Dan Matuszewski says the market’s winning setup is turning into its most crowded trade
Dan Matuszewski, co-founder of CMS Holdings and former head of Circle Trade, said the barbell trade that has defined much of this crypto bull cycle now carries a growing risk: crowding.
Speaking on The Rollup podcast, aired on Sept. 15, 2026, Matuszewski said he agrees with the broad framework. One side of the barbell is made up of hard-money or macro-sensitive stores of value. The other is made up of onchain businesses that keep generating cash and return it to holders through buybacks or distributions. His warning was not that the thesis is wrong. It was that too many people are now running the same playbook.
「This has been repeated into the same narrative by everyone. That doesn’t mean it’s wrong, but once everyone crowds into the same investment logic, the edge gets crowded out,」 he said.
The episode was hosted by The Rollup team. The Chinese transcript cited in the input was compiled and translated by TechFlow. The program notes also said CMS holds or has invested in several assets mentioned in the discussion, including BTC, ETH, ZEC, HYPE, Monad and Chainlink, and that the views were Matuszewski’s own rather than investment advice.
CPI day and macro noise
The episode was recorded on the day fresh CPI data came out. Matuszewski described the print as hot, but not as hot as expected, and said the market appeared relieved.
He argued that macro-heavy stretches are often the hardest periods to trade because participants get pulled around by information where they have little edge. He said the coming election cycle is likely to keep dominating headlines, leaving crypto exposed to outside forces for some time.
In his view, the old pattern of buying the rumor and selling the news has, at times, flipped into something closer to fear being priced in ahead of the event. By the time the event arrives, the outcome is often less damaging than feared, and prices can move higher. The pressure comes before the release, not always after it.
What signals he watches now
Matuszewski said funding rates no longer work as cleanly as they once did. He pointed to the BitMEX-dominated era, when negative funding could be treated as a straightforward long signal and annualized funding above 30% could be treated as a straightforward short signal.
Today, he pays more attention to aggregate open interest. If total altcoin open interest rises above Bitcoin open interest, he said, that is a sign the market has shifted into a more speculative phase and capital is moving toward higher-beta assets in search of excess returns.
He also watches two proxy indicators. One is the basket of relative winners, including names such as HYPE, NEAR, ZEC and Fartcoin. If those assets keep getting bought on dips and maintain strength, he takes that as evidence that money is still flowing in. The other is the stock market, because in his view crypto remains equity-led in the short term.
Why crypto cannot support itself
Asked why crypto cannot stand on its own, Matuszewski said the market simply does not have enough independent internal capital flows. He said there has not been a wave of new fund formation, nor a steady stream of fresh money entering the space. Retail comes quickly and leaves quickly, and sustained net inflows have been absent for a long time.
That leaves crypto unable to self-support in a real drawdown. If equities fall 10%, he said, crypto’s own inflows are not enough to absorb the shock.
He also pointed to DATs, or digital asset treasury companies. If the market stabilizes, he said, those vehicles should show beta on the way up. But they also show beta on the way down. The more meaningful signal would be a period of consolidation followed by premium trading while they continue raising capital and buying the underlying assets.
The “10-day rule” and staying invested
The hosts brought up Tom Lee’s “10-day rule”: since 2014, the best 10 days of each year accounted for 162% of market returns, while the other 355 days averaged a 14% annual decline.
Matuszewski said he fully agrees with that framework. For him, the takeaway is simple: be in the market. Hold assets. Do not obsess over catching the exact bottom.
Most of the return comes from a very small number of days, he said, and many investors end up mentally broken not because they got the direction wrong, but because they were not there when those days happened. Over the long run, he said, sitting on your hands and holding assets matters more than trying to trade every isolated move.
The barbell: hard money on one side, onchain cash machines on the other
The hosts said they had cut exposure near the end of the previous year, took roughly a 20% drawdown, but avoided the worst stretch that followed. Over time, they said, they arrived at a portfolio theory for this cycle: the winning setup was a barbell. Matuszewski said he agreed.
