ChainFeeds on Aug. 13 published a research roundup covering five separate themes: the weakening of Upbit’s listing effect, Li Lin’s latest push through UMX, the continued migration of Korean crypto demand offshore, the structure behind stablecoin transfer volumes, and whether perpetual DEXs need a second answer beyond token incentives.
Upbit’s listing effect has weakened as listing frequency climbs
According to Four Pillars, Upbit added 58 Korean won-denominated trading markets in the first seven months of 2026. Of those, 42 were added in the first half of the year, or roughly seven per month on average. The pace picked up after that. Upbit added 10 markets in July and another six in the first week of August. Between Aug. 4 and Aug. 7 alone, it listed QUID, HOME, GRVT, CAP, KMNO, and BSB. On Aug. 10, the exchange also announced BTC and USDT markets for six more tokens in a single notice.
The report links that acceleration to a sharp decline in trading activity. Upbit’s average daily trading volume fell from $1.9 billion in February to $590 million in July, then slipped further to $460 million so far in August. That puts the year-to-date drop at about 70%, and the decline from a year earlier at about 85%. Four Pillars said Korean retail capital has been rotating into domestic equities, leaving new listings as one of the few levers an exchange can still control directly to try to revive volume.
Up to June 14, the median first-day gain for newly listed tokens reached 42%, while the median first-day trading value was about KRW 142 billion. Ten tokens posted triple-digit first-day gains during that period. ELSA rose 144% on its first day in January, EDGE jumped 230% in March, and SLX climbed 198% on June 1. SLX, however, was the last listing to post that kind of outsized debut.
For the 14 tokens listed from June 15 onward, median first-day gains dropped to 10%, and median first-day turnover fell to KRW 54 billion. HOME was the best performer at about 90%, but eight of those 14 tokens closed their first day below the opening price. Even if the sample is limited to genuinely first-time listings, the median first-day gain still fell from 42% to 10%.
Four Pillars said the problem does not appear to be seller behavior. For tokens that rose more than 40% after listing, the day-seven close usually sat 20% to 65% below the first-day high, with a median drawdown of about 45%. That pattern was broadly similar across both periods. In other words, early holders were already using Upbit listings as an opportunity to sell before June; what changed after mid-June was weaker buyer demand.
The report’s most direct explanation is listing fatigue. The pool of capital chasing Korean exchange listing trades is limited. When six tokens go live within four days, both capital and attention get split across too many names for any one listing to become a market-wide event.
That matters because Upbit has long been one of the toughest major crypto exchanges to get listed on. The difficulty of securing a KRW market created what the report described as a certification effect. Korean retail traders were not just buying a token. They were buying Upbit’s screening signal, and the expectation that other traders would do the same. If that screening mechanism approves 58 projects in seven months, its selectivity no longer carries the same weight. Four Pillars said Upbit is now eroding the two factors that once powered listing rallies: the credibility and scarcity of the listing signal itself, and the concentration of capital around a single new token. For an exchange whose volume is down 85% year on year, the most dependable way it had to stimulate trading is no longer working as it once did.
From Avenir to UMX, Li Lin’s latest thesis centers on unified capital and risk management
Foresight News said Li Lin founded Avenir Group in Hong Kong in 2023, taking a name from French that means “a better future.” The report frames the move as more than a change in title. After running an exchange directly, Li shifted into the role of a multi-strategy family office allocator. That shift changed the questions in front of him.
Instead of focusing on user growth, trading volume, product lines, and liquidity, the piece said the allocator’s perspective highlights another set of issues: where capital sits idle, where assets are fragmented, why accounts cannot interoperate, and why risk cannot be managed in one framework. Those issues are harder to see from the vantage point of an exchange operator focused on its own venue, but much harder to ignore from the perspective of capital allocation.
Avenir’s investments over the past few years sketch out that line of thinking. Foresight News said the group accumulated more than a 10% stake in UP Fintech, the parent company of Tiger Brokers, participated as a core investor in the $300 million equity financing of Hong Kong-licensed platform OSL, invested in institutional order-routing company CoinRoutes and options infrastructure provider SignalPlus, led the investment in AI-native quant platform Inference Research, and signed a memorandum of understanding on multi-asset infrastructure with Tiger Brokers and AMINA Bank at Consensus Hong Kong.
The report places that activity in a broader four-year shift. During the same period, crypto finance and traditional finance have moved from observing each other to becoming more deeply embedded with one another. Spot Bitcoin ETFs won approval in the U.S., bringing digital assets formally into traditional allocation frameworks. Stablecoins moved from being gray-zone funding rails to channels accepted by legislation in major economies. RWA and stock tokenization began pulling treasuries and equities into crypto-native conversations.
As the boundaries between brokerages, custody, clearing, market makers, and trading venues are redrawn, a slice of professional users is now operating in both markets at once. The report said these users watch on-chain liquidity, but also Federal Reserve rates, tech earnings, ETF flows, and the dollar cycle. Their problem is no longer where to buy an asset. It is whether different assets can sit inside the same strategy, treasury, and risk framework.
