The Great Migration:
Two Migrations, One Destination

There is a question that nobody in crypto wants to answer honestly: where did everybody go?

Not the capital — that is easy to track. Not the institutions — they are louder than ever. The question is about the energy. The degenerate conviction that made someone ape into a dog token at 3am or mass-migrate liquidity to a new L2 on the strength of a Medium post. The raw, irrational speculative impulse that powered this industry from its inception.

It did not vanish. It split in two.

One stream found prediction markets — a better casino with clearer rules, real-world stakes, and outcomes that resolve in days instead of "next cycle." The other was captured by traditional finance, which stopped trying to build its own crypto products and simply started buying the ones that already worked.

These are not separate stories. They are two halves of a single structural shift. The market is rationalizing. And rationalization, it turns out, looks nothing like what crypto expected.

The Speculation Layer Moved. Most People Haven't Noticed.

Start with a number that should bother every token project founder reading this: 53.2% of all tokens listed on GeckoTerminal were functionally dead by the end of 2025. Not underperforming. Not in a drawdown. Dead. Of the 11.6 million tokens that launched, more than half ceased to have any meaningful trading activity.

The standard narrative frames this as a bear market problem — cyclical, temporary, fixable with the right catalyst. It is not. What happened to altcoins is what happens to any market where the expected value for new entrants goes negative and stays there long enough for behavior to change.

Here is the part that matters: people did not stop speculating. They stopped speculating here.

Polymarket hit 688,000 monthly active users. Kalshi's single best NFL weekend — $441 million in volume — eclipsed the record set during the 2024 US presidential election.

Why the Casino Moved

The surface explanation is that prediction markets offer "defined risk" and "transparent odds." This is true but insufficient. Plenty of financial instruments offer defined risk. Options have existed for decades. The deeper explanation requires understanding what crypto speculation actually was — and what participants discovered they were really paying for.

Altcoin trading was never purely about financial returns. It was a social game dressed in financial clothing. The thesis was the content. The community was the distribution. The token was the scorecard. When someone bought an obscure DeFi governance token, they were not making a DCF-based investment decision. They were joining a tribe, signaling beliefs about the future of finance, and hoping the tribe would grow large enough to make the early signal profitable.

This works until the tribes stop growing. And once more than half of all tokens die and 38% of survivors trade near all-time lows — a figure worse than the FTX collapse — the tribe formation mechanism breaks. Why would anyone join a new tribe when every recent tribe has been decimated?

Prediction markets offer something altcoins never could: a game where the skill component is legible and the outcome is objective. When you buy a contract on whether the Fed will cut rates in June, your edge comes from understanding macroeconomics, Fed communication patterns, and labor market data. That edge is researchable. It is expressible. It does not depend on whether enough people share your narrative to bid up a reflexive asset.

The binary structure — $0 or $1, yes or no — eliminates the most destructive feature of altcoin markets: the ambiguity about when you are wrong. An altcoin can decline 70% and its holders will insist the thesis is intact. A prediction market contract resolves. The argument ends. Capital is freed for the next trade.

For a generation of traders who spent three years watching "the thesis" deteriorate in slow motion while locked in illiquid positions, that finality is not a limitation. It is the entire point.

The Post-Election Test

The most revealing data point in prediction markets is not the growth curve — it is the recovery.

When Polymarket's volume crashed 84% overnight after the November 2024 US election, the consensus was obvious: prediction markets were a political novelty. Election toy. One-cycle phenomenon. By January 2025, daily active users had dropped to 24,000.

Eighteen months later, Google search interest for "Polymarket" hit 100 — its all-time peak — exceeding the November 2024 election high. This happened without a comparable political event. It happened because the platform expanded into sports, economic data, geopolitical events, and cultural markets, and participants discovered that the engagement loop — research, position, resolve, repeat — was inherently addictive regardless of the underlying event category.

