title: [ARFC] SVR Expansion: Next Phase of Multi-Network Expansion
author: @TokenLogic
created: 2026-07-16
Summary
This analysis evaluates the historical performance of Aave’s existing SVR deployments, identifies the most promising candidates for future SVR integrations, and highlights opportunities to improve the performance of existing deployments.
Since launching on Ethereum in 2025 and subsequently expanding to Base and Arbitrum, SVR has processed approximately $877.6M of liquidation volume, generating more than $21.3M in gross revenue, of which approximately $13.9M has accrued directly to the Aave DAO. Historical results demonstrate that recapture rates are influenced by differences in the underlying MEV supply chain, the auction architecture, and the distribution of liquidation sizes. As a result, Base and Arbitrum have achieved recapture rates of approximately 89%, compared to approximately 52% on Ethereum Core V3.
Based on the analysis below, we recommend prioritizing Avalanche, Polygon, and BNB Chain V3 deployments for future SVR integrations. Assuming 80% SVR coverage and an 80% recapture rate, these deployments would have generated an additional $1.81M in revenue for the Aave DAO over the past year, representing an approximately 11% increase relative to the revenue generated by existing SVR deployments over the same period.
Overview of SVR
SVR is a mechanism that enables Aave to capture a portion of the value traditionally extracted by liquidators during oracle driven liquidations. Rather than allowing searchers to compete for liquidation opportunities immediately following a price update, SVR auctions the right to execute transactions alongside the associated oracle update, allowing a portion of the resulting value to be redirected to the protocol.
Aave shares SVR proceeds with Chainlink, with 65% allocated to Aave and 35% to Chainlink, and has progressively expanded SVR coverage across its deployments. The first implementation was introduced on Aave V3 Ethereum in March 2025, initially covering a limited set of assets before being expanded through additional deployment phases in June and August 2025.
In March 2026, SVR was expanded to Aave V3 on Base and Arbitrum through Chainlink’s Atlas based cross chain auction framework. Aave V4 launched on Ethereum mainnet on March 30, 2026 using standard price feeds, SVR was added roughly six weeks later, on May 15, 2026, when eligible assets across V4’s spoke oracles had their price sources swapped to SVR feeds.
| Deployment | Chain | Since |
|---|---|---|
| Aave V3 Core | Ethereum | March 2025 |
| Aave V3 Prime | Ethereum | May 2025 |
| Aave V3 | Base | March 2026 |
| Aave V3 | Arbitrum | March 2026 |
| Aave V4 | Ethereum | May 2026 |
As of today, Ethereum, Base, and Arbitrum represent the primary production deployments from which historical SVR performance can be evaluated. Markets on these chains provide the basis for assessing recapture rates and estimating the potential revenue impact of extending SVR to additional Aave instances.
Historical Performance of Existing SVR Deployments
To assess the potential impact of extending SVR to additional Aave instances, it is useful to evaluate the performance of existing deployments.
Table below summarizes the performance of current SVR enabled deployments. In addition to liquidation volume and protocol revenue, the table includes the liquidation bonus opportunity, defined as the portion of liquidation incentives available to external liquidators. Specifically, liquidation bonus opportunity is calculated as the liquidation bonus less the protocol liquidation fee, which is fixed at 10% of the liquidation bonus. This represents the maximum value that could theoretically be recaptured through SVR auctions.
| Deployment | Launch Date | Liquidation Volume | Liquidation Bonus Opportunity | SVR Revenue | Aave Revenue | Recapture Rate |
|---|---|---|---|---|---|---|
| Ethereum Core v3 | Apr 2025 | $867.8M | $40.56M | $20.90M | $13.58M | 51.5% |
| Arbitrum V3 | Mar 2026 | $4.66M | $0.23M | $0.21M | $0.13M | 89.2% |
| Base V3 | Mar 2026 | $3.56M | $0.18M | $0.16M | $0.11M | 89.3% |
| Ethereum Prime v3 | May 2025 | $0.77M | $0.04M | $0.03M | $0.02M | 66.7% |
Since the initial oracle swap on Ethereum in April 2025, SVR has facilitated more than $877.6M of liquidation volume across Aave markets, generating over $21.3M in gross revenue, of which approximately $13.9M has accrued directly to the Aave DAO. Ethereum Core V3 accounts for the overwhelming majority of historical activity, reflecting both its longer operating history and substantially larger liquidation volumes relative to newer deployments.
While Ethereum V3 has generated the largest amount of absolute revenue, more recent deployments on Base and Arbitrum have exhibited significantly higher recapture rates. Both markets have recaptured approximately 89% of the available liquidation bonus opportunity, compared to approximately 52% on Ethereum Core V3.
Several factors contribute to this difference. First, the auction architecture differs across chains. On Ethereum, SVR liquidations occur within the MEV supply chain through Flashbots’ MEV Share that includes additional intermediaries, most notably builders and validators, which capture a portion of the value generated by liquidation opportunities. Under the current MEV Share configuration, 10% of every winning searcher bid is allocated directly to the block builder, who uses it to bid for inclusion in the next block, before the remaining proceeds are distributed between Aave and Chainlink. As a result, a portion of the available liquidation value is inherently unavailable for recapture by the protocol.
