Random Name Picker

A random name picker draws one entry from a list so that nobody has to choose. The draw uses crypto.getRandomValues, the browser's cryptographically secure random number source, rather than Math.random, and it runs entirely in the page, so the names are never sent anywhere.

Each row holds one entry: a person, a team, a topic, or a prize. Empty rows are ignored, rows can be added or removed at any point, and the list and pick history are saved in the browser between visits. A list pasted into any row, one name per line or a column copied from a spreadsheet, is split into one row per name, filling empty rows first. Leading and trailing spaces are trimmed, but spelling and capitalization are kept as typed, so "Ann" and "ann" count as two different entries.

With n names in the draw, each has a probability of 1/n. A 32-bit random integer is turned into a position by taking the remainder after dividing by n. Because 2^32 is not a multiple of most list sizes, a plain remainder would give the first few names a very slightly higher chance, so a draw that lands in the incomplete final block of values is discarded and redrawn, which makes every position exactly equally likely. By default the previous winner sits out the next draw, giving each of the other names a 1/(n - 1) chance. Settings holds the option to allow back-to-back repeats, the reveal animation, and the history display. A name entered on two rows is in the draw twice and has double the chance, the same as a second raffle ticket.

The reveal flashes random names from the whole list for just under three seconds before the winner appears. The flashes are decoration: the winner is drawn separately at the end, so a name shown during the animation says nothing about the result. Random also does not mean evenly spread over a few rounds. With 10 names and repeats allowed, the chance that 10 draws reach every name exactly once is 10!/10^10, about 0.04 percent, so a duty rotated fairly over a short run is better handled by deleting each winner's row after the draw.

With 5 names and back-to-back repeats allowed, each name has a 1/5 = 20 percent chance per draw. With the default setting the previous winner sits out, so each of the other four has 1/4 = 25 percent. For 5 names the cut-off is 2^32 - (2^32 mod 5) = 4,294,967,296 - 1 = 4,294,967,295, so only the single highest 32-bit value is ever redrawn, a 1 in 4,294,967,296 event.

Randomly selecting from a list removes the chooser's bias and settles who goes next, in contexts ranging from classroom activities to promotional giveaways.

• Classroom participation: A class list with one student per row gives each student the same chance to answer a question, present a project, or lead a group activity. Over a lesson, the history shows who has already had a turn.
• Giveaway and raffle draws: Contest entrants copied from a spreadsheet column become one row each, and the draw can run on camera or during a live stream. The reveal builds suspense for the audience, the timestamped history records every draw, and an entrant holding two entries can simply be listed twice.
• Team assignments: Drawing a name, deleting that row, and drawing again picks people without anyone being chosen twice. Alternating the drawn names between teams gives a random split for hackathons, sports leagues, or project groups.
• Meeting facilitators: Team member names kept in the saved list make choosing a facilitator, note-taker, or presenter a single draw at each meeting, and the default setting keeps the same person from being drawn twice in a row.
• Content scheduling: Topic ideas, blog post titles, or social media themes on separate rows turn the question of what to work on next into a draw, which breaks decision paralysis when every option is acceptable.
• Restaurant and meal selection: Favorite restaurants, cuisines, or recipes in the list settle the "where should we eat" debate. With the default setting, consecutive picks never return the same option twice in a row, which helps when building a weekly meal plan.
• Game night decisions: Board game or video game titles in the list resolve a group that cannot agree on what to play, and the history log shows what was chosen in earlier sessions.