The hand-raising problem has a specific shape: the same four students answer every question, a handful more answer when the material is easy, and the rest learn that being invisible is a viable strategy. Nobody in the room is doing anything wrong. The system just rewards being comfortable.
Replacing “who wants to answer?” with a draw changes what students are optimising for. It is a small change with a large effect, and most of the effect is in the first fortnight.
What Actually Changes
The barrier to answering is not the question. It is the risk of being picked badly. Take the picking out of the equation and participation rises across the board — not because the material became easier, but because being wrong stopped being evidence of something.
Quiet students get the floor. A volunteer system reliably surfaces the confident. Random selection finds people who did not intend to volunteer, which is the entire reason to do it.
You stop managing the optics. There is a real cost to appearing to favour particular students, and it lands on the students who notice. A draw is not a judgement you made, and that is easier to explain to a parent than a decision about a specific child.
Picking the Right Tool
Wheel of Names — the visual one, and the one students expect. Best where the announcement carries weight: cold-calling, reading rotations, “who presents next”. The anticipation is part of it.
Lucky Draw — a flat list, one winner, no theatre. Better for anything you repeat: a register round, a homework check, “who has not read yet this week”.
The choice is not important. Both draw from the same source and are equally fair. Pick on how much ceremony the moment deserves — speed for register, theatre for cold-calling.
Preparing the List
Type it before the lesson. Thirty names typed the night before takes five minutes. Typed live in front of thirty students it takes three, wastes the lesson, and gives them time to object to the list.
Use full names where there are duplicates. Two students called Alex means the draw cannot distinguish them, and a near-miss looks like rigging even when it is not.
Save it. Saved lists keeps up to five, works in your browser without an account, and syncs across devices if you sign in. One list per class per term is the useful unit.
Remove absentees before drawing. Otherwise a name comes up, you skip it, and the pool was not the pool.
Making It Land
Say that it is random. One sentence — “everyone has exactly the same chance” — removes most of the pre-emptive anxiety. Students who believe the mechanism cannot be influenced do not spend the question planning their escape.
Do not let it become a quiz. There is a line. Random selection works as a way of distributing attention across a lesson, not as a mechanism for putting individuals on the spot. If a student who cannot answer looks relieved rather than ashamed, you have it right.
Vary the stakes. Cold-calling for everything is exhausting and eventually produces a student who stops participating entirely.
Let students run it. The strongest version of all: hand the device to a student for the draw. They watch from the start, they can see the list was not edited mid-lesson, and they stop needing to trust that you rigged it. Thirty seconds, and it removes a suspicion you would otherwise spend the term managing.
Beyond Recitation
Group formation. Group Maker removes the “can I work with Sam” negotiation entirely. Use pot balance for anything graded, and seat the strongest students before you draw — a random split is not a balanced one.
Task assignment. In project work, certain roles are popular and others are not. Task Matcher pairs people with roles, and reports anything unassigned rather than dropping it silently.
Seating. Classroom Seater is better as a rotation than as a single event. Reassign on a fixed day and it stops feeling imposed.
Scoring. The Scoreboard turns a series of small tasks into something with a visible score, which people find considerably more engaging than the same points on a whiteboard.
Two Things Randomness Cannot Do
It cannot supply competence. Random selection for a task needing a particular aptitude will sometimes pick the wrong student, and no amount of fairness fixes a missing skill. Assign those deliberately; randomise everything around them.
It cannot equalise outcomes. A class of thirty will not have every student picked equally often over a term. Randomness is uniform, not balanced — over a long enough run the gaps close, and if the requirement is balance, a rotation is the right tool rather than a draw.
This one matters because a student will eventually notice it and ask. The honest answer is that the gaps close over a long enough run. Most students find that satisfying.
A Note on the Mechanism
Draws use the browser’s cryptographic random source rather than Math.random(), which is deterministic and can in principle be reconstructed to anticipate results. The fairness page covers it, including why uniformity across a range of 40 names needs rejection sampling rather than a scaled float.
The one-sentence version for a class: it uses your browser’s secure random generator, which cannot be predicted from previous results.
Start With One Thing
Use one draw in tomorrow’s lesson. One question, one student. Then notice whether the room relaxes.
It usually does within a fortnight.