Kalshi figured this out faster and more aggressively. Sports now account for 90% of its volume. The valuation trajectory — $2B to $5B to $11B in under a year, with $20B now on the table — looks less like a startup fundraising arc and more like the growth curve of a new exchange category.

The regulatory picture reinforces this. CFTC Chairman Michael Selig withdrew the proposed ban on political and sports event contracts — a complete reversal from the previous administration. The agency has received 17 Designated Contract Market applications, approved seven, with ten pending. Both the Kalshi and Polymarket CEOs sit on the CFTC's Innovation Advisory Committee. Polymarket received approval to reenter the US market through a registered intermediary.

The federal government is not tolerating prediction markets. It is institutionalizing them.

The counterargument — that prediction markets are just gambling with better branding — misses the structural point. The same was said about derivatives in the 1970s. What matters is that a $40 billion+ annual volume market with 97.5% concentration in two platforms and explicit regulatory endorsement is not going away. The speculative energy that used to flow into token launches is being rerouted here, and the plumbing is being built to keep it.

The Infrastructure Layer Was Acquired. Not Disrupted. Acquired.

The second migration is quieter and, in the long run, probably more consequential.

There is a version of crypto history where decentralized protocols gradually replace centralized intermediaries — where Uniswap eats the NYSE, where MakerDAO eats JPMorgan, where self-custody wallets eat Bank of America. That version of history is not happening. What is happening is the reverse: centralized intermediaries are eating crypto.

Not competing with it. Buying it.

$8.6 Billion in Acquisitions: What Was Actually Purchased

These are not technology bets. They are consolidation plays. Each company is replicating the conglomerate model that has governed traditional financial services for a century: own the exchange, own the custody, own the data, own the client relationship.

The top three M&A categories — Investing & Trading, Brokers & Exchanges, and Stablecoins & Payments — represented 96% of deal value at $42.5 billion. The market is not buying innovation. It is buying market share, regulatory licenses, and customer bases. This is how financial services consolidation has always worked. Crypto just took fifteen years to arrive at the same playbook.

The Banks Did Not Build. They Waited. Now They Are Moving.

While crypto-native companies consolidated each other, traditional finance did something more subtle and arguably more significant: they started offering crypto through existing distribution channels, making the native platforms less necessary.

Morgan Stanley's E*Trade is launching spot BTC, ETH, and SOL trading in the first half of 2026. Think about what that means. The brokerage platform sitting inside every Morgan Stanley wealth management relationship — managing trillions in assets — will offer crypto alongside stocks, bonds, and options. No new app to download. No KYC with an unfamiliar exchange. No self-custody anxiety. Just crypto, in the same portfolio view as everything else.

Charles Schwab rolled out 24/7 crypto trading. SoFi became the first US bank to offer retail crypto trading under the new regulatory framework. Bank of America is advising wealth clients to allocate up to 4% of portfolios to crypto. Robinhood generated $680 million in crypto revenue in just the first nine months of 2025, up 154% year-over-year.

Coinbase's inclusion in the S&P 500 — the first crypto company in the index — crystallized this dynamic. Bernstein estimated $16 billion in buying pressure from passive funds alone. The stock jumped 24% the day after inclusion. But the deeper implication is that every retirement account, every pension fund, every target-date fund that tracks the S&P 500 now has crypto exposure whether the account holder knows it or not.

$33 Trillion: The Stablecoin Number Nobody Talks About

The stablecoin market cap — $314 billion, with USDT at $184 billion and USDC at $78 billion — gets cited regularly. The number that should receive ten times the attention is transaction volume: $33 trillion in 2025. That is a 72% year-over-year increase. That is on par with Visa's annual throughput. That is not a crypto metric. That is a global payments metric.

And the issuers are changing. JPMorgan launched JPMD on Base. Ten of the world's largest banks — Bank of America, Goldman Sachs, Citi, UBS, Deutsche Bank, MUFG, Barclays, TD, Santander, BNP Paribas — are exploring a consortium for G7-currency stablecoins. Nine European banks are building a MiCA-regulated euro stablecoin for mid-2026 launch. Stripe's Tempo platform, built on its $1.1 billion Bridge acquisition, launched with Visa, Deutsche Bank, and Shopify as partners.

