Everyone on Kalshi wants the same thing: to buy a contract for less than it's actually worth. That's the entire game. The trouble is most people never define what "actually worth" means — they trade on vibes, headlines, or whichever side feels obvious. This guide walks through how to find positive expected value (+EV) trades on Kalshi the way disciplined traders do: methodically, and after fees.
What +EV actually means
A Kalshi contract settles at either 100¢ (the event happens) or 0¢ (it doesn't). The price you pay is the market's implied probability — a contract at 40¢ is the market saying "about 40% likely."
Expected value is simply the average outcome if you could make the same trade thousands of times:
EV per contract = (your probability × (100 − price)) − ((1 − your probability) × price)
If you think something is 50% likely and you can buy it at 40¢, you're paying 40 for something worth 50 — that's +EV. If you're paying 60¢ for that same 50% shot, you're the one overpaying, and no amount of "I have a good feeling" changes the math.
The whole skill is in one comparison: your probability vs. the market's price. Everything below is about making that comparison honest.
Step 1: Estimate the true probability — honestly
This is the hard part, and it's where almost all the edge (or ruin) lives. Your EV is only as good as your probability estimate. A few ways to ground it:
- Use a sharper market as your anchor. Sportsbooks set lines for a living. If the consensus moneyline implies a team is 57% to win and Kalshi is pricing it at 50¢, that gap is worth a look. (Strip the bookmaker's vig first — more on that below.)
- Be brutally realistic, not hopeful. The fastest way to blow up is to inflate your own number because you want the trade to be good.
- Write the number down before you look at the price. Anchoring to the market price first will quietly drag your estimate toward it.
Step 2: De-vig the market you're comparing to
Sportsbook odds include a built-in margin — both sides add up to more than 100%. If you compare Kalshi to a raw, vigged line, you'll see edges that aren't real. You need the de-vigged (fair) probability.
The quick version: take both sides' implied probabilities and normalize them so they sum to 100%. Our odds converter does this conversion for you, and the Edge Scanner does it automatically across every game — comparing Kalshi's live price to the de-vigged sportsbook consensus and ranking the biggest disagreements.
Step 3: Let the scanner do the heavy lifting
Manually checking dozens of markets is slow. The Kalshi Edge Scanner pulls live Kalshi prices and the current sportsbook consensus, removes the vig, and surfaces every market where the two disagree — sorted by edge. It won't tell you the trade is right (that still depends on whether the market or your read is correct), but it tells you where to look, which is most of the work.
Treat a flagged gap as a starting point, not a green light. Ask: do I have a reason to trust Kalshi's price over the books here, or vice versa? Is one side reacting to news the other hasn't? The scanner finds candidates; you still do the thinking.
Step 4: Subtract the fees — every time
This is where a lot of "edges" quietly die. Kalshi charges a taker fee that's largest on coin-flip contracts near 50¢. A 2¢ edge can be eaten entirely by the fee on entry, turning a "good trade" into a break-even or losing one.
Before you fire, run the numbers through the EV calculator — it computes your expected value after the taker fee and shows your net EV and ROI. If you plan to exit early rather than hold to settlement, check the round-trip cost in the fee calculator too. A trade has to be +EV after fees, not before.
Step 5: Size it with Kelly, not your gut
Finding a +EV trade is only half the job — betting the right amount is the other half. Bet too much and variance can wipe you out even with a real edge; bet too little and you barely grow.
The Kelly criterion turns your edge and bankroll into an optimal stake. In practice, most experienced traders use half-Kelly or less, because it keeps most of the growth while cushioning the damage when your probability estimate is off (and it will sometimes be off).
A realistic example
Say the de-vigged sportsbook consensus has a team at 57% to win, and Kalshi is offering the "Yes" at 50¢.
- Raw edge: you're paying 50 for something worth ~57 — about 7¢ per contract.
- After the taker fee (largest near 50¢), your net edge shrinks, but it's likely still positive — the EV calculator will tell you exactly.
- If it's +EV after fees, Kelly tells you how much to stake. You take the position, and over many similar trades, the math works in your favor — even though any single trade can lose.
That last point is the mindset that separates traders from gamblers: +EV is about being a long-run favorite, not about winning this one bet.
Common ways people fool themselves
- Trusting a vigged line. Always de-vig before comparing.
- Ignoring fees. A pre-fee edge is not an edge.
- Overestimating their own read. If your probability is wrong, the EV math is worthless.
- Overbetting a small edge. Size kills more accounts than bad picks do.
- Chasing illiquid markets. A great price you can't actually fill (or exit) isn't a great price.
The repeatable process
- Estimate the true probability before looking at the price.
- De-vig your comparison line.
- Use the Edge Scanner to find disagreements fast.
- Confirm it's +EV after fees with the EV calculator.
- Size it with Kelly, at half-Kelly or less.
Do that consistently and you stop guessing and start trading a process. That's the whole edge.
For informational purposes only. Not financial advice. Always confirm live prices before trading.