The only comparison that means anything
"Is 7,000 a good Gridshot score?" has no answer, for the same reason "is 42 a good golf score" has none until you say the course and the par. Every trainer counts its own way: dots hit, targets per second, weighted accuracy, timing bonuses. A score of 80 in one app and 8,000 in another can describe the same hand. So the first rule of reading any aim score is to compare it only against the same drill, on the same trainer, at the same settings, or against your own history.
That is why the Benchmark is built on fixed pars rather than player data. Each drill's score is divided by a transparent reference value, capped at 1, and the four ratios are averaged into a rank. The pars are engineering choices, score-versus-reference marks in the same spirit as par on a golf hole, not a survey of how other players perform. Nobody is looking at a leaderboard behind the scenes. There isn't one.
What the four numbers are for
The Benchmark runs your four core skills for 60 seconds each, precisely because they fail differently. Gridshot is speed: how fast your hand moves between respawning targets. Precision is accuracy: whether the little corrections land. Tracking is sustain: whether you stay on a moving bot. Flick is reach: how quickly you solve wide angles.
| Drill | Par | Reads |
|---|---|---|
| Gridshot | 7000 | raw speed and reset between kills |
| Precision | 4200 | micro-corrections on small targets |
| Tracking | 4500 | sustained fire on a strafing target |
| Flick | 8000 | large corrections to distant targets |
Average the four ratios and the ladder reads: NOVA at 0.90 and above, DIAMOND 0.78, PLATINUM 0.62, GOLD 0.45, SILVER 0.28, BRONZE below. Two notes on that ladder. It is deliberately demanding, since NOVA means averaging within ten percent of every par, including the ones where you are weakest. And the number under each tier is a ratio to fixed references, not a percentile: "Diamond" describes your distance from four constants, not the distance between you and another player. Nothing here is a scouting report.
"Good" mostly means "stable"
Once you accept you're only comparing to yourself, the useful property of a score changes: the spread matters more than the peak. A player whose Gridshot lands between 6,600 and 7,000 every run has a stronger, more usable skill than a player oscillating between 5,800 and 7,400, even though the second player owns the better single run. Competitive CS2 rewards the floor, because that is the version of you that shows up in round eleven of a close match.
The trainer keeps the last fifty runs per drill with a sparkline, and that trail is the real test. When you review a week, look at the band your scores live in, not the spike on Tuesday. If the band is creeping upward month over month, the training is working long before any single new personal best shows it.
A good score is one you can reproduce on a mediocre day. Everything above that is bonus, and everything below it is weather.
How to read a bad day
You'll have runs that come in 15 percent under the band, and they will arrive exactly when you feel most ready to draw conclusions. The short list of what to do: note it in the history, leave your sensitivity alone, do not "fix" your grip mid-run, and check the next session before believing anything. Sleep, food, tilt and mousepad temperature move scores more than a Tuesday-to-Wednesday technique change ever could.
Three bad sessions in a row is a different signal, and that one is worth acting on. Run the Benchmark fresh after a real break, and if the whole board is down together, look at training load or life stress before you look at your aim technique. If only one drill is down, that is the rare case where the number is telling you something mechanical; give that drill the three-set treatment from the improvement plan for two weeks and re-test.
What a score cannot prove
This part matters more than any number above it. A benchmark measures one hand moving a crosshair, and CS2 kills happen for dozens of reasons besides that. Not measured, not even approximated:
- Crosshair placement: where your aim starts before the duel. A stable Diamond hand with lazy placement loses to a Gold hand that is always already on the head. That's its own guide, because it deserves one.
- Decisions: which fights you take, when you fall back, whether you re-peek. Unmeasurable in a range and worth more than the last five percent of flick speed.
- Pressure: the same hand, watched by four teammates and an enemy AWPer, is not the hand that produces clean ratios at 2 a.m. Perfect proxies for this don't exist; the closest is scrimming.
- Movement and util: counter-strafing into a real fight, smoke timings, flashbangs, the whole layer of CS2 that decides engagements before anyone shoots.
So treat the rank like a smoke detector: it tells you, cheaply and honestly, whether the hand is being trained. It will never tell you that you are good at Counter-Strike. When a score and a match result disagree about you, the match is more right, but if the score is falling while the match is fine, believe the score and get some rest.