Why 3 Pools Left Trackers

What Happens When A Pool Disappears From Yield Tracking

In the third week of August 2026, three Curve pools stopped appearing on yield aggregators: frxUSD/msUSD, PYUSD/crvUSD, and pmUSD/crvUSD. The first lost 164,000 gauge weight votes in a single week. The second saw liquidity incentives expire without renewal. The third experienced a bank run after questions emerged about the redeemability of its collateral backing.

If you held positions in any of these pools, you discovered the problem the same way most liquidity providers do: you checked your dashboard, saw zero emissions, and realized your yield had stopped accruing. By that point, you faced exit slippage between 1% and 3% on six-figure positions, depending on which pool you were in and how fast TVL had already drained.

The question readers ask is not why pools fail in general terms. The question is how to recognize the pattern two weeks before a pool becomes uneconomic, while exit liquidity still exists at reasonable cost.

The Three Mechanisms That Remove Pools From Tracking

Curve pools disappear from yield trackers for three distinct reasons, and the warning signals differ for each.

Gauge Vote Collapse

Gauge weights determine what percentage of CRV emissions flow to each liquidity pool. veCRV holders and Convex vote on these allocations weekly. When a pool loses votes, its emissions drop. When emissions drop below the fee-based yield floor, trackers stop listing it as an income opportunity.

The frxUSD/msUSD pool shed 164,000 gauge weight in week 32 of 2026, the largest single-week loss recorded that month. The msETH/WETH pool lost 85,200 in the same period. Both pools had been unwinding for three weeks before trackers delisted them.

The pattern: vote-locked CRV holders redirect emissions toward newer stablecoin pairs or pools with higher base volume. Older pools with stagnant TVL or redundant pair composition lose votes incrementally, then catastrophically once a threshold is crossed.

The early signal you can track: a pool losing more than 20,000 gauge weight per week for two consecutive weeks will likely drop below economic viability within a month. That gives you exit runway before slippage compounds.

Liquidity Incentive Expiration

Some pools rely on external incentives from protocols that issue their own tokens to attract TVL. When PYUSD/crvUSD incentives expired in late August 2026, the pool lost $21.8 million in a single week, the largest outflow of any tracked pool that period.

This was not a technical failure. The pool’s smart contracts functioned normally. The peg held. The issue was economic: without additional PYUSD rewards, the base yield from trading fees and CRV emissions fell to 2.1% APY, below what LPs could earn in competing stablecoin venues.

Incentive-dependent pools often appear at the top of yield leaderboards when rewards launch, then disappear within 60 days when the incentive budget is exhausted. If a pool’s APY is more than 80% derived from a single external token, and that token has no established secondary market or liquidity backstop, the pool’s yield duration matches the protocol’s treasury runway.

The early signal: declining incentive token price combined with stagnant or falling pool TVL. If incentive APY drops by half while TVL remains flat, it means LPs are not rotating in to capture cheaper rewards. That precedes full incentive withdrawal by 10 to 15 days.

Peg Instability and Collateral Doubt

The pmUSD/crvUSD pool lost peg stability and TVL after market participants questioned whether pmUSD’s backing collateral was redeemable at par. The result was a classic bank run: LPs exited faster than arbitrageurs could restore balance, the pool’s virtual price decoupled from its target, and trackers removed it to avoid listing a broken peg as a yield venue.

This failure mode is the fastest. Gauge vote collapse takes weeks. Incentive expiration is telegraphed by token price and treasury disclosures. Collateral doubt triggers exit within 48 hours.

Curve’s PegKeeper system is designed to stabilize crvUSD pairs by injecting liquidity when pools trade below peg. In August 2026, PegKeeper reserves climbed to $65 million, an increase of $26.5 million from the prior month, as the mechanism absorbed sell pressure across multiple crvUSD pools. But PegKeeper only stabilizes crvUSD itself. It does not defend the collateral quality of the asset paired against crvUSD.

When pmUSD’s collateral redeemability was questioned, PegKeeper could not prevent the run. The pool’s TVL evaporated, and LPs who exited late faced slippage north of 2%.

