The Final Bell Curve: What Does Crypto Look Like in 1 and 5 Years?

Bell Curve is ending its run as a podcast, and this final episode focuses on where crypto is headed over one-year and five-year horizons, with the hosts offering concrete predictions, disagreements, and structural critiques of the industry.

On a one-year view, the speakers argue that distribution is the dominant force in crypto and that all roads lead back to Coinbase, Kraken, Binance, and Robinhood. These platforms are uniquely vertically integrated in a way that does not exist in traditional finance, combining custody, brokerage, and exchange under one roof. Trading has functioned as a customer acquisition mechanism rather than a core business, analogous to how Fidelity makes most of its money on 401K packages and asset management rather than on trading fees, which dropped from five dollars to zero without meaningfully hurting the firm. Coinbase and Robinhood are evolving into app stores for financial products, and both have already expanded into savings products and captured millennial wallet share. Robinhood launched Robinhood Ventures as a public vehicle for access to large private companies like Stripe and Anthropic, and Coinbase launched pre-IPO perpetuals. Secondary market volume for private assets is up 40 to 50 percent in the past year, and the speakers predict a tipping point when a major distributor secures exclusive access to a high-quality issuer, causing users to shift money away from incumbents like Fidelity.

Building a blockchain is no longer technically difficult and could be accomplished with AI tools today, meaning the scarce resources in crypto are now liquidity and distribution rather than the ability to build a chain. No successful infrastructure launch in the past 24 months has performed even remotely well. The barrier to entry for new L1s has never been higher, and few if any new ones are likely to emerge. Arbitrum and Optimism are better positioned than new entrants because they have existing brand, de-risked technology, and flagship customers including Base, Robinhood, and Tempo. Robinhood chain is described as the first good example of what it looks like when an incumbent launches a chain successfully, and after multiple incumbent chains launch, competition will intensify between incumbents themselves rather than primarily between incumbents and original crypto projects. Hyperliquid is cited as one of the most successful projects of the past 24 months but is predicted to face a fork in the road from its current full-stack position.

The smile curve framework is applied to argue that value accrues at the distribution end and at the scarce infrastructure end, with the middle increasingly difficult for startups to compete in. Incumbent TradFi and payment organizations including Stripe and Ramp are well positioned in stablecoin payments, custody of tokenized securities, and trading. Among startups, the biggest winners have caught waves of adoption momentum on the infrastructure side. Commonware is cited as what Cosmos and Atom should have been, Canton is benefiting by powering Wall Street tokenization initiatives, and Rain caught the neobank wave by pivoting from B2C to powering hundreds of incoming neobanks. Prediction markets and stablecoins are already working and expected to keep growing, while on-chain asset management and structured financial products are expected to emerge and boost Ethereum and Arbitrum in particular.

On a five-year horizon, the speakers express genuine uncertainty about how products will be accessed when agents, whether run by a person, a business, or operating as self-sovereign entities, become the primary users. A major startup opportunity is identified in becoming the distribution layer for agents in crypto. Apps are predicted to transition into skills within a harness or operating system rather than standalone interfaces. The overall prediction is that crypto gets smaller before it gets bigger, meaning fewer companies and fewer things that work, but the things that do work will do so unbelievably well.

Ethereum is expected to win the rollup market with corporate chains eventually becoming rollups on Ethereum, but whether Ethereum can generate significant fees from that activity is unclear. Ethereum is characterized as too passive in attracting use cases, too fragmented in development direction, and unclear about what it wants on its chain. Solana is described as cheap, functional, and having a roughly comparable Lindy effect to Ethereum for certain use cases, and is seen as deliberately repositioning toward institutional credibility. Neither Ethereum nor Solana is expected to win the infrastructure layer that AI agents most heavily transact on, with a credibly neutral layer focused on operating AI agents as self-sovereign entities seen as more fitting for that role.

Crypto currently performs worse than traditional finance in terms of asset information transparency and the extractiveness of intermediaries. Infrastructure has been set up specifically to dump tokens, and even good projects are structurally set up to have their tokens decline due to token-based incentive spending. Projects feel compelled to spend tens or hundreds of millions of dollars on incentives to compete, but a project that spent nothing would likely have been in a far better position from a token standpoint. Many of these structural problems are expected to be solved in the next one to two years, ideally before institutions truly adopt the technology and asset class.