Random Name Picker

P(a given name is drawn) = drawCount / nameCount and the draw is a seeded shuffle taking the first drawCount names

nameCount

How many names the list holds after it is split. The separators are a line break, a semicolon or a comma — a space is not one, so a name with a space in it stays a single entry and a numbered roster typed with spaces collapses into one name

drawCount

How many names to take out. It cannot exceed the size of the list, because the draw is without replacement: taking four names from a roster of three is not an unlikely outcome but an impossible request

seed

The starting state of the generator, up to 4,294,967,295. The same list, count and seed always produce the same names, which is what lets a published result be re-run and checked; change the seed and the whole draw changes

drawCount / nameCount

The chance that any one particular name comes out. It is not a promise about any single draw — a name with a 37.5% chance can miss ten times in a row — but over many independent draws the share of them a name appears in converges on this figure

picked

The names drawn, in the order the draw produced them, separated by commas. The order is part of the output rather than a presentation choice: the same seed reproduces the same order, so it can be used for a speaking order or a first-to-last list

Use it whenever a visible, defensible draw is needed and the fairness has to be checkable afterwards: picking a winner, deciding an order, assigning tasks, choosing which of several options to take. The seed is what separates this from an opaque random button — publish the seed along with the result and anyone can reproduce the identical draw, so a dispute about whether the picking was fair becomes a question of arithmetic rather than of trust. Two things to keep in mind when choosing the settings. Draw the whole list when the point is a complete ordering rather than a subset, since drawing every name is the same as a shuffle and the output becomes a running order. And check the list the page thinks it has before trusting the draw: the count reported beside the result is the number of entries actually parsed, which is the fastest way to catch a separator that did not split the way you expected.

1. Three names out of eight

   1. The list is split on the semicolons into eight entries
   2. The seeded draw takes three of them without replacement: 6, then 1, then 5
   3. Every name had the same chance going in: 3 / 8 = 0.375, so 37.5%
   4. Running the same draw again with seed 1 returns the same three names in the same order

   The first thing to look at here is the count of 8, not the three names. It is the page telling you how many entries it actually parsed, and it is the quickest check that the list was split the way you intended — a roster that silently became one entry still produces a confident-looking result. The 37.5% figure is the chance for each name individually, not a running total: the eight of them add up to 300%, because three names are being drawn and each draw is a separate claim on the list.

2. Names with spaces in them

   1. Each name sits on its own line, so the list holds four entries — not eight
   2. The draw takes two of the four: Mary Jane and Peter Parker
   3. Each name's chance is 2 / 4 = 50%
   4. Writing the same four names as John Smith; Mary Jane; Peter Parker; Bruce Wayne gives the identical result

   Four entries rather than eight is the whole point: a space does not separate names on this page, because most real names contain one. Semicolons or line breaks do, and either form gives the same draw from the same seed. If the count beside the result ever comes back larger than the number of people you pasted, the list has been split too finely — which happens when a comma is used as a separator inside a name written the other way round.

3. Drawing the whole roster

   1. Eight names drawn from a list of eight, without replacement
   2. Every name must come out exactly once, so the result is the whole list in a new order
   3. The chance for any given name is 8 / 8 = 100%, which is the honest reading of a full draw
   4. The output order is the draw order, so this doubles as a shuffle

   A chance of 100% looks like a mistake until the question is read carefully: it is the chance that a given name is drawn when every name is drawn, and it says nothing about where in the order that name lands. This is the setting to use when you want a shuffled running order rather than a subset — the same output serves both purposes, and the order is reproducible from the seed like everything else on the page.

4. Numbers typed with spaces, read as one name

   1. A space is not a separator on this page, so the list has exactly one entry
   2. Drawing one name from a list of one returns it unchanged
   3. The chance is 1 / 1 = 100%
   4. Rewriting the same roster with semicolons gives eight entries and the intended draw

   This is what the space rule looks like when it bites: eight numbers typed with spaces are one name, so the draw returns that single entry and reports a count of 1 and a chance of 100%. Nothing on the panel looks broken, which is why the count beside the result deserves a glance every time. The fix is a separator that is not a space — 1; 2; 3; 4; 5; 6; 7; 8, or one per line.

