If you’re comparing Kalshi vs Polymarket perps, one number does most of the work. Kalshi caps Bitcoin perpetuals at around 5.8x. Polymarket, which launched its own perps on September 4, goes to 20x. Same product category, same month, roughly a fourfold gap in what you’re allowed to borrow.
That gap isn’t a difference in temperament. It’s a clearinghouse. Once you see why, the number on the leverage slider tells you more about a venue than any of its marketing does.
| Kalshi | Polymarket | |
|---|---|---|
| Perps live since | June 3, 2026 | September 4, 2026 |
| Max leverage | ~5.8x on BTC, ~4.4x on ETH, 2.1x to 3.4x on smaller listings | Up to 20x |
| Markets | BTC, ETH, XRP, SOL, LINK, HYPE | 67, including stocks, indexes and commodities |
| Regulation | CFTC-approved, own clearinghouse and futures broker | Offshore, outside US reach |
| US traders | Yes, with full verification | Blocked and redirected |
I wrote in July that leverage was coming to prediction markets and that it should make you nervous. It arrived faster than I expected and in a shape I didn’t predict. Here’s what happened, why the two numbers are so far apart, and the part of my own article I got wrong.
What actually happened, in order
May 29. The CFTC approved Kalshi’s Bitcoin perpetual contract. Not tolerated, not left alone in a grey zone. Approved.
June 3 to June 10. Bitcoin perps went live, then Ether, then XRP. Volume hit $100 million in the first 24 hours and crossed $1 billion inside the first week.
August 18. Kalshi filed to extend perps to equity indexes.
September 4. Polymarket launched its own perps: 67 markets at up to 20x, covering crypto, single stocks including Tesla, Nvidia and Coinbase, the S&P 500 and Nasdaq 100, and commodities down to Brent and WTI crude. US traders are blocked.
Six weeks, start to finish. That’s the pace this industry moves at, and it’s why an article with a date on it stops being true faster than the writer would like.
Why the regulated venue gives you less
Kalshi’s caps look almost disappointing next to an offshore crypto exchange. Under 6x on Bitcoin, barely 2x on the small listings. Polymarket, operating outside US reach, offers 20x on assets Kalshi doesn’t list at all.
Kalshi isn’t being cautious by temperament. It’s carrying a cost Polymarket doesn’t. Kalshi runs perps through three registered entities: the exchange itself, a futures commission merchant that holds your margin, and its own CFTC-regulated clearinghouse. A clearinghouse has to be able to absorb a bad day. When somebody stands behind the trades, somebody has to size the risk, and the leverage cap is where that requirement shows up on your screen.
The offshore venue has no such obligation and prices accordingly.
So the multiple on offer is now the clearest single signal of what kind of venue you’re standing in. High number, nobody behind you. Low number, somebody is on the hook.
That’s the opposite of the intuition most people bring to this, which is that the unregulated place is where you find restraint and the regulated one is where you find rules. Here the rules are the restraint, and you can read them off the slider.
What I got right in July
The mechanism. I wrote that leverage changes what you’re betting on, that unlevered you’re forecasting an outcome and levered you’re forecasting the path, and that a thin market plus borrowed money is how a correct call becomes a liquidation anyway.
Kalshi’s own risk page puts it more bluntly than I did. Positions can be closed without notice, and in a fast or illiquid market the fill can come in worse than the liquidation trigger, which can leave the account negative. Read that twice. The floor I described in July, where the worst case is losing your stake, is gone on these products.
What I read wrong
I expected regulators to look at 10x perps and recoil. I wrote that a state attorney general reading about them wouldn’t come away thinking this was a sober forecasting tool.
The federal regulator went the other way. The CFTC approved the contract, and then approved the clearinghouse behind it. Whatever else that is, it isn’t a regulator recoiling.
I still think the state-level fight I described is real. I got the direction of the federal one backwards, and the reason matters: the CFTC appears more comfortable with a leveraged product it can supervise than with an unleveraged one it can’t.
The part nobody is saying out loud
Look at Polymarket’s 67 perp markets again. Bitcoin, Tesla, Nvidia, SpaceX, the S&P 500, gold, Brent crude.
None of that is a prediction market. There’s no event, no resolution source, no question with an answer. It’s a crypto derivatives exchange sharing a brand and a login with a prediction market, and the two sit one tab apart.
That matters beyond tidiness. The legal case these platforms have been making, the one I covered in the state-by-state breakdown, rests on event contracts being financial instruments rather than bets. A perpetual on WTI crude isn’t an event contract by any reading. The argument that wins the prediction market fight doesn’t obviously cover the product now sitting beside it.
What this changes for you
Nothing, if you came here to forecast events. The unlevered markets work exactly as they did in July, and the costs and order-book notes still describe what you’re paying. If you’re weighing the platforms themselves rather than their perps, that’s the Kalshi review and the Polymarket review.
If you’re tempted by the perps tab, one thing before you open it. You’d be leaving prediction markets entirely and entering a product where being right about the outcome doesn’t protect you. Different skill, different failure mode, and the fact that it sits one tab away from the thing you understand is the design decision I’d argue with hardest.
My July advice holds, and I’d put it more strongly now. Leverage is a feature to ignore while you’re learning. The plain version of these markets is already enough to lose money on.
This article updates Leverage Is Coming to Prediction Markets, published 27 July 2026. Figures verified 5 September 2026. Leverage caps and market listings change often, so check the platform before acting on anything here.
Nothing here is financial advice. Prediction markets involve real risk, so never trade money you can’t afford to lose.