The early signal: widening spread between a pool’s traded price and the oracle price of its underlying assets. If that spread persists for more than six hours without arbitrage correction, it indicates either a liquidity crisis or collateral doubt. Both justify immediate exit.

Why This Matters More In Emerging Markets

In dollar-native economies, losing 2% to exit slippage is a cost you can model and accept. In peso, lira, or naira economies, that 2% is not denominated in the currency your cost basis is calculated in. You are converting stablecoin losses back into a depreciating local currency, and your real loss is the dollar loss plus the opportunity cost of holding an illiquid position while local inflation runs.

I have seen this play out in Turkey and Argentina. A Turkish LP holds a position in a Curve pool yielding 8% APY in USDT terms. Gauge votes shift. Yield drops to 3%. The LP delays exit because 3% still beats lira deposit rates. Two weeks later, the pool is delisted, TVL has drained, and exit slippage is 2.5%. The LP exits at a 2.5% loss in dollar terms, but the lira has depreciated another 4% in the same two-week window. The combined loss is 6.5% in purchasing-power terms, and the opportunity cost of reallocating that capital earlier is another three weeks of yield at a functioning venue.

This is why recognizing pool death two weeks early matters more when your income is denominated in stablecoins but your expenses are not. The exit cost is not just slippage. It is slippage multiplied by currency depreciation and compounded by foregone yield during the period you were immobilized.

The Warning Signals You Can Track Yourself

You do not need proprietary data to see a pool unwinding. The signals are public, updated weekly, and visible on Curve’s own dashboards and third-party trackers.

Track gauge weight changes weekly. Curve publishes vote results every Thursday. If a pool you hold loses more than 15% of its gauge weight in a single week, or loses weight for three consecutive weeks, you are watching managed decline. Exit within the next vote cycle.

Monitor TVL outflows relative to market TVL. If your pool loses 10% TVL while Curve’s aggregate TVL is flat or growing, your pool is losing competitive position. If it loses 20% in a week, you are in the early stage of a run. DefiLlama updates this data daily.

Watch for stale virtual price growth. Virtual price is the internal accounting measure of fees accrued per LP token. If a pool’s virtual price stops growing for more than 72 hours, it means trading volume has fallen to the point where fee income is near zero. That precedes formal delisting by one to two weeks.

Check PegKeeper reserve levels if you hold a crvUSD pair. If reserves are being deployed to your specific pool and the peg is still trading 20 basis points or more below target, it means sell pressure exceeds what the stabilization mechanism can absorb. That is a liquidity crisis in progress.

Track the incentive token price if your pool relies on external rewards. If the token falls 30% while the pool’s APY remains high, it means the protocol is issuing more tokens to maintain displayed yield. That is not sustainable. The pool will either lose TVL or lose incentives within 30 days.

What To Do When You See The Pattern

Exit before TVL falls below 50% of its 30-day average. Once a pool crosses that threshold, slippage becomes nonlinear. The difference in exit cost between a pool at 60% of average TVL and a pool at 40% is often more than the difference between 100% and 60%.

Do not wait for tracker delisting. By the time a pool disappears from DefiLlama or Curve Monitor, the exit window has closed for large positions. Trackers delist pools when they fall below minimum TVL thresholds, which means liquidity is already impaired. You want to exit while the pool is still tracked but showing decline.

Calculate your exit slippage in advance. Curve pools display estimated slippage for withdrawal amounts. If your position size would incur more than 0.5% slippage at current TVL, and TVL is falling, your real exit cost will be higher by the time you execute. Plan the exit in tranches or accept the cost now.

Move to pools with demonstrated gauge stability. The pools that gained the most gauge weight in week 32 of 2026 were reUSD/scrvUSD (+90,500), crvUSD/sfrxUSD (+43,900), and crvUSD/sUSDe (+37,300). These are not inherently safer, but they represent current veCRV holder preference. Pools that gain votes for multiple consecutive weeks are less likely to lose them abruptly.

Treat incentive-heavy yields as short-duration positions. If more than half your APY comes from a token that launched in the past 90 days, plan to exit before the incentive budget is exhausted. The real return calculation for these positions should include a maximum hold period of 60 days.

