Saturday 24 October 2026 Rebuilt nightly from the WCA's official results · export of 24 October 2026 · About
The Middle Three
People → Brazil → Miguel Yoshimura da Costa

People · Brazil

Miguel Yoshimura da Costa

Skewb · top 2.0% of 74,164 all-time · competing since November 2024 · 2024COST21

Miguel Yoshimura da Costa has been competing for a year. His best event is the Skewb: a personal-best single of 2.24, set at MS ao Cubo 2025, puts him in the top 2.0% of the 74,164 people who have ever been ranked in the event, and 26th in Brazil. Until February 2014, no official Skewb solve in the world had been that fast. Across 5 competitions since November 2024, he has put 253 official solves on the record across nine events. Solve to solve, his 3×3 times vary about 12% around a typical one, steadier than 71% of the 36,406 competitors with ten or more counted rounds. His 3×3 best has come down from 10.60 in November 2024 to 8.76, a 17% improvement. Miguel has entered five competitions, more competitions than 83% of everyone who has ever competed.

Skewb
best event — top 2.0% of 74,164 all-time
2.24
signature single · MS ao Cubo 2025
253
official solves on the record
12%
solve-to-solve spread · steadier than 71% of the field

Standing, event by event

vs everyone ever ranked
EventPB singlePB averageAll-time standingBrazil
Skewb2.247.42
top 2.0%26 / 1,493
2×21.713.12
top 3.2%112 / 5,285
3×38.7610.07
top 4.3%199 / 9,942
Pyraminx3.367.24
top 4.9%113 / 3,311
3×3 one-handed14.6017.44
top 7.4%98 / 2,045
4×441.6147.12
top 14.1%172 / 1,884
Megaminx1:36.001:47.99
top 32.1%267 / 1,082
5×52:49.692:54.81
top 74.5%553 / 855
Clock28.2642.25
top 82.2%582 / 741

Percentile against all competitors ever ranked in each event, a fairer measure than raw rank, because the pools differ by a factor of ten.

The line so far

Campeonato Brasileiro de Cubo Mágico 2024: 10.60Presidente Prudente Open 2025: 10.34MS ao Cubo 2025: 8.96UFMS Space Rubik 2026: 9.73Presidente Prudente Open 2026: 8.768.7610.6015 Nov 20245 Sep 2026

Best single per competition, 3×3. Faster is higher — the axis is the right way up.

The story so far

Competitions

Reading this page: a single is one solve's time · an average is the middle three of five solves, best and worst discarded · a PB is a personal best · a DNF (did not finish) is an attempt with no valid result: a popped piece, an unsolved face, an over-limit memorisation · top X% compares against everyone ever ranked in that event, which is fairer than raw rank because event pools differ tenfold.

The circuit

The people whose scorecards keep turning up next to his.

Next targets

  • A sub-2 Skewb single — 1,031 people have one; it is 0.24 away.
  • A sub-1 2×2 single — 1,050 people have one; it is 0.71 away.

Targets move as he improves.