Solve Lab
Solve·Lab

Analyse your solves.

Drop a timer export in. Everything is computed in your browser — nothing is uploaded, nothing loads until you say so.

Drag & drop / click to choose
.csv  ·  .txt  ·  paste with ⌘V
Reads
CubeTime · csTimer · Twisty Timer · ChaoTimer · plain lists — comma, semicolon or tab separated. Columns are sniffed: time (12.34, 1:23.45, DNF(12.3), 14.2+, millis), date, scramble, comment, penalty.
Ao100, first to latest
—

Every practice day, coloured by its mean

Weekly mean, then one cell per practice day
Range
From
To
Session
Drop above
Showing

Solve timeline

Average of N 100
Trim rule
Trim each end 5%
X axis
Scale
Trend
Filter
Process noise
Window
Options

Activity

Personal bests — single

Personal bests — averages

Where it is going

least squares on every solve in the filter, projected forward

Daily mean & best

one point per practice day

Monthly summary

mean · best · solve count

Sub-X rate

Options
Threshold
Window 100

Consistency

rolling spread of the same window
Options
Window 100
Measure

Period comparison

first block vs last block of the filtered set
Options
Block size 250

Record book

best average of every N, and when it happened

Barriers broken

first solve under each mark

Every single PB

Distribution

Options
Bin width

Percentiles

cumulative — read off any sub-X rate

Percentile table

and the shape of the tail

Solve-to-solve

is a good solve followed by a good solve? lag-1 correlation
Options
New session after a gap of 60 min
Sessions

Within a session

mean time by position in the session — warm-up and fatigue

Practice calendar

Hour of day

mean time · solve count

Day of week

mean time · solve count

Streaks & volume

Scramble effects

mostly noise — that is the point. Anything far from the line needs a lot of solves behind it.

By first move

mean ± standard error

By scramble length

moves in the scramble

Move frequency

how often each face turn appears, and the mean time when it does
Search
Show