Skip to the solver

WCA rules · session kept on your device

Speedcubing Timer

Press space to start the 15-second inspection, hold space when you are ready, release to go, and hit anything to stop. Averages are trimmed the way the WCA trims them — the best and worst of every five are thrown out — and each solve gets a fresh random-state scramble automatically. Nothing leaves your browser.

Scramble

L2 D2 B2 R' F2 R B2 R' U2 L2 F2 D' B L F' D R B' U F2

0.00

Press space to start the 15-second inspection.

Solves

0

Best

--

Ao5

--

Middle three of the last five

Ao12

--

Middle ten of the last twelve

Session mean

--

Best Ao5

--

Best Ao12

--

Std deviation

--

Consistency, lower is better

What the numbers mean

StatisticHow it is worked outWhat it tells you
BestFastest single solve, penalties appliedYour ceiling on a good scramble
Ao5Middle three of the last fiveCurrent form. The number cubers actually quote
Ao12Middle ten of the last twelveForm with the luck mostly averaged out
Session meanArithmetic mean of every solveIncludes the disasters. Usually the honest one
Best Ao5The best such window across the whole sessionA personal best that is hard to fluke
Std deviationSpread of your times about their meanConsistency. Falls as recognition improves

Inspection, and why to practise with it

Regulation A3 gives a competitor 15 seconds with the scrambled cube before the timer starts. Going over is a +2; going past 17 seconds is a DNF. This timer applies both, measured from the moment inspection began to the moment the solve started.

It is tempting to switch inspection off for practice. Do not, if you intend to compete: planning a cross in fifteen seconds is a distinct skill from executing one, and people who practise without it consistently lose several seconds at their first competition discovering that.

Rough benchmarks

Ranges are indicative. Progress is not linear and most people sit at one plateau for a long time.
Ao5Where you probably areThe next thing to fix
Over 2 minutesLearning the beginner's methodFinish the method without a guide open
60–120 sBeginner's method, unaidedStop rotating the cube between every step
40–60 sFluent beginnerLearn F2L intuitively — the single biggest win
25–40 sF2L, beginner last layerTwo-look OLL and two-look PLL
17–25 sFull CFOP, still recognising slowlyLookahead: keep turning while you find the next pair
Under 17 sCompetitiveFull OLL, finger tricks, and a better cube

Timing a solve you cannot finish

Mark it DNF rather than deleting it — a deleted solve flatters your average, and the whole point of a trimmed mean is that it already handles the occasional disaster fairly. If you want to see how the solve should have gone, enter the cube on the solver and step through it.

Common questions

How is an Ao5 calculated?
Take the last five solves, throw away the fastest and the slowest, and average the middle three. That is WCA regulation 9f — it is a trimmed mean, not an arithmetic one, so a single lucky or disastrous solve cannot move it. An Ao12 does the same thing with the middle ten of twelve.
What happens to a DNF inside an average?
It counts as the slowest result, so one DNF in an Ao5 is simply trimmed away and the average still stands. Two DNFs make the average itself a DNF, because there is no longer a valid slowest result to discard.
What is the 15-second inspection for?
It is the time a competitor gets to look at the scrambled cube before starting. Regulation A3 allows 15 seconds; going past that is a +2 penalty and past 17 seconds is a DNF. This timer applies both automatically, so practising with inspection on gives you numbers that mean the same thing as competition numbers.
Where are my times stored?
In your browser's local storage, on your device. There is no account and no server — which also means clearing your browser data clears your session, and the times will not follow you to another device.
Why is my average so much worse than my best?
Because your best is a lucky scramble and your average is you. Speedcubers quote both for exactly that reason. If the gap is very wide, look at the standard deviation below the averages: a high number usually means recognition pauses rather than slow turning, and recognition is the cheaper thing to fix.

Keep going