Statistics

Five random Chinese characters make no word more often than not

In a Chinese word puzzle the pieces are characters, and a random handful of five makes no word at all in a little over half of draws from an open dictionary. Where characters do rearrange, the result can be a new word, and some idioms turn out to be shuffles of each other.

4 Nov 2026 · 6 min read

Statistics

Student's t-test and Welch's go astray in different places

Student's t-test pools two groups' variances and Welch's keeps them apart. With groups of 10 and 40 and a fourfold spread, the pooled test's false-positive rate is 29.4% or 0.08% depending on which group is the noisy one, while Welch's stays near 5%.

3 Nov 2026 · 6 min read

Statistics

Roman numerals on Roman stones fail today's strict standard form

Roman numerals today follow a strict standard form that allows six subtractive pairs, no letter more than three times, and nothing above 3,999. Numerals carved on Roman legionary stones break that form, and software writes numbers past 3,999 in several different ways.

19 Oct 2026 · 6 min read

Statistics

How reading-time estimates turn words and Chinese characters into minutes

A reading-time label divides a text's length by an assumed reading speed, and the speeds in use for English run from 200 to 300 words a minute. For Chinese the bigger difference is what a tool counts: characters, segmented words or text split at spaces.

18 Oct 2026 · 6 min read

Statistics

Chinese reign era names were reused by rival courts, and most Ming and Qing emperors began theirs at New Year

Chinese emperors named their years by era names, and a date was written as the era plus a year count. Catalogues disagree on how many names there were, many were used by several courts, and the first year of an era was often not the year an emperor came to the throne.

16 Oct 2026 · 6 min read

Statistics

Riegel's race-time formula comes within about 2% of world records and predicts recreational marathons too fast

Riegel's formula scales a known race time to another distance using an exponent, usually 1.06. Tested against world records it misses by about 2% at most, but for recreational runners in a published 2014 survey it predicted marathons a median 5.5% too fast.

15 Oct 2026 · 5 min read

Statistics

Descriptive and algebraic chess notation, and the rules FIDE and PGN files disagree on

Old chess books write moves in descriptive notation, naming each square from the mover's side, while FIDE's Laws recognise only algebraic for its own events. Game files in PGN format use an algebraic of their own, which differs from FIDE's on castling, promotion and which symbols are required.

15 Oct 2026 · 6 min read

Statistics

Hundreds of subject combinations can add up to the same PSLE Score

Singapore's PSLE Score adds four Achievement Levels into a total from 4 to 32, where lower is better, and MOE sorts those totals into three Posting Groups. Counted across every Standard-level grade combination, hundreds of different subject profiles share each score in the middle of the range.

15 Oct 2026 · 6 min read

Statistics

Singapore's golf tees turn one handicap index into very different stroke counts

A Course Handicap rescales a golfer's Handicap Index by the Slope Rating of the tee being played. Across the 88 tees the Singapore Golf Association rates over 18 holes, that Slope part alone turns an 18.0 Index into 17.4 to 24.7 strokes.

15 Oct 2026 · 6 min read

Statistics

Net run rate tables from the ICC add up only when overs are counted in balls

Net run rate is a cricket side's runs per over minus the runs per over scored against it, across every match with a result. Counting overs in balls, we reproduced all 13 figures the ICC published for three groups of the 2026 Men's T20 World Cup.

14 Oct 2026 · 6 min read