The Platform War That Could Leave Kalshi and Polymarket Holding Empty Bags

The prediction market industry just spent two years convincing everyone they were the future of financial infrastructure. Now the future is arriving — and it’s wearing someone else’s face.

The Stack Is Moving, and the Pioneers Aren’t Driving

Here’s what nobody at Kalshi or Polymarket wants to admit out loud: the infrastructure layer of prediction markets is up for grabs, and the companies that built the user-facing products may not be the ones who capture the value.

We’ve seen this movie before. The early internet had Netscape. Search had AltaVista. Social had MySpace. The pattern is always the same — pioneers prove the concept, then better-capitalized players with existing distribution eat the market. DraftKings just entered the prediction market arena, and they brought 20 million existing customers with them. Meta is experimenting with prediction features for its 3 billion users. Robinhood has signaled interest. Interactive Brokers is watching.

The stack war isn’t coming. It’s here.

What does “stack” even mean in this context? Think of it like a layer cake. At the bottom, you have the exchange infrastructure — the matching engines, the clearing mechanisms, the settlement rails. Above that sits the market-making layer, where liquidity actually gets created. Then comes the data and oracle layer, determining what happened and when. Finally, at the top, the consumer-facing applications where regular people click buttons and place bets.

Kalshi and Polymarket both assumed they’d own multiple layers of this stack. Kalshi built its own CFTC-regulated exchange. Polymarket’s latest markets run on crypto rails with decentralized resolution mechanisms. Both believed vertical integration was the moat.

But moats only matter if the castle is worth taking.

The Regulatory Arbitrage Window Is Closing Fast

For years, prediction markets operated in a regulatory gray zone that rewarded first movers willing to push boundaries. Polymarket built a billion-dollar operation offshore, explicitly barring U.S. users while winking at the VPNs everyone knew they were using. Kalshi waged trench warfare with the CFTC, fighting state by state through constitutional challenges to establish legal footing.

That era of regulatory uncertainty created opportunity. It also created vulnerability.

Now that prediction markets have proven they can generate real volume — and real political attention — the regulation picture is crystallizing rapidly. The CFTC has opened formal inquiries. States are writing their own rulebooks. And here’s the thing about regulatory clarity: it benefits the players with the deepest compliance infrastructure, not necessarily the ones who got there first.

DraftKings already operates in 25 states with full licensing. They have compliance teams, government affairs operations, and relationships with every gaming commission in America. The sports betting giant has already demonstrated it can integrate prediction markets into existing infrastructure without missing a beat.

For Kalshi, the regulatory fight has been expensive and exhausting. Every court victory requires another court victory. Illinois wrote the first real state rulebook, and Kalshi immediately had to file suit challenging it. That pattern — win federal approval, fight state pushback — could continue for years.

And while Kalshi fights, larger players build.

The Liquidity Problem Nobody Wants to Talk About

Here’s something the prediction market evangelists rarely acknowledge: these markets are still shockingly thin.

During the 2024 presidential election, Polymarket saw record volumes. Headlines everywhere. Yet the actual dollar amounts being traded were a rounding error compared to traditional financial markets. The S&P 500 sees $500 billion in daily volume. Polymarket’s biggest day ever was maybe $300 million. That’s not a criticism — it’s context.

Thin markets create two related problems. First, they’re easy to manipulate. A single large trader can move prices in ways that have nothing to do with actual probability assessments. Polymarket’s security breach earlier this year highlighted how vulnerable these platforms remain to bad actors with moderate resources.

Second, thin markets can’t support serious institutional participation. The hedge funds and trading firms that provide liquidity in mature markets need depth to operate. They need to know they can get in and out of positions without moving prices against themselves. Until prediction markets have that kind of depth, they’ll remain retail playgrounds — interesting, but not transformative.

The path to liquidity runs through distribution. And distribution belongs to the platforms that already have it.

What Happens When the Giants Actually Show Up

Mark Zuckerberg told his team to chase the prediction market industry. Let that sink in for a moment.

Meta has 3 billion monthly active users. If they integrate prediction features into Facebook or Instagram — even as gamified, points-based systems that skirt gambling regulations — the user base of every existing prediction market becomes irrelevant overnight. Volume follows attention. Attention follows distribution. Distribution belongs to Big Tech.

The crypto maximalists will argue that decentralization protects Polymarket from this fate. Maybe. But decentralization also means no moat around the user experience. Anyone can fork the smart contracts. Anyone can build a front-end that taps into the same liquidity pools. The permissionless ethos that makes crypto exciting also makes it impossible to defend a market position through technology alone.

Kalshi’s recent push toward a $40 billion valuation suggests they understand the stakes. That number isn’t based on current revenue — it’s a bet that they can become the infrastructure layer for a much larger market. But infrastructure plays only work if you can lock in the data, the liquidity, or the regulatory relationships before someone else does.

Right now, none of those locks are secure.

The Uncomfortable Historical Parallel

Remember when Netflix was “just” the DVD-by-mail company that was going to get crushed by Blockbuster? That analogy cuts both ways.

Yes, scrappy startups sometimes eat incumbents. But the reverse happens just as often. Instagram was “just” a photo filter app until Facebook acquired it. YouTube was “just” a video hosting site until Google turned it into the second-largest search engine on Earth. The technology mattered less than the distribution and resources that followed acquisition or competition from larger players.

The CFTC’s recent focus on Polymarket suggests regulators are paying attention to who actually controls these markets. That attention could accelerate consolidation. If operating a prediction market requires the kind of compliance infrastructure only large financial institutions can afford, the natural endpoint is obvious: the pioneers get acquired, regulated into niches, or simply outcompeted.

Kalshi’s regulatory fight](https://kalshi.com/blog) has been genuinely impressive. They won where others didn’t bother to compete. But winning the right to operate isn’t the same as winning the market.

Where This Actually Goes

The prediction market industry has maybe 18 months before the window of opportunity closes definitively. Either the existing players — Kalshi, Polymarket, and the scattered smaller competitors — find ways to lock in advantages that matter, or they become case studies in business school classes about first-mover disadvantage.

The smart money says one of three things happens. Scenario one: major financial institutions acquire the successful prediction market operators, folding them into existing infrastructure the way banks absorbed fintech startups a decade ago. Scenario two: Big Tech builds competing products that overwhelm through distribution, leaving current players as niche operators serving crypto enthusiasts and trading obsessives. Scenario three: regulatory fragmentation creates protected markets where different players dominate different jurisdictions, leading to a messy equilibrium nobody particularly wanted.

None of these scenarios involve Kalshi or Polymarket continuing to grow independently into dominant global platforms. That outcome remains possible but increasingly unlikely as the competitive landscape shifts.

The prediction market industry proved something important: people want to bet on real-world outcomes, and those bets generate meaningful price signals. That insight is now public knowledge. And public knowledge gets exploited by whoever has the resources to exploit it.

The stack war has begun. The pioneers might not survive it.