Fast solvers do not turn much faster than you — they stop less. Two skills make that possible: hands that execute without regripping, and eyes that are already on the next piece.
Finger tricks
One goal: never regrip mid-algorithm. Home grip — thumbs on the front, index fingers resting on top. Every common turn has a finger that reaches it from there:
- U turns — index flicks. Left index pushes U, right index pushes U’. For U2, roll index then middle in one motion.
- R and L — from the wrist. The whole hand rotates the face; the grip never changes.
- F turns — thumb pushes. Reach the thumb down to the front layer and push, rather than rotating the cube into a new hold.
Practice on the trigger you know best. R U R' U' should collapse into one gesture: wrist, flick, wrist, flick — four turns, zero regrips.
Lookahead
While your hands run an algorithm they know, your eyes have nothing to do there. Send them ahead: during each F2L insertion, find the next pair; during the last pair, read the top for OLL. An algorithm only frees your eyes once it runs without thought — if you still watch your hands through T-Perm, drill it blind first.
The drill is deliberately slow solving. Set a metronome pace — about one turn per second — and solve without ever stopping. If you must pause to find the next piece, you were turning faster than your eyes. Slow the hands until the solve is one continuous stream, then let the pace creep up.
Honest metrics
| Average | What it actually takes |
|---|---|
| 60s | 2-Look OLL/PLL from memory, beginner-speed F2L, pauses everywhere. No finger tricks required. |
| 40s | Intuitive F2L and clean finger tricks. You still stop before every pair. Full PLL not needed. |
| 25s | Cross planned in inspection; lookahead carries F2L with only brief stops. Full PLL helps here. |
Each level is reached by removing pauses, not by adding hand speed — which is exactly what the slow-solving drill trains.