What this test measures
Point at a thing, click it, repeat. That is the whole task, and it sounds trivial until you watch your own numbers. This aim test shows you thirty circular targets, one at a time, each appearing at a random spot inside the panel. You click each one as fast as you can, and the moment you hit it, the next appears. Your score is how many targets you hit per second, with the average time per target in milliseconds next to it. A click that lands inside the panel but off the target counts as a miss. The page shows your accuracy next to your time.
Two separate abilities are being combined here. The first is visual search: your eyes have to find the new target, which appears somewhere you cannot predict. The second is motor execution: your hand has to translate what your eyes found into a fast cursor movement that ends inside a 48-pixel circle. Our reaction time test measures a simpler kind of speed. There, you answer one signal with one press, and no aiming at all. This test is about pointing, which is a different skill with different bottlenecks.
It also forces a trade-off. Move faster and you miss more; slow down for accuracy and your average climbs. Psychologists call this the speed-accuracy trade-off, and it is the reason the page shows both numbers. A quick average with mediocre accuracy and a slower average with perfect accuracy describe two different strategies, not necessarily two different levels of skill.
There is no time limit, but the clock keeps running until each target is clicked, so a pause lands straight in your average. If you get interrupted mid-run, finish it, ignore that score and run it again.
Why there is no percentile here
This site follows one rule about percentiles: they appear only when a published research norm exists for the task, measured the way you play it. Even simple reaction time, with decades of laboratory data behind it, carries no percentile here — browser and device delay varies too much between setups for that comparison to be honest. For a browser aim test with 48-pixel targets, thirty trials, and random placement inside a panel, no published normative distribution exists. Nobody has measured a proper sample of people on this exact task and published the results. So this page shows your score, your personal best, and a nickname leaderboard — but no percentile.
Inventing a distribution would be easy. Plenty of sites display a smooth curve and tell you that you beat some precise-sounding share of players. Those curves come from whoever happened to visit, or from nowhere at all. Their meaning flips depending on who is in them — shooter-game veterans, or people on laptop trackpads. We would rather show you nothing than show you a number we cannot defend.
What the page does track is the comparison that actually holds up: you against yourself. Your personal best and your history come from the same hands, the same mouse, the same screen, and the same panel. When your average drops after a week of practice, the improvement is real. When a stranger posts a faster score, you learn nothing, because you do not know what hardware they were holding.
Fitts's law: why target size and distance decide your time
In 1954, a psychologist named Paul Fitts had people tap back and forth between two metal plates with a stylus. He kept changing how wide the plates were and how far apart they sat. He found a strikingly regular pattern: movement time grows as the distance to a target grows, and shrinks as the target gets wider. The relationship is logarithmic, which means doubling the distance does not double your time; it adds a roughly constant amount. The original paper is Fitts, P. M. (1954), "The information capacity of the human motor system in controlling the amplitude of movement," Journal of Experimental Psychology, 47(6), 381–391, and it is one of the most repeated and confirmed results in psychology.
The intuition behind it: a pointing movement is not one smooth motion. You make a fast, coarse movement toward the target, then small corrective adjustments at the end. A distant target lengthens the first phase. A small target demands more of the second, because the tolerance for error at the endpoint is tighter. Far and small together is the worst case.
In this trainer, target width is fixed at 48 pixels in normal mode, so the width term in Fitts's equation never changes. Distance does. Each circle appears in a random spot. So the gap from your cursor to the next target can be a short hop or a full diagonal across the panel. That is why some runs feel fast and others slow even when you played the same. Part of your average is just the luck of where the circles landed. Averaging over thirty targets matters for exactly this reason, since the random distances even out across a full run.
The difficulty modes are Fitts's law in action: Hard shrinks the circle to 32 pixels and Very hard to 24, and nothing else changes. Smaller width means a tighter endpoint tolerance, so the corrective phase of every movement costs more. Each mode keeps its own record — a 24-pixel average is a different measurement from a 48-pixel one, and comparing them would be exactly the mistake this page warns about.
Mouse settings that change your score
Your score depends on your gear and settings more than most people expect. Before you compare a result to anything, know what is under your hand.
DPI and sensitivity. DPI is how many counts your mouse sends per inch of physical movement; sensitivity is how far the system moves the cursor per count. Together they decide how much hand motion a long traverse costs you. Set them too high and the cursor flies past a 48-pixel circle, forcing extra corrections. Set them too low and the long movements eat your time. There is no single correct value, but there is a value your motor system has learned, and changing it throws away some of that learning.
Mouse acceleration. With acceleration on (Windows calls it "Enhance pointer precision"), the cursor travels farther when you move the mouse quickly. The mapping from hand to cursor stops being consistent, so the same physical flick lands in a different place depending on its speed. Fast pointing relies on your brain predicting where a movement will end, and acceleration makes that prediction harder.
Polling rate and refresh rate. A mouse that reports its position infrequently, or a monitor that redraws slowly, adds a small delay between your movement and what you see. These delays are small, but this test is scored in milliseconds.
Trackpad, mouse, or touchscreen. A trackpad gives you friction, a small surface, and a click that takes real force. A touchscreen removes the cursor entirely; you point with your finger, which is a different motor task. Comparing a trackpad score to a mouse score, or either one to a phone score, is not meaningful. Treat scores from different devices as different tests that happen to share a page.
How to actually get faster
Pick a sensitivity and leave it alone. Every change forces your motor system to relearn the mapping between hand and cursor, and that learned mapping is most of what this test measures. Consistency beats endless tweaking.
Use your arm for the big movement and your wrist and fingers for the final correction. Wrist-only aiming works for short hops but runs out of room on long diagonals. Dragging a planted wrist across the desk is slower than a loose swing from the elbow.
Keep your eyes on the target, not the cursor. Your eyes are built to guide your hand to the spot you are looking at. Watching your own cursor adds a tracking job you do not need. Snap your eyes to the new circle the instant it appears and let peripheral vision handle the rest.
Warm up before you judge yourself. The first run of a session is usually the slowest, and scores tend to improve over the next few as your movements calibrate. If you want a fair read on your ability, ignore the opening run.
Expect a plateau, because everyone hits one. Early gains come quickly, mostly from fixing strategy: gaze habits, grip, an unsuitable sensitivity. Once those are settled, you are chipping away at small margins of raw motor speed, and progress slows to a crawl. A flat average after weeks of practice is the normal endpoint, not a sign you are doing something wrong.