SQL is a shift-heavy language
Most languages are lowercase with punctuation on top. SQL, by convention, is uppercase keywords with lowercase identifiers, which means your left pinky is holding shift for roughly a third of every query.
That constant toggling is why experienced developers still mistype column names in a query they have written a hundred times. The problem is not knowledge, it is the cost of switching case every few characters.
The expensive characters
- Keywords in capitals — SELECT, FROM, WHERE and GROUP BY keep your pinky in constant motion.
- Asterisks in SELECT statements — shift plus eight, on a finger that rarely does anything else.
- Single quotes around string literals — a pinky reach that has to happen twice per value.
- Underscores in column names — the most common character in a real schema and the least practised.
- Commas and semicolons — cheap individually, expensive when one is missing and the query runs for a minute before failing.
Short text, high stakes
A query is short compared with a paragraph of prose, so SQL scores tend to sit lower than English but not as low as JavaScript: there are fewer brackets, but more case switching. Between 35 and 55 WPM is a normal working range.
Practise the queries you write in real life. Joins, aggregations and date filters are where the punctuation actually concentrates, and they are also where a typo costs you the most.