Your ADR is fine and you still lose. Here is why.

A 74 ADR at Level 9 with a 50% win rate is one of the most common profiles there is, and it is genuinely confusing to the person living in it. The damage is there. The kills are there. The wins are not.

What ADR actually measures

Damage dealt, divided by rounds played. That is it. It does not know whether the damage arrived at a useful moment, whether it was traded, or whether the round was already decided.

Here are the real medians, measured from 435 public FACEIT profiles rather than estimated:

Level	1	3	5	6	7	8	9	10
Median ADR	61.6	74.4	75.4	81.1	80.7	81.2	81.7	82.5

Sitting at the band for your level means your mechanics are level-appropriate. It says nothing about whether you win.

But look at the shape of that row, because it is the most important thing on this page.

ADR stops discriminating at about Level 6

From Level 1 to Level 6 the median climbs 19.5 points. From Level 6 to Level 10 it climbs 1.4.

Ninety percent of ADR’s entire explanatory power is spent in the bottom half of the ladder. Above Level 6 it is flat: the median Level 7 and the median Level 9 do within a point of the same damage per round, and they are 400 ELO apart.

The row is not even monotonic — Level 6 medians above Level 7, and Level 4 above Level 5 — which is what a metric looks like when the differences between adjacent groups have shrunk into the sampling noise of a 40-profile median.

So if you are Level 7 and wondering why you are not Level 9, ADR cannot answer the question. It is not that your damage is too low. The players a level or two above you are not doing more of it.

For contrast, over the same range: opening duel win rate moves from 50.6% to 52.4%, and headshot percentage from 47% to 51% — the only column in the whole table that never goes down. The full measured table is here.

Where damage and winning come apart

Damage in lost rounds counts the same as damage in won rounds. Deal 250 damage across a round your team loses 5v2 and your ADR looks great. This is why players who die late in losing rounds — after doing chip damage on the way out — post inflated numbers.

Unconverted damage is nearly worthless. Eighty damage on someone who escapes costs them a moment and costs you a position. The scoreboard credits you for the eighty.

Damage without trades leaks value. You do 90, you die, nobody kills the enemy, they take the site with 10 HP. Great ADR, lost round.

Timing beats volume. 40 damage in the opening five seconds that lets your team take space is worth more than 120 in a post-plant you lose. Nothing in ADR knows the difference.

Stats that predict winning better

Opening duel balance. First kills minus first deaths. Strongly tied to round win rate, because whoever wins the opening plays the rest of the round a man up.

1v1 clutch rate. Measures decision-making with the round on the line and no team to carry you.

Utility damage per round. A proxy for whether you contribute in rounds where you never fire a shot.

Trade rate. How often your death gets traded — partly your team, but mostly a function of where you choose to stand.

None of those are in the big number on your profile, and all of them separate a 50% player from a 57% player far better than ADR does.

Two players, same ADR

Player A: 74 ADR, opening balance +150, 1v1 at 33%, utility damage 6.5. Wins the opening, struggles to close, contributes when not fragging. Fix the endgame and this player climbs.

Player B: 74 ADR, opening balance −190, 1v1 at 41%, utility damage 3.5. Loses the fights they start, wins the fights they choose, adds nothing on non-fragging rounds. Fix the role and this player climbs.

Identical ADR. Opposite problems. Opposite fixes. Give either one the other’s advice and it makes them worse — which is exactly what generic improvement content does, and why “work on your aim” is the most expensive free advice in Counter-Strike.

Both of those players exist, and here they are

Player A and Player B above are constructed to make the point. These are not — they are real profiles from our worked examples, anonymised:

A Level 8 at 74.0 ADR, 1,279 matches, 51% win rate. Their ADR is 7 points below the Level 8 median of 81.2, which by the usual reading makes them a weak Level 8. But their opening duel rate is 17.5% against a 20.4% median — they are not overextending — and they are 20 ELO from Level 9. The report’s top finding for them is utility output at 3.5 per round against a 4.57 median, worth an estimated 1–2 points of win rate. Nothing about their aim is the issue.

A Level 10 at 79.4 ADR, 522 matches, 57% win rate. Below the Level 10 median on ADR, and winning more than anyone else in our set. What they have instead is map mastery — Inferno at 67% over 52 matches — and what they have wrong is that they take 26.2% of their team’s opening duels against a 19.3% median while winning only 47.1% of them.

Two accounts, both below the ADR median for their level, one at 51% and one at 57%. The number did not predict either.

The uncomfortable version

If your ADR matches your level and your win rate is 50%, mechanics are not your problem. Aim training will not move it, because there is nothing there to move.

What is left is decisions: which fight, when, with what information, having bought what. Those are unglamorous, they are not satisfying to practise, and they are worth several percentage points of win rate — which, as the ELO maths shows, is the entire difference between stuck and climbing.

When ADR is the right thing to look at

None of this makes the number useless. It makes it a measure of one range and one question.

Below about Level 6, it is the most useful stat you have. The medians climb 19.5 points from Level 1 to Level 6, which means real headroom: a Level 3 player at 65 ADR genuinely does have a mechanical gap to close, and closing it genuinely does move their rank. Aim practice pays in this range, and it pays visibly.

As a floor check at any level. If your ADR is more than about eight points below the median for your level, mechanics are plausibly a constraint and worth some time. If it is within a few points, they are not, and the same hour spent elsewhere returns more.

As a tracker of your own change, against yourself. Your ADR six months ago compared to now is a cleaner signal than your ADR compared to anyone else’s, because the role and the map pool are held roughly constant.

What it cannot do is tell you why you lose. That question needs the numbers that are still moving at your level, and above Level 6 those are opening duel conversion and headshot percentage — not the one printed largest on your profile.

Find out which player you are

Opening balance, clutch rate and utility output, read together rather than one at a time: run your profile.