One side includes macro-sensitive stores of value such as BTC, ZEC and gold, with ETH also fitting into that bucket. The hosts said that on the hotter CPI day, BTC rose about 10% and ZEC surged, which they framed as the fiat-debasement side of the trade.
The other side includes onchain businesses that keep “printing money” and return capital to holders, including names such as HYPE. The hosts said they built positions early in HYPE, ZEC and Venice, and still hold many of them. Their approach was to buy those two buckets in spot and think about trimming only once the market moved into a more speculative phase.
Matuszewski’s addition was blunt: watch the crowding.
Crowding risk and the memory of 2021
When asked whether this looked like mid-cycle crowding or late-cycle crowding, Matuszewski did not pin down a precise stage. Instead, he pointed back to 2021, when nearly everyone loaded into what he called “that basket” as the future of the market. Then came violent deleveraging. Solana came back later, he said, but the other two-thirds of those positions were effectively wiped out.
His broader point was structural. Crypto does not have the same persistent inflow base as equities, where retirement products and other institutional channels keep money moving in. Crypto, by contrast, behaves more like a closed pool of capital that expands and contracts with inflows and outflows.
That means when everyone crowds into the same trade and the whole asset class starts shrinking, the most crowded names can fall the hardest.
He also pushed back on the idea of rotating into second-tier names that have not yet moved. In theory, he said, capital can spill over from the leaders into laggards such as Lighter or Hyperliquid. But he described that as a dangerous game because it amounts to guessing where the next marginal dollar will go. If those rotations start happening faster and faster, he said, that itself is a sign the cycle is getting older.
The next marginal buyer
Matuszewski said market participants should be careful not to confuse Crypto Twitter attention with actual capital. Drawing on four to five years in OTC markets, including the 2017 bull run, he said some of the largest pools of money come from people most traders do not know and never see online.
On institutional adoption, he was skeptical about how far things have really gone. 「To say any large institution is really putting these assets on the balance sheet is basically still a joke right now. Other than IBIT, there’s almost nothing,」 he said.
He added that traditional large allocators still have not truly entered, and that BTC is the only asset that has genuinely crossed that threshold. At the same time, he said, that leaves a long runway ahead, with capital likely to arrive as a slow multi-year tailwind.
His ranking of the next marginal buyers was as follows:
- crypto-native capital and liquidity funds, which he said are still building positions in a FOMO-driven way and are roughly halfway done;
- smart high-net-worth individuals allocating through ETFs;
- larger Wall Street institutions;
- and finally retail, coming in through FOMO and social trading rather than “dinosaur coins.”
ETH, Solana and the monetary value of L1 assets
Matuszewski said the value of layer-1 native assets comes from usefulness inside their own systems. That usefulness, in his view, is what allows value to accrue.
He used several examples. BNB gained utility through IEOs because users needed BNB. ETH moved from the ICO era into DeFi as a base asset inside the system. Solana, he said, became a reserve currency for meme trading, with large amounts of capital locked in pools. What these chains share is a large enough surface area for the native asset to become deeply useful. That, he said, is what investors are really buying.
When the host said that framing made him extremely bullish on ETH because “they have value for a reason,” Matuszewski said he agreed and that he had simply not articulated the logic clearly before.
On Monad, he said he is constructive because the team will keep iterating until it finds a value-capture mechanism for the token. The only real risk, he said, is running out of money before getting there.
He extended that logic to other L1s and L2s as well. The risk is usually one of two things: either the founders get tired and stop, or the project genuinely runs out of capital. As for Solana, he said a large part of its long-term value proposition rests on Anatoly Yakovenko continuing to iterate relentlessly.
Ethereum’s identity problem
The hosts suggested Ethereum may have hurt itself by being too committed to decentralization and too unwilling to pick winners. Matuszewski agreed. He contrasted that with a market that has become more comfortable with centralization, pointing to Hyperliquid’s 24 nodes as something many participants now accept.