Fee Floors and What They Signal About Pool Longevity

In late August 2026, Michael Egorov, Curve’s founder, proposed doubling core crvUSD pool fees from 0.01% to 0.02%. The proposal reflects a structural issue: at current trading volumes, many pools do not generate enough fee income to justify LP capital allocation, even with moderate CRV emissions.

Curve pools charge between 0.01% and 0.04% per trade, with 50% of fees going to LPs and 50% to veCRV holders. At a 0.01% fee, a pool needs $100 million in weekly volume to generate $50,000 in annual LP fee income. On $10 million TVL, that is 0.5% APY from fees alone.

If your pool’s base fee APY is below 1%, and CRV emissions are the only reason the pool shows positive yield, you are holding a position that exists because of vote incentives, not because of organic trading demand. When those votes shift, the pool becomes uneconomic within one emissions cycle.

The crvUSD/WBTC pool added $2.9 million in TVL during the same week that frxUSD/msUSD lost gauge weight and PYUSD/crvUSD lost incentives. That TVL moved to a pool with higher fee-based yield and stable gauge support. This is not a liquidity crisis. It is capital reallocation toward pools with better income sustainability.

The fee floor matters because it defines the minimum trading volume required for a pool to survive without subsidies. Pools below that floor depend on governance votes or external incentives. Both can disappear faster than TVL can migrate.

When A Pool Disappearing Does Not Matter

If you hold less than $5,000 in a pool, and exit slippage is under 1%, the cost of monitoring gauge votes and TVL flows weekly exceeds the cost of simply exiting when yield drops. The income mechanics described here are relevant for positions large enough that 1% to 3% slippage is a material cost.

If you are providing liquidity as a short-term arbitrage or liquidity mining strategy with a planned hold period under 30 days, pool disappearance risk is not your primary concern. You are exiting before gauge vote cycles or incentive expirations become factors.

If you are in a pool with sustained gauge weight growth, stable TVL, and fee-based APY above 2%, the risk of sudden delisting is low. The pools that disappeared in August 2026 all showed at least two of the three warning signals, multiple weeks in advance.

What The Pattern Tells You About Curve’s Current State

Curve tracks 548 pools with an average supply APY of 16.8% as of late August 2026. That average is heavily skewed by new incentive-driven pools. The median pool yields closer to 4% to 6%, and a significant portion of that comes from CRV emissions subject to weekly vote reallocation.

The three pools that disappeared represent different failure modes, but the common thread is that none had economic sustainability independent of external support. Gauge votes, incentives, and collateral trust are all external variables that LPs do not control.

The pools gaining TVL and gauge weight in the same period are predominantly crvUSD pairs with established stablecoins, higher base volume, and fee structures that generate positive returns before emissions are counted. This is not a Curve-wide liquidity crisis. It is a culling of pools that should not have been economically viable in the first place.

The lesson is not that Curve pools are risky. The lesson is that not all Curve pools are the same, and the difference is visible in public data weeks before tracker delisting occurs. If you are in a pool that depends on votes, incentives, or collateral trust, and any of those three shows deterioration, you are holding a position with a countdown timer.

For LPs treating stablecoin yield farming as income rather than speculation, the cost of recognizing that countdown two weeks early is zero. The cost of ignoring it is 1% to 3% of your position size, multiplied by however many positions you hold in pools with the same structural dependency.

The Takeaway

The frxUSD/msUSD pool lost gauge votes for three weeks before delisting. PYUSD/crvUSD showed declining incentive token price and stagnant TVL before liquidity collapsed. pmUSD/crvUSD traded below peg for six hours before the bank run started. All three disappearances were visible in public data before exit costs spiked. The pattern repeats because the mechanisms are structural, not random. Track gauge weight weekly, monitor TVL against market TVL, and watch for virtual price stagnation. If your pool shows two of these signals, you have 10 to 15 days of exit liquidity at reasonable cost. After that, you are paying 2% to 3% to leave, and if you are converting back to a depreciating local currency, your real loss is higher. The projects building sustainable EM stablecoin income infrastructure are the ones that do not depend on vote incentives or external reward budgets to keep pools economically viable. Those are the venues that still have your exit liquidity when you need it.

FAQ

Why do Curve pools suddenly stop appearing on yield trackers?

