A crypto protocol can generate $100M in annual revenue and still be one of the worst tokens to hold through a bear market. The numbers behind crypto’s revenue cycles are pretty brutal. Across 227 protocols, median revenue contracted by more than 60% during major market downturns. + DEX-volume downturn: -62% + Bitcoin-market downturn: -65% A protocol earning $10M monthly could see that fall toward $3.5M under comparable conditions. Now imagine valuing that business using earnings from its strongest quarter. The multiple might look attractive, but the underlying revenue could be far less durable than it appears. I think this is one of the biggest problems with how crypto fundamentals are evaluated. Over the past 30 days, protocols generated approximately $1.22B in revenue. + @tether : $506M + @circle : $203M + Combined contribution: ~58% Two stablecoin issuers accounted for more than half of tracked revenue. Their economics are quite different from those of DEXs, derivatives exchanges and launchpads, where earnings depend heavily on trading activity and leverage. Stablecoin issuers earn primarily from reserve assets and outstanding balances. Lending protocols depend on borrowing demand, utilization and interest rates. These businesses can all experience revenue declines, but they don’t necessarily contract for the same reasons or at the same pace. And there’s another complication. A protocol generating more revenue doesn’t automatically have better token economics. Consider two hypothetical businesses: + Protocol A: $50M annual revenue, with $40M allocated to token buybacks. + Protocol B: $100M annual revenue, with no direct tokenholder distribution. Now assume A’s revenue falls 25% during a downturn while B’s declines 70%. A would still generate $37.5M annually. B would fall to $30M. The smaller business would now earn more through the downturn, while maintaining a clearer mechanism for returning economic value to its token. That doesn’t guarantee A’s token will outperform. Entry valuation, dilution and buyback execution still influence returns. But it shows why revenue size alone tells us very little about the quality of a token investment. A derivatives exchange reporting record fees during a leverage-driven rally could be an excellent business. The question is how much of that revenue remains when leverage disappears and trading activity contracts. The same applies to launchpads and protocols that rely heavily on incentives to attract users. Personally, I’d rather evaluate a business generating $30M consistently across market conditions than one earning $100M at the peak and $10M when activity dries up, assuming comparable valuations and tokenholder economics. Because a high-revenue protocol isn’t necessarily a durable business, and a durable business isn’t automatically a good token investment. The strongest fundamental opportunities may come from protocols that can preserve earnings through downturns and convert those earnings into tangible value for tokenholders.

.@Raydium’s revenue more than tripled in Q3, but memecoins are no longer its biggest source of swap fees. According to Blockworks Research, Raydium generated $8M in net revenue during Q3 2026, with tokenized assets becoming its largest source of swap fees. The numbers: + Q3 net revenue: $8M + Revenue growth: More than 3x + Annualized net revenue: $32M + Largest swap fee category: Tokenized assets For a DEX historically tied to Solana’s speculative trading cycles, the changing revenue mix is arguably more important than the growth itself. Consider $1B in trading volume. + At 5 bps: $500K in gross swap fees + At 20 bps: $2M in gross swap fees That’s four times the fee generation without any increase in volume. Of course, gross swap fees aren’t Raydium’s net revenue. LP allocations and protocol take rates determine how much the business retains. But this illustrates how a different trading mix can improve fee generation without requiring equivalent growth in volume. Tokenized assets also introduce different liquidity requirements, market-making costs and trading behavior, so their long-term profitability remains unproven. Still, Raydium is developing another meaningful revenue source beyond memecoin trading. If that continues across market cycles, it could change how investors assess Raydium’s business and the economics behind $RAY.

30-day inflows can tell one story. The next 7 days can tell another. Across 242 lending pools tracked by @DeFiStar, $348.6M flowed into growing markets, while $305.3M left declining ones. After adjusting for interest, net deposits came to just $29.8M. RLUSD led stablecoin inflows at $199.8M, while @SkyEcosystem recorded $85.8M in net outflows. This is why gross deposits alone can be misleading. Capital moving into one market doesn’t mean DeFi is attracting the same amount of new capital overall. The real signal is where liquidity settles, whether borrowers absorb it, and whether the resulting yields can sustain that demand.

bStocks has won the first round: distribution. Keeping that lead after the incentives fade is a different problem. That’s where I’d judge the tokenized-stock race. As of Oct 22, RWA.xyz puts tokenized-stock distributed value at $3.58B: • @bstocksfinance: $1.30B • @OndoFinance: $908M • @xStocksFi: $628M Together, they account for roughly 79% of the market. bStocks reached $1B in 125 days. Binance Research says the total market grew 119% since launch; excluding bStocks, growth was still 45%. It attributes over 60% of the increase to bStocks. That points to market expansion alongside share gains. Distribution explains the speed, but “it’s just campaigns” is too simple. Binance Research reports 56% of bStocks trading happened outside US market hours, weekly volume averaged 1.65x AUM, and margin collateral rose 6.5x since June 30. That’s activity, not proof of retention. The wider market has a counter-signal. RWA.xyz’s latest 30 day snapshot shows distributed value up 13.4%, but transfer volume fell 74.2% and active addresses fell 49.6%; holders rose 14.8%. That doesn’t prove demand is fading. It does show AUM and activity need separate scorecards. Wallet holders aren’t the same as users. One person can hold multiple addresses, and a large custodial wallet can represent many customers. Holder counts alone don’t tell us retention, average balance or whether users come back after rewards stop. RWA.xyz shows 1,472 for xStocks under “RWA count”; Kraken’s FAQ lists 131 assets. Those measures aren’t directly comparable, so I wouldn’t present 1,472 as 1,472 distinct stock choices. Asset counts need the same definition. Same ticker doesn’t mean same product: backing, legal claims, redemption and country access vary. My take: bStocks leads distribution, Ondo has breadth and mint/redemption, and xStocks brings multichain reach. The durable winner keeps repeat users, tight liquidity and net inflows after incentives fade.

NFT minting gets a different twist with @chimpchampsfun on Robinhood Chain Testnet. Instead of minting straight away, you explore a pixel island, dig up treasure and discover chests that can be minted as NFTs. Each chest gives you a choice: keep the NFT or burn it for BANANA rewards. 🥉 Bronze: 1,000 BANANA 🥈 Silver: 2,000 BANANA 🥇 Gold: 4,000 BANANA Rewards decrease with every new epoch, so timing matters. This is an NFT collection, not a PONS token launch. The experience is still on testnet, and the assets have no real-world value. Ready to dig? 👇