Foresight News argued that the previous generation of platforms competed on access, liquidity, and asset coverage. The next round may center on accounts, capital, and risk management. The report treats that as a structural signal rather than a passing narrative, saying the era of crypto as a sealed-off system is fading and a new zone is forming where the two markets meet. In its telling, that new zone still lacks a proper port.
The article then moves to UMX’s proposed user experience. Under the conventional path, a user holding USDT and BTC in a crypto account who wants to buy U.S. stocks would need to withdraw stablecoins, convert into fiat, wait for banking settlement, and then fund a securities account. Each step adds cost and delay.
In UMX’s unified account, the report said USDT can be converted into U.S. dollars in one click and transferred into a securities account to create immediate buying power. A user who does not want to sell BTC can instead borrow against existing holdings, transfer borrowed USD, and keep the crypto position intact.
What users buy, according to the report, is not a CFD or tokenized proxy but real U.S. equities with full shareholder rights. The system supports 12 order types, eight options combination strategies, fractional shares from $1, and pre-market and after-hours trading.
The return path for capital is where the design becomes more notable. The report said NVIDIA shares held by a user can be turned into corresponding stock tokens through a stock-to-token conversion, counted into a unified margin pool at the applicable haircut, and then used to support crypto derivatives and leveraged trading. That lets a user keep U.S. equity exposure while deploying the tokenized representation as collateral, with the option to convert back into the underlying stock. Yield-bearing products such as cash management and term products can also continue accruing yield while serving as margin. In the account structure described by Foresight News, capital is no longer meant to sit idle.
Korean crypto demand keeps moving offshore
In a long English-language thread, Tiger Research said the global digital asset market is evolving into different regional forms. Institutional capital is entering what was once a retail spot market, and blockchain use cases are extending across the wider financial value chain. Spot Bitcoin and Ether ETFs have broadened the route in for institutions. Perpetual futures and options have grown quickly. Stablecoins are expanding beyond crypto trading into payments and remittances. Tokenized real-world assets are reshaping issuance and distribution.
Those shifts have not happened everywhere at the same speed. Tiger Research said South Korea still revolves around retail spot trading, with limited institutional participation. Derivatives and newer financial services that are scaling globally have yet to form a market of meaningful size domestically. Global platforms have moved beyond spot into perpetuals and other products, while Korean activity remains concentrated in spot. The structural gap between Korea and the broader market is widening.
The report said falling market share does not mean Korean investors have lost interest in digital assets. It says the opposite. Products and services that are difficult to access within Korea are growing quickly abroad, and investors are turning to offshore exchanges and on-chain venues to meet that demand.
Working with Chainalysis, Tiger Research tracked crypto assets moving from Korea to overseas exchanges using on-chain data. It estimated that $530 billion moved through those channels between 2021 and 2026. Of that, about $120 billion left in 2025, and about $52 billion is projected for 2026. The report adds that because the figures capture only identifiable flows, the real total may be larger.
Even though the absolute size of outflows has come down, the net outflow ratio has risen. In Tiger Research’s reading, that means capital is leaving faster than the local market is shrinking. The group estimates Korean investors generated about $3.5 billion in fee revenue for overseas exchanges in 2025 and another $900 million in the first half of 2026. Trading demand sourced from Korea is being converted into revenue outside Korea.
The same dynamic is showing up in decentralized derivatives. Based on on-chain analysis from January 2024 through July 2026, wallets controlled by Korean investors deposited about $1.64 billion into Hyperliquid, Lighter, and Variational. Those venues focus on perpetuals and active leveraged trading, giving Korean users access to derivatives products that are difficult to obtain domestically.
In July 2026 alone, around 1,200 Korean investor wallets generated $4.97 billion in notional trading volume on Hyperliquid. Tiger Research noted that leverage inflates notional figures, but a monthly volume close to $5 billion still points to meaningful participation by Korean users in on-chain derivatives trading.
The set of products traded on Hyperliquid also extends beyond crypto assets. From January through July 2026, products tied to SK Hynix, Samsung Electronics, and crude oil ranked among the instruments with the highest trading volumes for Korean investor wallets. The report said high leverage and the ability to trade outside regular market hours may be important drivers. In practical terms, Korean investors want to trade traditional assets in ways the domestic market does not offer, and that demand is heading on-chain.
Stablecoin transfer volume is huge, but the underlying demand is mixed
Coin Metrics said stablecoins have processed $41.7 trillion in adjusted on-chain transfer volume so far in 2026. Supply growth has slowed recently, but each dollar of stablecoin supply is turning over more frequently than in previous years. The report said that points to a broader set of uses, from exchange liquidity management and DeFi collateral transfers to consumer payments and B2B transactions.
Since 2025, stablecoin settlement volumes have also started to decouple from crypto exchange trading volumes. Adjusted stablecoin transfer volume this year at one point topped $250 billion in a single day, while exchange trading volume had fallen to roughly $18 billion per day.
