Tool
Cold Call Student Picker
Draws through the whole roster before anyone repeats — and remembers between lessons.
Cold Call Student Picker
The cycle is stored in this browser only. It does not follow you to another device, and clearing site data or using a private window resets it — worth knowing before you rely on it in September.
Paste the register once. Every draw takes a name out of the pool, so the whole class is asked before anyone is asked twice — and the cycle survives closing the tab, so it carries across lessons rather than resetting every period.
This is the only tool on the site that breaks the equal-chance rule on purpose, and it is worth being direct about it: the first draw of a thirty-name roster is 1-in-30, and the last draw of the cycle is a certainty.
A teacher does not want equal odds. A teacher wants everyone asked, which is a coverage guarantee, and coverage and equal odds are different properties that cannot both be maximised.
How the Cold Call Student Picker works — and why it's fair
The roster is held as a pool of names not yet drawn in the current cycle. Each draw takes an unbiased uniform pick from that pool and removes it. When the pool empties, it refills from the full roster and the cycle counter increments — so a term is a sequence of complete passes rather than an unbounded stream of independent draws.
The consequence is that the odds rise as the cycle proceeds, and the tool prints the current odds rather than leaving them implied. That rise is the mechanism doing its job: the twenty-ninth draw of a thirty-name cycle is 1-in-2 precisely because twenty-eight names have already been used, which is the guarantee you wanted.
Compare this against the plain name picker, which draws independently every time. Over thirty independent draws from a thirty-name roster, the expected number of students never picked is thirty times (29/30)^30, which is about 11 — more than a third of the class untouched after a month, while somebody has been picked four times. That gap is the entire reason this tool exists separately.
The cycle is stored in this browser under a key tied to the roster contents. Changing the roster — adding a student, correcting a spelling — starts a fresh cycle, because a half-finished cycle over a different list is not meaningfully half-finished. That is a deliberate choice and it does mean a mid-term correction resets your progress.
Duplicated names are treated as two entries, so a name entered twice genuinely gets two turns per cycle. That is occasionally what a teacher wants for a student who needs more practice, and it is more often a paste error, so the tool does not silently deduplicate.
The memory is local and fragile, and the page says so. It does not follow you to another device, it does not survive a private window, and clearing site data wipes it. Any teacher relying on it across a term should know that before September rather than in November.
When the Cold Call Student Picker is fair — and when it is not
What it does guarantee
- Every student is asked exactly once per cycle — a guarantee an independent draw cannot make.
- Within the remaining pool, every unasked student is exactly as likely as every other.
- Progress through the cycle and the current odds are both displayed rather than implied.
What it does not
- It is not an equal-chance draw. The odds change as the cycle empties, and the final draw is a certainty.
- Order within a cycle is unpredictable, so a student can be asked at the end of one cycle and the start of the next — twice in two lessons.
- The memory is in this browser only. It does not sync, does not survive a private window, and is lost if site data is cleared.
Two worked examples
A 30-name roster, first lesson
- Draw 1 is 1-in-30 for each student. Draw 10 is 1-in-21 from the remaining pool.
- Draw 30 is a certainty — one name is left, and it is asked.
- After 30 draws, every student has been asked exactly once.
The same roster with an independent picker
- 30 independent draws from 30 names leaves an expected 30 × (29/30)³⁰ ≈ 11 students never picked.
- Meanwhile somebody will have been picked three or four times.
- That is the difference between equal odds and guaranteed coverage.
Why coverage beats equal odds in a classroom
The everyday problem is not fairness in a single lesson — it is that the same six students answer every question in every lesson, and the teacher's own attention is the mechanism. A picker with memory removes the teacher from the selection entirely and guarantees the quiet two-thirds are reached.
It also disarms the most common objection to cold-calling from students, which is that it is targeted. A visible pool that visibly shrinks is a much stronger answer than an assurance, and the counter on screen is doing that work.
University seminars and training groups use it identically, and often with a smaller group, where the cycle completes in a single session and the guarantee is immediately visible.
For a genuinely independent draw where repeats are acceptable, the plain name picker is the right tool. For rotating seats rather than questions, the random seat rotation uses a derangement; and the classroom use-case page covers what to do when a student freezes.
Frequently asked questions
Why do the odds change during a cycle?
Because drawn names leave the pool. That shrinking pool is exactly what guarantees everyone is asked before anyone repeats.
Can a student be asked twice in two lessons?
Yes — at the end of one cycle and the start of the next. The guarantee is per cycle, not across cycle boundaries.
Does the cycle survive closing the browser?
Usually. It is stored in this browser under a key tied to your roster, so a normal tab close and reopen continues where you left off.
What happens if I correct a name mid-term?
The cycle resets, because a partially completed cycle over a different roster is not meaningfully partial. Correct the roster early if you can.
Does it sync between my laptop and the classroom computer?
No. The memory is local to one browser on one device and does not follow you anywhere.
What if I paste a name twice?
It gets two turns per cycle. That is sometimes intentional, so the tool does not deduplicate silently.
How is this different from the name picker?
The name picker draws independently every time, which leaves about a third of a 30-name class unpicked after 30 draws. This one guarantees coverage instead.
Can I see who has already been asked?
Yes, the list of names used this cycle is shown below the result, most recent first.
Does it work for something other than students?
Yes — any list where coverage matters more than independence. Rotating presenters, reviewers, or duty assignments all fit.