What Is Speedcubing and How Do You Get Started?

Speedcubing is solving twisty puzzles for time rather than just for completion. You start by learning a beginner layer-by-layer method on a decent speed cube, then move to an advanced method like CFOP, Roux, or ZZ once you can solve without looking at instructions. The gear that actually matters at the start is a cube with good corner cutting, some lube, a timer, and a mat; everything else is optional until you know you'll stick with it.

Speedcubing vs. casual solving

A casual solver is done when the cube is solved. A speedcuber is done when the timer stops, and the number on the timer is the point. That single shift changes everything about how you practice:

• You measure solves and track averages instead of just finishing.
• You learn algorithms as muscle memory, not as steps you read off a page.
• You care about fingertricks, because how you turn a face matters as much as which face you turn.
• You compare your times against your own past times, not against whether the puzzle got solved.

The main events and puzzle types

Speedcubing is not one event. The Speed Cube Database covers algorithms for 2x2 through 6x6, plus Square-1, Pyraminx, and Megaminx, which gives a sense of how wide the competitive range is.

You do not need to pick an event on day one. Most people start on 3x3 because that's where the method resources, algorithm sets, and community knowledge are deepest, then branch out.

Method progression: beginner to advanced

The usual path runs from a beginner layer-by-layer method to one of the advanced methods. The beginner method breaks the solve into many small, easy steps with few algorithms. Advanced methods cut the step count and raise the algorithm count, which is why they're faster once the algorithms are automatic.

• Beginner layer-by-layer. Solve the first layer, then the second layer, then orient and permute the last layer. Low algorithm count, high step count. This is where you should start.
• CFOP. Cross, F2L (first two layers), OLL (orient last layer), PLL (permute last layer). The most widely documented advanced method, and the one most algorithm databases are organized around.
• Roux. Block building on the left and right, then last-layer and middle-slice work. Fewer rotations, more emphasis on block-building intuition.
• ZZ. Starts with an edge-orientation step, which sets up a last layer with fewer cases and lets you solve without rotating the cube as much.

The right time to switch is when your beginner-method solves feel automatic and your times have plateaued, not before. Switching early means learning advanced algorithms while still thinking about basic turning.

Gear that actually matters

• A speed cube. The defining feature is corner cutting: the cube lets you turn a face even when a layer isn't perfectly aligned. A cube that locks up on slight misalignment will cap your improvement no matter how well you know your algorithms.
• Lube. Changes the feel and speed of the cube. Different lubes give different results, so treat this as tuning rather than a requirement.
• A timer. Needed to measure solves and track averages. Without recorded times you have no feedback loop.
• A mat. Protects the cube and the surface, and gives you a consistent place to solve.

Everything beyond this list is preference. Do not buy gear to solve a practice problem.

Using algorithms and algsheets

An algorithm (alg) is a fixed sequence of moves that produces a specific result, written in standard notation. An algsheet is a collection of algorithms organized by case, usually one sheet per step or per puzzle. The Speed Cube Database is a free online speedcubing algorithm and reconstruction database covering every algorithm for 2x2–6x6, Square-1, Pyraminx, and Megaminx, and it also hosts video content.

A workable way to use them:

1. Identify the case you keep getting stuck on.
2. Look up the alg for that case on an algsheet or database.
3. Learn the move sequence slowly, then drill it until it's automatic.
4. Practice the fingertricks for that alg so the turns flow instead of stopping between moves.
5. Add it to your solves only after it's faster than what you were doing before.

Learning algs you never encounter in real solves is wasted effort. Learn the cases that show up in your own solves.

A simple practice routine

• Timed solves. Do sets of solves with the timer running. This is your raw data.
• Track averages. A single fast solve means little; an average over many solves shows whether you're actually improving. Watch the trend, not one number.
• Learn fingertricks. Practice turning technique on their own, separate from full solves, so the movements become automatic.
• Drill weak cases. When one step keeps costing you time, isolate it and repeat it rather than doing more full solves.
• Review reconstructions. A reconstruction is a step-by-step breakdown of a solve. Comparing your solve to a faster one shows where you lost time.

Getting started, in order

1. Learn the beginner layer-by-layer method until you can solve a 3x3 without instructions.
2. Get a cube with good corner cutting, a timer, and a mat.
3. Start timing your solves and tracking averages.
4. Learn fingertricks for the moves you already use.
5. Pick one advanced method (CFOP is the most documented) and learn it in stages.
6. Use an algsheet or algorithm database to learn cases as you meet them.
7. Branch into other events once 3x3 feels comfortable.