USDC stands out on turnover. So far in 2026, USDC’s annualized velocity based on adjusted supply has reached 741x, about 10 times USDT’s 74x, even though USDT’s market capitalization is more than $100 billion larger than USDC’s. Relative to its circulating base, USDC is moving on-chain far more frequently.
Circle’s USDC overtook Tether’s USDT in adjusted transfer volume in 2024, and the gap widened further this year. As of August 2026, USDC had processed about $32 trillion in transfers, equal to 77% of the stablecoin market total. USDT stood at about $8 trillion, or 19%. Circle’s second-quarter 2026 earnings showed USDC on-chain transaction volume rose 151% year over year to $14.8 trillion, even though supply growth was much slower. Reserve income still accounted for about 95% of Circle’s revenue.
Coin Metrics said a closer look at USDC activity shows that not all of that volume comes from payments. On Base, more than 90% of USDC transfer volume came from just three contracts. DEX liquidity provision accounted for about 69% of total volume, while flash loan activity through Morpho made up about 23%. Everything else was about 8%. On one day in June alone, USDC flash loan volume on Base exceeded $500 billion.
USDC activity on Ethereum was even more concentrated in flash loans. Roughly 65% of transfer volume came from flash loans, nearly three times the share seen on Base. DEX liquidity provision contributed just 0.3%, centralized exchange-related flows 2%, and the remaining 33% fell into other categories. USDT on Ethereum showed a similar structure, but with more visible exchange-related movement: flash loans were 46%, DEX liquidity provision 0.3%, centralized exchange flows 9%, and other activity about 45%.
USDT on Tron looked very different. Flash loan activity was almost negligible, DEX liquidity provision was just 0.2%, and flows tied to deposits and withdrawals at centralized exchanges including Binance, OKX, and Bybit represented about 19%. The remaining roughly 80% could not be classified into those buckets. Coin Metrics said that remainder may include payments, remittances, and other uncategorized fund activity.
Perp DEXs may need a value loop that reaches traders more directly
Foresight Ventures argued that a perpetual DEX has to win traders, liquidity, and distribution at the same time during its cold start. Early platforms rarely have stable revenue, and they are not well positioned to fund large cash subsidies for long. Tokens offered a more efficient coordination tool. A platform could translate the contributions of traders, market makers, and early participants into future claims, using future value to buy present-day volume, liquidity, and attention.
dYdX is presented as one of the earliest complete models. In its initial token supply, 25% was allocated to trading rewards, 7.5% to historical users, and 7.5% to liquidity rewards. Trading rewards were tied to fee contribution and open interest. Liquidity rewards measured quote uptime, two-sided depth, bid-ask spread, and the number of covered markets. Hyperliquid, Aster, and Lighter later adjusted that framework in different ways.
The report said those platforms were never only about tokens, but tokens remained the core coordination mechanism in the cold-start phase. The advantage was not merely that users got rewards. It was that distinct forms of contribution could be translated into future rights, while much of the customer acquisition and liquidity cost was pushed back to the TGE. The mechanics vary, but the trade is largely the same: future claims and upfront subsidies are exchanged for current volume, liquidity, and market attention.
Points and airdrops change more than the decision to enter a platform. They also change the reason for trading on it. Under normal conditions, traders compare venues on liquidity, execution stability, fees, asset coverage, and risk management. Once points enter the equation, those standards still matter, but the weighting shifts. Users also calculate how many points a unit of trading cost can generate, whether a potential airdrop will cover fees and capital usage, and when the reward might materialize.
Foresight Ventures said that does not make points-driven participants any less professional. Many professional traders evaluate points rationally. The issue is that once points become a meaningful variable in platform selection, the volume a platform attracts no longer reflects product competitiveness alone. It also reflects how users price future rewards.
In an earlier retail survey on perpetual DEXs cited by the firm, 69% of respondents ranked points or airdrops as an important factor in choosing a venue. That was higher than reliability at 61%, liquidity and depth at 58%, and fees at 47%. In that framework, points and airdrops are no longer side benefits outside the product. They compete directly with reliability, liquidity, and fees as a core selection variable. Rewards can lower the threshold for joining a platform, but they can also lower the threshold for leaving it.
After TGE, incentive systems usually move from distributing tokens to supporting the token. Platforms often use fee revenue for buybacks, burns, staking yield, and holder discounts. Value starts flowing back, but the basis of distribution shifts from trading contribution to token ownership. A trader generates revenue through trading, then must buy or stake the platform token to qualify for a share of that revenue.
That creates a dual-market problem. Platforms end up maintaining both a trading market and a platform-token market. The two can reinforce each other, but they do not always align. Professional traders may contribute fees over a long period while refusing to take additional token price risk. In that setup, the platform gradually rewards not just trading activity but also capital committed to the token itself.
When revenue sharing, fee discounts, and product access all hinge on holding the token, a perpetual DEX is effectively running two businesses at once: the trading venue and the token economy around it. Foresight Ventures said that produces a basic tension. Is the platform’s top priority the trading product, or the value loop built around the token? The report’s answer is to separate the two logics clearly: the trading market is judged on execution quality, liquidity, and cost, while the token market depends on demand, scarcity, and price expectations.
