on another.
Aim and reaction metrics
Time to damage describes how quickly damage is dealt after an engagement begins, according to the provider’s measurement. Preaim and crosshair placement describe angles. A lower value may reflect better preparation, but weapon choice, positioning and the encounter still matter.
Headshot percentage, head accuracy and overall accuracy use different denominators. Read the tooltip and source label before comparing them. A high headshot percentage with a rifle is not directly comparable to the same player’s AWP games. Histograms on cs.ninja describe the available match sample, not a global ranking of every CS2 player.
Use performance as context
Look for patterns over multiple games rather than a single outlier. Strong mechanics alone do not establish suspicious behavior. Review rank history, account evidence and the confidence shown with the trust assessment alongside performance.