Prediction markets are (also) developer platforms

Last week we announced our investment into Gondor, a company building financial services for prediction market users. In our investment post I wrote about how prediction markets are not just end user destinations, but also developer platforms and I wanted to expand on that.

The two leading prediction markets today, Polymarket and Kalshi, both have an ecosystem of third party developers building on top of them. Developers choose to do so because they can leverage the existing user-base and liquidity of these underlying platforms to build their own products and services.

Some categories of apps emerging on top of prediction markets include:

  • Trading terminals: interfaces that aggregate markets across different prediction markets and allow users to place trades through a single UI.
  • Bots: trading bots on telegram or other social media / messaging services through which users can place trades.
  • Data / analytics: services that provide users with enriched data on markets.
  • DeFi tooling: leverage, lending and other tools for prediction market users.
  • Developer tooling: APIs, SDKs etc to make building on top of prediction markets easier.

What’s worth noting is that under the hood, Polymarket and Kalshi have different architectures.

Polymarket is built natively onchain using Gnosis smart contracts. It is like a DeFi protocol in its construction. Every position on Polymarket is a token that is settled on the Ethereum blockchain. Polymarket also has an API that developers can use, but they don’t need to. They can build on top of Polymarket positions directly without asking for Polymarket’s permission.

Kalshi on the other hand started by launching an offchain platform, which means that developers need to build on its API as the positions aren’t tokenized and settled onchain. However, just last week, Kalshi announced the release of an onchain version of its product on Solana. Positions, like on Polymarket, will be tokenized on the Solana blockchain. There’s still not a lot of detail on this and it'll be interesting to see how it develops.

Over the next couple of months I expect to see a lot more movement in this space as more companies want a piece of the pie. Robinhood recently announced plans to build a prediction market of its own and the large sports betting companies, DraftKings and FanDuel, are also working on their own prediction market offerings.

More platforms will emerge, a mix of onchain and offchain, and the applications built on top will have to cater to both architectures.

My personal view is that the onchain, permissionless markets will support the more vibrant ecosystems. Developers want assurances that the platform underneath them can’t suddenly change the rules on them and blockchains provide that kind of guarantee. The benefits of a thriving third party app ecosystem on top will ultimately accrue to the prediction markets underneath.