Pools disappear when gauge votes collapse, external incentives expire, or peg instability triggers TVL flight. Trackers delist pools that fall below minimum TVL thresholds or show zero yield for multiple days. The frxUSD/msUSD pool lost 164,000 gauge weight in one week, PYUSD/crvUSD lost incentives, and pmUSD/crvUSD suffered a bank run after collateral doubt emerged. All three showed warning signals weeks before delisting.

What is the earliest signal that a Curve pool is becoming uneconomic?

Gauge weight loss of more than 20,000 per week for two consecutive weeks, or TVL decline exceeding 20% in one week while market TVL is flat, indicates managed decline or early-stage capital flight. Virtual price stagnation for more than 72 hours means trading volume has collapsed and fee-based yield is near zero. Any of these signals gives you 10 to 15 days to exit before slippage becomes punitive.

How much does it cost to exit a pool after it loses most of its TVL?

Exit slippage ranges from 1% to 3% for six-figure positions once TVL falls below 50% of its 30-day average. Small positions under $5,000 may see slippage under 1%, but large LPs face nonlinear costs as liquidity drains. The pmUSD/crvUSD bank run left late exits with slippage above 2%, and that does not include opportunity cost from holding an illiquid position while yield stops accruing.

Why does pool disappearance matter more for emerging market users?

LPs in peso, lira, or naira economies face dollar-denominated exit slippage plus local currency depreciation during the period their capital is immobilized. A 2.5% dollar loss combined with 4% currency depreciation in the same two weeks results in a 6.5% real loss in purchasing-power terms, plus foregone yield. Recognizing pool decline two weeks early allows reallocation to functioning venues before local currency depreciation compounds the cost.

Can you recover funds if a Curve pool is delisted from trackers?

Yes. Delisting from yield trackers does not mean smart contract failure. You can still withdraw liquidity directly from Curve’s interface. The issue is not access but cost. Once a pool is delisted, TVL has usually fallen to the point where exit slippage is 2% or higher for meaningful positions. Early recognition allows exit while liquidity still supports reasonable slippage.

Frequently Asked Questions

Why do Curve pools suddenly stop appearing on yield trackers?

Pools disappear when gauge votes collapse, external incentives expire, or peg instability triggers TVL flight. Trackers delist pools that fall below minimum TVL thresholds or show zero yield for multiple days. The frxUSD/msUSD pool lost 164,000 gauge weight in one week, PYUSD/crvUSD lost incentives, and pmUSD/crvUSD suffered a bank run after collateral doubt emerged. All three showed warning signals weeks before delisting.

What is the earliest signal that a Curve pool is becoming uneconomic?

Gauge weight loss of more than 20,000 per week for two consecutive weeks, or TVL decline exceeding 20% in one week while market TVL is flat, indicates managed decline or early-stage capital flight. Virtual price stagnation for more than 72 hours means trading volume has collapsed and fee-based yield is near zero. Any of these signals gives you 10 to 15 days to exit before slippage becomes punitive.

How much does it cost to exit a pool after it loses most of its TVL?

Exit slippage ranges from 1% to 3% for six-figure positions once TVL falls below 50% of its 30-day average. Small positions under $5,000 may see slippage under 1%, but large LPs face nonlinear costs as liquidity drains. The pmUSD/crvUSD bank run left late exits with slippage above 2%, and that does not include opportunity cost from holding an illiquid position while yield stops accruing.

Why does pool disappearance matter more for emerging market users?

LPs in peso, lira, or naira economies face dollar-denominated exit slippage plus local currency depreciation during the period their capital is immobilized. A 2.5% dollar loss combined with 4% currency depreciation in the same two weeks results in a 6.5% real loss in purchasing-power terms, plus foregone yield. Recognizing pool decline two weeks early allows reallocation to functioning venues before local currency depreciation compounds the cost.

Can you recover funds if a Curve pool is delisted from trackers?

Yes. Delisting from yield trackers does not mean smart contract failure. You can still withdraw liquidity directly from Curve’s interface. The issue is not access but cost. Once a pool is delisted, TVL has usually fallen to the point where exit slippage is 2% or higher for meaningful positions. Early recognition allows exit while liquidity still supports reasonable slippage.