What is a prediction-market signal?

A signal is the simplest, most useful idea in this whole site. It's a real-world number that tells you when to act — the condition that has to be true before a bot places a bet. "Only buy when the yield curve is inverted." "Only trade crypto on a day the whole market is panicking." "Only act when the forecast high has already cleared the bracket." Each of those is a signal: a checkable, real-data trigger. This is a plain-English guide to what a signal actually is, why it matters more than any prediction, and how you drop one into a paper-trading bot here in a few clicks — no code, no real money.

1. The quick answer

Most people think the hard part of trading is predicting — guessing where the price goes. It usually isn't. The hard part is knowing when your edge is actually there, and sitting on your hands the rest of the time. A signal is how you write that down. Instead of "I have a feeling about this market," a signal turns your reason into a rule a computer can check every few minutes:

Because it's a rule and not a hunch, you can test it on history, watch it run live, and — crucially — find out when it's wrong. A vague feeling can't be proven wrong. "Only buy when the VIX is below 18" can.

2. On a prediction market, a price is already a probability

To see why signals matter, remember what a prediction-market price is. A contract trades between 0¢ and 100¢, and that price is the crowd's estimate of how likely the event is — a YES at 62¢ means "about a 62% chance." The market has already done the averaging for you. So your only real job is to find the moments when the crowd's number is off — and a signal is how you pick those moments out of the noise instead of betting on everything.

A signal doesn't have to predict the outcome itself. Often it just tells you that the conditions for an edge are present — the market is stale, a fresh number just dropped, attention is spiking, or a weather bracket is already settled in all-but-name — and lets you act only then.

3. What a signal actually looks like

Signals aren't abstract. They're concrete, public numbers you could look up yourself — we just pull them live, keep them honest, and let you click them into a bot. A few real building blocks from our library, in plain English:

Each one is a free, real, live feed — and there are 40+ of them across 9 markets. The example thresholds above are illustrative; you choose the exact level when you build.

4. A good signal can be proven wrong

This is the part that separates a signal from a superstition. Because a signal is a precise rule on real data, you can backtest it — replay it over history and see whether acting on it would actually have made money — and then watch it trade live in paper money, in the open. If the idea is bad, the numbers say so. We hold our own house bots to the same bar: each one writes down, in advance, the condition that would prove it wrong, and we retire it if that condition trips.

5. The signal library: 40+ blocks, plain English, free

The whole point of this site is that you don't have to gather these numbers, write the code, or babysit the feeds. Our signal library has 40+ free signals across 9 markets — macro, crypto, weather, energy, politics, sports and more — each already live, each described in plain English. You add one to a bot by clicking it, the way you'd add an ingredient to a recipe. "Only when the yield curve is inverted" becomes a real, running rule — no code. New signals get added regularly, and they're all free.

6. From signal to bot in a few clicks

A signal on its own is just a green light. To put it to work you wrap it in a tiny bot: which markets it watches, which way it bets when the signal fires, and how much (everything here is simulated paper money). In the builder you assemble that as a plain-English sentence — pick a signal, pick a market, pick a side — then the bot backtests on history and, if you turn it on, paper-trades live on the public leaderboard so you can watch the idea prove out (or not) in the open. Not sure where to start? Starter bots are pre-built around a single signal — fork one and change the threshold to make it yours.

New to prediction markets? Start with how it works for the mechanics of contracts, pricing, and settlement, or compare the two big books in Kalshi vs Polymarket. Want to go deeper on one category? See the crypto, sports, and weather primers.