In contrast, Base and Arbitrum do not have the same MEV supply chain as Ethereum, as block building, transaction ordering, and block proposal are performed by a centralized sequencer. Consequently, there are fewer intermediaries capturing value between the liquidator and the protocol, allowing a larger share of the liquidation opportunity to be recaptured through SVR. This structural difference is one of the primary factors contributing to the materially higher recapture rates observed across both deployments.
Second, liquidation characteristics also differ materially across deployments. Average liquidation sizes on Ethereum are approximately 25 to 30 times larger than those observed on Base and Arbitrum.
Larger liquidations are generally more difficult to execute profitably. Liquidators must unwind substantially larger collateral positions, which increases expected slippage, execution risk, inventory management costs, and the capital required to complete the liquidation. These higher execution costs reduce the amount of the liquidation bonus that can be competitively bid through SVR auctions.
At the same time, larger liquidations tend to attract fewer participants. Historical data shows that the number of distinct liquidators declines sharply as liquidation size increases, suggesting that fewer market participants possess the capital, inventory, and infrastructure required to execute these opportunities. Reduced competition may allow the remaining liquidators to retain a larger share of the available surplus, further limiting the portion of the liquidation bonus that is ultimately recaptured through SVR.
This relationship can be observed empirically across historical liquidations. As liquidation size increases, the proportion of the liquidation bonus recaptured through SVR generally declines, indicating that larger liquidation opportunities leave less surplus available for competitive auction bidding. The significantly smaller average liquidation sizes on Base and Arbitrum therefore provide another structural explanation for the higher recapture rates observed on those deployments.
Together, these observations establish the assumptions used throughout the remainder of this analysis. Historical recapture rates vary meaningfully across chains and are influenced by differences in the underlying block production pipeline, particularly which participants control block building and block proposal, as well as the characteristics of liquidation opportunities. These empirical observations provide the basis for estimating the revenue potential of extending SVR to additional Aave instances.
Prioritizing Future SVR Deployments
Unlike interest income or protocol fees, SVR only generates revenue when liquidations occur. Consequently, markets with large TVL but predominantly correlated collateral and debt positions may experience relatively few liquidations despite their size. Conversely, markets with more directional collateral and borrowing activity can generate meaningful liquidation opportunities despite having substantially smaller lending markets.
To identify the most promising candidates for future SVR deployments, we rank markets based on their historical liquidator bonus over the past 365 days, which serves as a proxy for the maximum value available for recapture through SVR auctions. This approach directly measures the value historically captured by external liquidators, making it a more suitable indicator of future SVR revenue potential.
Before estimating future revenue, it is also necessary to account for SVR coverage. Not every historical liquidation would have been induced by SVR oracle updates. Historical deployments indicate that approximately 76% of liquidation volume on Ethereum and 86% on Arbitrum occurred through SVR. To estimate future deployments, we therefore assume an 80% SVR coverage ratio, meaning that 80% of historical liquidation activity is considered attributable to SVR.
For the portion of liquidation activity covered by SVR, we assume an 80% recapture rate based on the empirical performance of existing deployments while remaining modestly conservative relative to the approximately 89% historical recapture rates observed on Base and Arbitrum. This assumption is supported by two factors.
First, future deployments are expected to utilize the same Atlas based auction architecture currently deployed on Base and Arbitrum, avoiding the builder and validator related value leakage present on Ethereum.
Second, historical liquidation sizes across the recommended deployments are relatively modest, suggesting stronger competition among liquidators and consequently higher expected recapture rates.
The resulting revenue estimates therefore assume both:
- 80% SVR coverage, representing the share of historical liquidation activity expected to be processed through SVR.
- 80% recapture rate, representing the share of the available liquidator bonus expected to be captured through SVR auctions.
Using these assumptions, the estimated annual revenue opportunity for top markets is shown below.
| Deployment | Active Loans | Historical Liquidation Volume (365D) | Historical Liquidator Bonus (365D) | SVR Attributable Bonus (80% Coverage) | Estimated Annual Aave Revenue (80% Recapture) |
|---|---|---|---|---|---|
| Avalanche V3 | $171M | $36.68M | $2.27M | $1.82M | $0.95M |
| Polygon V3 | $33.5M | $22.83M | $1.22M | $0.98M | $0.51M |
| BNB Chain V3 | $67M | $10.75M | $0.83M | $0.66M | $0.35M |
| Sonic V3 | $2.9M | $9.91M | $0.76M | $0.61M | $0.32M |
| Optimism V3 | $27.6M | $10.70M | $0.61M | $0.49M | $0.25M |
| Linea V3 | $7.9M | $4.17M | $0.21M | $0.17M | $0.09M |
| Plasma V3 | $871M | $3.43M | $0.16M | $0.13M | $0.07M |
The results further demonstrate that market size alone is not an effective indicator of SVR revenue potential. Plasma, for example, represents one of Aave’s largest lending markets with approximately $871M in active loans, yet generated only $0.16M in liquidator bonus over the past 365 days. This reflects the predominantly correlated nature of borrowing activity within the market, resulting in relatively few liquidation opportunities despite its large size.