He described Ethereum’s “identity crisis” as the result of building much of the early infrastructure without seeing enough real use stick to it. In response, the ecosystem shifted toward stablecoins, tokenized assets, perpetuals and prediction markets. In his telling, many of those directions emerged out of that same identity crisis.
He also discussed the survival challenge facing infrastructure projects such as bridges, ZK and intent-based systems in the $50 million to $150 million market-cap range that are not profitable. Whether they survive, he said, depends on whether teams are willing to keep reinventing themselves. He cited Ondo’s move from impermanent-loss hedging into RWA as an example.
Why he thinks tokenized stocks can outgrow stablecoins
Matuszewski was clear that he dislikes pairing memes and stocks on Robinhood in a way that makes crypto look unserious just as it enters traditional finance. But he was equally clear that tokenized equities themselves can keep moving higher for years and, in his view, become larger than stablecoins.
His argument was not about convenience for U.S. investors who can already open brokerage accounts. It was about access. Just as many people around the world want dollars but cannot easily hold them directly, many also want exposure to U.S. equities such as Nvidia and cannot get it for a range of reasons.
Turning those equities into bearer-style tokenized assets, he said, is powerful. It opens a new global capital base for U.S. companies without giving up governance rights.
When the host described that as “exporting wealth creation,” Matuszewski agreed. The U.S. already exports dollars, he said, and can now export its financial markets as well. The discussion also referenced the idea that a China version of the S&P 500 has been roughly flat over the past 20 years, while owning U.S. companies has been the compounding machine. On that basis, he said tokenized stocks can open up the global capital base for U.S. companies and eventually exceed stablecoins in size.
Still, the episode noted one practical issue: trading volume after the Robinhood chain launch has been weak, leaving questions about sustainability.
Who benefits from $35 billion in onchain tokenized assets
The hosts said onchain tokenized assets currently stand at $35 billion and asked who benefits. Matuszewski said that is exactly where large pools of capital will want to look, but the answer is still hard to pin down early on.
He said the likely early beneficiaries include the L1s that host those assets, oracle networks such as Chainlink, and some issuers, including a few that are already public and investable.
He also said the companies already benefiting from “legalization” are those that have been inside the CFTC’s innovation advisory process and have been named by Trump, citing Hyperliquid as an example.
At a broader level, he said L1s could benefit again if trillions of dollars of assets move onchain, including stablecoins, short-term Treasuries, money market funds and stocks. More activity would mean more TVL, more fees, more earnings and more buybacks. The hard question, he said, remains the same: who actually outperforms.
Social trading is taking share from gambling, not investing
Near the end of the episode, one host said he had sent $400 to $500 worth of meme tokens to several friends who were sports bettors, and that some of those accounts had already gone up 3x to 4x. He said he gave them rules: do not add more principal, always be selling, keep a moon bag, rotate gains from smaller caps into larger caps to build a base, and treat it as a portfolio rather than a one-off bet.
Matuszewski said social trading is fundamentally taking share from existing forms of gambling: sports betting, casinos and even a large portion of perpetual futures trading, which he said is itself often gambling. In that sense, it is a substitute product, not an investment product.
He added that FOMO and Pump.fun may charge a heavier take, but the odds of a hit may be slightly better than in sports betting, making the experience closer to a lottery or parimutuel-style game. In either case, he said, participants should expect to lose money. The difference is mostly in how much the system takes out.
The core warning
Matuszewski did not argue that BTC, ETH, ZEC, HYPE, Solana or tokenized equities have no place in the market. His repeated point was narrower and more tactical. A strategy can work and still become dangerous once it turns into consensus positioning.
In his framework, the bigger risk now is not that the hard-money plus onchain-cash-flow thesis suddenly stops making sense. It is that too much capital is now concentrated in the same template. In a market he describes as a closed pool of capital, the most crowded positions are often the first to crack when the pool starts shrinking.