The GENIUS Act gave banks the regulatory clarity they needed. They are not experimenting. They are building production systems. When the world's largest banks issue stablecoins on public blockchains, the word "decentralized" requires significant qualification.

Tokenization: Blockchain as Plumbing

BlackRock's BUIDL fund holds $18 billion in AUM across nine blockchain networks, listed as collateral on Binance, live on Uniswap. JPMorgan launched MONY, a tokenized money-market fund on Ethereum. Tokenized US Treasuries alone account for $8.7 billion. Tokenized equities surpassed $1 billion, up 2,900% year-over-year.

What is already accurate is the directional bet: traditional assets are moving onto blockchain rails, managed by traditional institutions, distributed through traditional channels. The blockchain is the database. The institution is the operator. The user may never know they are interacting with a distributed ledger.

This is the trojan horse working in reverse. Crypto's innovation — programmable, composable, 24/7 settlement infrastructure — is being absorbed into the existing financial system rather than replacing it. The rails are open. The access points are gated. (For more on how we think about digital asset infrastructure and strategic advisory in this new landscape, see our work.)

Two Migrations, One Destination: The Market Is Rationalizing

Here is the unifying thesis: both migrations — speculation to prediction markets, infrastructure to TradFi — are expressions of the same underlying process. The crypto market is rationalizing, and rationalization means the separation of speculation from infrastructure.

For fifteen years, these were bundled together. You could not participate in crypto's infrastructure without also participating in its speculation. The infrastructure was the speculation. The speculation funded the infrastructure. They were inseparable.

They are separating now.

Speculation is migrating to venues purpose-built for it. Infrastructure is migrating to operators purpose-built for it. What remains in between — the token layer — is being compressed from both sides. From above, by institutions that offer exposure to digital assets without requiring token ownership. From below, by prediction markets that offer speculation without requiring belief in a protocol's long-term value.

What This Means for Different Participants

The IPO Pipeline Is the Scoreboard

The upcoming IPO cycle will test whether the market's rationalization holds. Kraken filed at a $20 billion valuation. Consensys is working with Goldman and JPMorgan on a mid-2026 listing. Ledger is targeting $4 billion. Animoca is going through a Nasdaq reverse merger. Bithumb is aiming for KOSDAQ.

The early results are cautionary. Of the 2025 crypto IPOs, only Circle and Galaxy Digital remain in profit. eToro is down 58% from listing. Bullish is down 52%. Gemini is down roughly 80%. The "peak fast, bleed slow" pattern that destroyed altcoin investors is showing up in equity wrappers too — strong day-one pops followed by sustained declines when fundamentals do not support the valuation.

This is actually healthy. It means the equity market is applying the same discipline to crypto companies that it applies to every other sector. Companies with revenue, moats, and regulatory positioning are being rewarded. Companies without clear differentiation are being punished. The market is rationalizing.

What Remains

There is a temptation to frame all of this as the death of crypto's original vision — the cypherpunk dream consumed by the very institutions it sought to circumvent. That framing is partially correct and entirely incomplete.

These are not numbers that describe a failed experiment. They describe an experiment that succeeded in ways its architects did not intend and cannot fully control.

The permissionless settlement layer still exists. The composability still works. The 24/7 markets still run. What changed is who operates on top of them and for what purpose. The rails are open-source. The businesses built on the rails are not.

For participants who came to crypto for the speculation, prediction markets offer a better version of what they were looking for. For participants who came for the technology, the technology is being deployed at a scale that was unimaginable five years ago — just by operators who look a lot more like JPMorgan than like a pseudonymous Discord server.

That was always supposed to be the point. It just took the death of 11.6 million tokens and the birth of a $40 billion prediction market for the industry to figure it out. If you are building at this intersection, we should talk.