Three practical limits are worth knowing before the result is used for anything that matters. The list holds up to 200 names, and the draw can take at most as many as the list contains. A space never separates names, because real names contain them — the separators are a line break, a semicolon and a comma — so a comma inside a name written surname-first will split it in two, and the count beside the result is the only warning you get. The seed is a reproducibility device rather than a security one. The generator behind it is a small, fast, fully deterministic one of the kind that belongs in a page like this: it is entirely adequate for picking a standup order or a chore rota, and it is not suitable for a draw with real money at stake, where someone who knows the seed and the algorithm could work out the result in advance and where a verifiable public source of randomness is the right tool instead. Finally, a chance is not a schedule. Every name in the list has the same probability of being drawn, but a 1-in-8 chance does not mean a name comes up once every eight draws — it means that over a very long run of independent draws the share converges on one eighth, which is a statement about the long run and not about the next attempt.

Is the draw actually random, or does it favour some names?

Every name in the list has the same chance, and the page will only draw names without repeats, so no name can come out twice in a single result. What you can check is the count beside the result: it reports how many entries the page parsed, and if that number does not match the number of people you pasted, the list was split differently from what you intended and the draw is fair over the wrong set. That, rather than any hidden bias, is the usual reason a draw looks wrong — and it is why pasting a roster one name per line is the safest habit.

Why does the same seed produce the same names?

Because the generator is deterministic: the seed sets its starting state, and from there the sequence is fixed. That is a feature rather than a limitation, since it turns a claim about fairness into something anyone can verify — share the list, the number drawn and the seed, and the identical result comes back on any device, in any language, with no record kept. A different seed produces a completely different draw, so if you want a fresh result without changing the list, change that one number. The same determinism is also why this is not a suitable draw for stakes where someone could act on knowing the result in advance.

Does a one in eight chance mean the name comes up once every eight draws?

No, and this is the most common misreading of any probability. A one in eight chance means that over a very long run of independent draws the share of them that include that name settles near one eighth, but any particular run of draws can miss it many times in a row or include it twice in succession. Eight draws are far too few for the average to have any force. The number is still useful — it is what makes two names comparable, and what tells you that drawing four of eight gives every name a 50% chance — but it is a statement about the long run rather than a schedule.

Why has my list of numbers been read as a single name?

Because spaces do not separate names on this page. Most real names contain a space, so treating one as a separator would split John Smith in two, and the rule that keeps names intact is the rule that collapses 1 2 3 4 5 6 7 8 into one entry. The count beside the result shows it: it will read 1 rather than 8, and the chance will read 100%. The fix is a separator that is not a space — semicolons, commas or one entry per line — and any of the three gives the same draw from the same seed.

Can the same name be drawn twice?

Not in a single draw. Names are taken out of the list as they are chosen, so each result is a set of distinct names, which is what a raffle or a speaking order requires — announcing the same person twice would be a bug rather than a surprise. The one consequence to know is that the number drawn cannot exceed the size of the list: asking for four names from a roster of three describes a procedure that cannot be carried out, and the page says so instead of silently returning the three. Draw as many as the list holds and the result is the whole roster in a new order.

Why is there no table or list of example draws on this page?

Because the result depends entirely on the names you paste and the count you choose, and a table cannot see either. A printed set of example draws would be a result for some other list, and where it disagreed with the panel above it, the panel would be the one that is right — a page arguing with itself is worse than a page with nothing extra on it. The panel already is the answer, and the two numbers beside it are the only general facts worth tabulating here: how many entries the list actually has, and the chance each of them had of being drawn.

• Math.random() — JavaScript reference (a built-in generator with no seed parameter, and the explicit warning that it is not cryptographically secure) — MDN Web Docs, Mozilla
• 3.1 Terminology — Introductory Statistics 2e (the law of large numbers: the relative frequency of heads on a fair coin approaches one half as the number of tosses grows) — OpenStax, Rice University
• 1.3.6.1. What is a Probability Distribution — e-Handbook of Statistical Methods (the frequency reading of probability, which is what "a one in eight chance" is a statement about) — National Institute of Standards and Technology (NIST)