Writing Contests for Worldbuilders: How to Pick One, How to Win It, and What Winning Is Worth

How to tell a good contest from a bad one

Start from the assumption that you will lose. This is not pessimism, it is arithmetic — a contest with four hundred entries and five placements has a 98.75% loss rate, and no amount of talent makes that number personal. Once you accept it the question stops being can I win this and becomes what do I keep when I don't, which is a far better question and one most contests answer badly.

By that test, four things matter more than the prize:

• What the terms take. The clause to find is not about copyright — you keep your copyright almost everywhere — but about rights and licences. A contest taking first publication rights over every entry, sometimes for a year or longer, has made your piece unsellable for that whole period in exchange for nothing — because you did not win. A contest taking nothing has cost you only time.

• Whether there is a fee. Fees are not automatically predatory; magazines use them to pay readers and fund prizes. But they make the expected value of entering reliably negative, and paying to be read is a habit that scales badly across a career.

• Whether the criteria are published. A contest that tells you what it is looking for is a contest you can aim at. One that does not is asking you to guess at a moving target, and guessing is not a skill you can practise.

• Whether the work survives the result. Does entering leave you holding something — a finished piece, a world you can keep building, a public page people can find — or does the entry disappear into an inbox forever?

There is a second, subtler filter: how many entries the judges can plausibly read properly. A contest that receives three thousand entries and lists four judges is not lying to you, but the first pass on your entry is going to be very fast indeed. Smaller and newer contests are frequently a better bet than famous ones, for the boring reason that fewer people enter them.

Why constraint-based contests are a different game

Most open calls ask for your best short fiction and then ask judges to rank a haunted-house story against a satire about municipal parking. Those judges are doing their honest best, and the comparison is still close to meaningless. Taste fills the gap, which is why open-call results so often feel arbitrary — because in a real sense they are.

A constraint-based contest removes that problem by handing everybody the same starting material. Same prompt, same three elements, same opening line; the specifics vary, the principle does not. When every entry begins from identical inputs, the only variable left is what the writer did with them, and that turns out to be startlingly legible. Put fifty entries built from the same three elements side by side and the differences stop being a matter of taste and start being visible.

This changes what you should optimise for. In an open call the premise does a lot of the work — an arresting concept buys attention before the prose has to earn it. In a constrained contest everybody has your premise, so it buys you nothing at all. What is left is interpretation and execution: how far you got from the obvious reading, and whether the thing held together once you got there.

It also changes what a loss teaches you. Losing an open call tells you almost nothing, because you cannot separate this was weaker from this was not to their taste. Losing a constrained contest to an entry built from your exact elements is one of the most useful pieces of feedback available to a writer, and it costs nothing.

What judging looks like from the other side

It helps enormously to picture the reader rather than the prize. Somewhere there is a person with a large pile and a finite evening, and your entry is somewhere in it.

The first pass is fast, and it is mostly elimination. Not because readers are lazy but because a majority of entries in any open contest disqualify themselves — they ignore the brief, they open on a paragraph of weather, they are a fragment of a longer thing that assumes context nobody has. Surviving this pass is a far lower bar than winning, and an enormous number of entries fail it for reasons that have nothing to do with the quality of the writing.

The second pass is where coherence beats flair. Entries that got through are read properly, and what separates them at this stage is usually whether the piece knows what it is. A modest idea executed cleanly beats an ambitious one that collapses in the third act almost every time, because the collapse is the part the reader remembers.

The final pass is a conversation, and it is about defensibility. Somebody has to be able to say why this one. The entries that win are frequently the ones easiest to argue for, which is not the same as the ones people liked most — and it is why a clear, strange, complete piece often beats a beautiful fragment.

The practical consequence: your opening carries far more weight than its word count suggests, your ending carries more than you think, and the middle is where you can afford to be interesting. Most writers allocate their effort in precisely the opposite order.

What separates the entries that place

Across constrained contests specifically, a few patterns hold well enough to plan around.

• Answer the brief completely, then depart from it. Judges notice when an element has been included grudgingly. If a contest gives you three things to use, the entries that place are the ones where removing any one of them collapses the story — not the ones where the third turns up in a sentence near the end.

• Go past your first idea. Your first response to a prompt is, statistically, also the first response of a hundred other entrants. The obvious reading is not bad; it is crowded, and the reader arrives at yours already tired of it.

• Be strange in one direction only. Entries that place tend to make a single unexpected choice and then follow it with total discipline. Entries that make six unexpected choices read as noise, because there is no stable ground left to measure the strangeness against.

