Method
How timed fact tests work
Counts, limits, fact ranges, and what the number at the top does not tell you.
A weekly ritual with a clock on it
A timed fact test is a page of single-step arithmetic answered against a stopwatch. In most American elementary classrooms it runs once a week, often on a Friday, often under a name like Mad Minute or minute math. The sheet is handed out face down, a timer starts, and children answer as many problems as they can before it stops.
The mechanics matter more than they look. Because the limit is fixed, the score is a count of problems finished correctly rather than a percentage — a child who answers 18 of 20 in a minute and a child who answers 18 of 100 in a minute have the same numerator and nothing else in common. The denominator on a timed sheet is time, not problems. That is the single most misread thing about the format, and it is why the sheet on every page here shows the elapsed time against the limit as its own fraction.
What twenty, fifty or a hundred problems change
Two of the four presets change the sheet itself; two change the clock. The count decides how much of the fact set gets sampled, and the limit decides how much of that is reachable. The pair together sets a required rate, and that rate is what a child either can or cannot sustain.
| Problems | Limit | Problems per minute | Seconds per problem |
|---|---|---|---|
| 20 | 1 min | 20.0 | 3.0 |
| 20 | 2 min | 10.0 | 6.0 |
| 20 | 5 min | 4.0 | 15.0 |
| 50 | 1 min | 50.0 | 1.2 |
| 50 | 2 min | 25.0 | 2.4 |
| 50 | 5 min | 10.0 | 6.0 |
| 100 | 1 min | 100.0 | 0.6 |
| 100 | 2 min | 50.0 | 1.2 |
| 100 | 5 min | 20.0 | 3.0 |
Anything above about 30 problems a minute assumes recall with no working step at all. Below about 10 a minute, the clock stops being the constraint and the sheet is measuring something else.
Twenty problems is a sample; a hundred is a census. With thirteen facts in a single range, a twenty-problem sheet draws each one once or twice, so one genuinely missing fact may not appear. A hundred-problem sheet draws each about eight times, and a missing fact shows up as a cluster in the missed list rather than as a single wrong answer that could have been a slip.
The fact range is the fourth preset and the one that changes what the sheet can tell you. A single range holds one operand steady, so a child can lean on the pattern in that table; mixed and 0–12 take the pattern away and ask whether the fact is there without it.
Two sentences about knowing facts from memory
The Common Core standards mention memorized facts in exactly two places, and both are short. They say nothing about speed, nothing about a time limit, and nothing about weekly testing.
2.OA.B.2 — page 19 “By end of Grade 2, know from memory all sums of two one-digit numbers.”
3.OA.C.7 — page 23 “By the end of Grade 3, know from memory all products of two one-digit numbers.”
These sheets are not aligned to a grade or to a standard and we do not claim that they are. The quotations are here so you can read the actual wording rather than a summary of it. Note what the wording covers: one-digit numbers. The 10s, 11s and 12s that appear on most classroom sheets, and on these, are outside both sentences — a local convention, not a requirement.
Why NCTM says to avoid timed tests
The largest professional body of mathematics teachers in the United States takes a direct position on this format, and it is not a qualified one.
NCTM, Procedural Fluency in Mathematics “Timed tests do not assess fluency and can negatively affect students, and thus should be avoided.”
youcubed, Fluency Without Fear “For about one third of students the onset of timed testing is the beginning of math anxiety.”
Both claims are worth reading before the score on any sheet is. The first is about what the format measures: recall speed on isolated facts is not the same thing as fluency, which includes choosing a method and knowing when a method fits. The second is about cost — the format is not neutral for every child who sits it.
Many classrooms run one every week regardless, and that is the situation this site is built for. If a weekly timed sheet is going to happen, the sheet may as well be newly generated, correctly keyed, and scoped to the range a child is actually working on. That is the whole argument for this tool, and it is a narrower argument than "timed tests are good".
Two pages out of one print command
Pressing Print worksheet + answer key sends two pages. Page one is the problem sheet with a header for name, date and score. Page two is the same grid in the same order with the answers filled in. They go out together so the key cannot end up separated from the sheet it belongs to.
The layout is set for portrait paper with a 12 mm margin and fits both US Letter and A4 without anything being cut off, including at 100 problems. To get a file rather than paper, choose Save as PDF in the browser print dialog — we do not generate a PDF ourselves, which is also why the sheet can be regenerated as many times as you like before you print it.
Everything that is not the sheet is dropped from the printout: the header, the presets, the buttons, the result panel, the article text and the advertising.
Nothing, and there is nowhere for it to go
No score, answer or name leaves the browser. There are no accounts, no sign-in, no saved history, and no database behind this site. The problems are generated in the page, the scoring happens in the page, and reloading discards all of it.
That is a design decision rather than a policy statement. A site that stored a child's results would need an identity to attach them to, and asking children for identifying information is exactly what we do not want to do. The cost is that there is no progress chart; the benefit is that there is nothing to leak. The privacy page sets out what the advertising on these pages does involve, which is the one thing that is not nothing.
Five that come up
- How many math facts per minute is normal?
- Classroom sheets commonly assume somewhere between 20 and 40 problems a minute for a child who has the facts memorized, which is one and a half to three seconds per problem. There is no single published figure, and the number a school uses is set locally rather than by any standard.
- How long should a timed test be?
- One minute is the usual length because it is short enough to run weekly without eating a lesson. Five minutes with 100 problems is the other common form, used when the point is to cover a whole fact set rather than to sample it.
- Is 100 problems in five minutes the same as 20 in one minute?
- The rate is the same — 20 problems a minute — but the sheets behave differently. Twenty problems sample the fact set, so a single weak fact may not appear at all; a hundred problems draw every fact several times, so a genuine gap shows up repeatedly instead of once.
- Does a timed test measure fluency?
- NCTM says it does not: "Timed tests do not assess fluency and can negatively affect students, and thus should be avoided." A timed sheet measures recall speed on isolated facts, which is one narrow part of what fluency is usually taken to mean.
- Do these sheets line up with a grade level or a standard?
- No, and we do not claim they do. The standards are quoted on this page so you can read what they actually say; choosing which range belongs in front of a particular child is a judgement the sheet cannot make.
What the standards and the critics actually say
Common Core State Standards for Mathematics (PDF), page 23 — 3.OA.C.7 “By the end of Grade 3, know from memory all products of two one-digit numbers.” thecorestandards.org — CCSSI_Math Standards.pdf
NCTM — Procedural Fluency in Mathematics (position statement) “Timed tests do not assess fluency and can negatively affect students, and thus should be avoided.” nctm.org — Procedural Fluency in Mathematics
youcubed — Fluency Without Fear “For about one third of students the onset of timed testing is the beginning of math anxiety.” youcubed.org — Fluency Without Fear