• Finish it. An enormous share of entries are openings — atmospheric, promising, unresolved. A complete small thing outperforms an incomplete large one so consistently that it is close to a rule.

• Make it legible at a glance. A title that means something, an opening that establishes who and where inside two sentences, and a shape the reader can feel. This is not dumbing down; it is respecting that yours is not the only entry being read tonight.

What a placement is actually worth

Honestly? Less than the announcement makes it look, and considerably more than cynics claim.

The prize money is rarely the point, and treating it as wages produces a miserable hourly rate. What a placement genuinely buys is standing — a line that means something to people who do not know you, a reason for an editor or a collaborator to open your next email, and a credential that keeps working long after the money is spent. That is real, and it is not nothing.

It also buys attention on a specific piece of work, which is the part writers most consistently waste. A placement generates a short window in which people are actively looking for what you made. If what they find is a PDF on a contest site with no way to follow you, that window closes with nothing behind it. If what they find is a world you are still building, with somewhere to follow along, the placement compounds instead of expiring.

What it does not buy is a career on its own, and contests implying otherwise are selling something. Treat any single result as one data point. The writers for whom contests work are the ones entering several a year, keeping everything they write, and building an audience that survives any individual result.

How SANDBOX works

SANDBOX is CanonBoard's own constraint-based story pitch contest, and it is built around the arguments above rather than in spite of them.

Every entrant in a round gets the same element sets — each one a location, an object and a moment. You pick a set, pitch a story that uses all three, and what comes back never looks the same twice. The sets are graded, so there is a low-floor way in and a genuinely difficult tier for anyone who wants the harder problem: in the difficult sets the three elements resist each other, and solving that collision is the whole test.

The parts that matter under the tests earlier in this post:

• Entering is free. No entry fee, no subscription requirement, no purchase of any kind.

• Your world stays yours. Entering transfers nothing — no ownership, no exclusivity, no publishing rights. What you grant is permission to feature the entry while promoting the contest, and that is the whole of it.

• You keep the work whether you win or lose. An entry is a public CanonBoard project on your own account, not a file in somebody's inbox. It stays live, followable and buildable after the round ends — which is the what do I keep when I lose test, answered.

• People judge it. Nothing is scored, ranked, screened or summarised by AI at any stage, and there is no public vote. The criteria are published in full.

• You can withdraw at any time, from project settings, and the project itself is untouched when you do.

How much you build is up to you. A logline is a complete entry and will be read. Drafting the key scene, or putting characters, locations and world rules on the canvas, is how an entry stops being a premise — and a world with real structure behind it reads completely differently in a pile of hundreds.

The current round's element sets, deadline, prizes and full rules are on the SANDBOX page, and the contest terms set out the rights position in plain English. If you have never entered a contest before, a constrained one is the right place to start: the brief does half of the hardest work for you.

Frequently asked questions

Are writing contests worth entering?

Some are. The test is not the prize size, it is what you keep when you lose — which is almost always what happens. A contest that leaves you with a finished piece of work you own, on a platform where people can find it, has already paid you something. A contest that takes a reading fee, holds first publication rights over everything submitted, and returns a form rejection eight months later has taken more than it gave. Read the terms before the prize.

Should I ever pay an entry fee?

Occasionally, and only with your eyes open. Legitimate contests run by literary magazines and nonprofits use fees to fund the prize and pay their readers, and some of those are worth it. But a fee changes the maths: you are buying a lottery ticket whose expected value is negative, and paying for the privilege of being read is a bad habit to build early in a career. Free contests with real prizes exist. Start there.

What is a constraint-based writing contest?

One where every entrant is handed the same starting material — the same prompt, the same three elements, the same opening line — and judged on what they do with it. It removes the biggest source of noise in ordinary contests, which is that judges are comparing a space opera to a divorce story and pretending that is a fair comparison. When the inputs are identical, the only variable left is the writer.

Do I lose rights to my story if I enter a contest?

It depends entirely on the terms, and this is the single clause worth reading twice. Many contests take first publication rights over every entry, not just the winners, sometimes for a year or longer — which can quietly make a piece unsellable elsewhere. Others take nothing at all. The words to search the terms for are not just 'copyright' but 'rights', 'licence', and 'exclusive'. If a contest cannot tell you plainly what it takes, that is your answer.

How do judges actually choose a winner?

In stages, and the first stage is brutal. Most entries are eliminated in under a minute by a reader working through a large pile, which means the opening is doing more work than anything else you write. Entries that survive the first pass get read properly, and at that point the decision is usually about coherence rather than flair — whether the thing holds together and knows what it is. Judging is subjective, but it is not random, and the criteria are usually published. Read them.