Incremental Revenue Impact
To better quantify the opportunity, we estimate the incremental revenue that would have been generated if SVR had been enabled on the three highest-ranked deployments in the table above over the past 365 days.
Existing SVR deployments generated approximately $13.98M in revenue for the Aave DAO over the past 365 days. However, because Base and Arbitrum only enabled SVR in March 2026, this figure understates the revenue generating capacity of the current deployment footprint. To create a like for like comparison, we normalize the historical revenue by assuming that Base and Arbitrum had also been SVR enabled throughout the entire evaluation period. Under this assumption, historical Aave SVR revenue increases to approximately $16.22M.
Applying the same methodology to the three highest ranked deployments indicates that Avalanche, Polygon, and BNB Chain would have generated an additional $1.81M in revenue for the Aave DAO over the same period. Relative to the adjusted historical baseline, enabling SVR on these deployments would have increased cumulative Aave SVR revenue by approximately 11% over the past 365 days.
Improvement Areas
While the previous sections focus on expanding SVR to additional Aave deployments, the historical data also identifies opportunities to improve the performance of existing deployments. The most significant opportunity lies in increasing recapture rates for large liquidation events on Ethereum.
Historical liquidation data shows a clear inverse relationship between liquidation size and SVR recapture rate. As liquidation size increases, both the proportion of the liquidation bonus recaptured through SVR and the number of liquidators able to participate in these auctions decline materially, as illustrated in the “Recapture Rate and Distinct Liquidators by Liquidation Size Bucket” chart presented in the previous section.
This effect is particularly pronounced for liquidations exceeding $5M in size, where the average recapture rate falls to approximately 36.7%, substantially below the rates observed for smaller liquidations.
The impact of this decline becomes apparent when comparing the revenue generated by these liquidation size buckets. Although the $5M-$50M bucket represents the largest source of liquidation volume, totaling approximately $378M over the past year, it generated $4.2M in protocol revenue due to its relatively low recapture rate of 36.7%. By comparison, the $500K-$5M bucket processed a smaller liquidation volume of $318M, yet generated $5.6M in protocol revenue because it maintained a higher recapture rate of 57.7%.
This disparity suggests that large liquidation opportunities remain one of the largest untapped sources of incremental SVR revenue.
Several factors contribute to the lower recapture rates observed for larger liquidations. Large liquidations require substantially greater capital commitments, expose liquidators to higher execution risk and slippage, and often require more sophisticated inventory management and financing strategies. These higher execution costs reduce the amount of liquidation value that can be competitively bid through SVR auctions. In addition, fewer liquidators possess the capital and operational infrastructure necessary to compete for multi million dollar liquidation opportunities, resulting in reduced auction competition relative to smaller liquidations.
Despite these structural constraints, historical data indicates that there remains meaningful room for improvement. Increasing competition for large liquidation opportunities, could materially increase protocol revenue.
Illustratively, if the average recapture rate for liquidations between $5M-$50M increased from 36.7% to 55%, Aave would have generated approximately $2.09M in additional SVR revenue over the past year.
Similarly, the $500K-$5M liquidation bucket represents another meaningful opportunity for improvement. Its current recapture rate of 57.7%. Increasing the average recapture rate for this bucket to 70% would have generated approximately $1.2M in additional SVR revenue over the same period.
| Liquidation Bucket | Current Recapture Rate | Illustrative Target | Additional Aave Revenue (365D) |
|---|---|---|---|
| $5M-$50M | 36.7% | 55% | +$2.1M |
| $500K-$5M | 57.7% | 70% | +$1.2M |
These results suggest that improving auction competition for large liquidations may represent one of the highest-return opportunities for Ethereum development. While structural factors such as execution costs and slippage are likely to impose an upper bound on achievable recapture rates, even modest improvements could generate protocol revenue comparable to, or exceeding, the impact of several additional SVR deployments.
Recommendations
Based on the historical liquidation activity over the past year and the corresponding estimated revenue opportunity, Avalanche V3, Polygon V3, and BNB Chain V3 emerge as the strongest candidates for SVR deployment among the remaining Aave markets.
Although Sonic V3 ranks fourth by historical liquidator bonus, we do not recommend prioritizing it at this time. The market has experienced a substantial decline in active loans, and more than half of its historical liquidation activity during the past year occurred within a single day. Consequently, the observed liquidation bonus appears to be driven by an isolated event rather than sustained liquidation activity, reducing confidence that similar opportunities will persist.
The remaining markets, including Optimism V3, Linea V3, and Plasma V3, are not recommended at this time due to their comparatively limited revenue opportunities compared with the recommended deployments.
Deployments such as X Layer, whose market composition is expected to be dominated by uncorrelated collateral and debt positions, should continue to be monitored as they mature. Because these markets are structurally more likely to generate liquidation opportunities, they may present a compelling business case for SVR once sufficient lending activity has developed.
In parallel with expanding SVR to additional deployments, we recommend exploring mechanisms to improve auction competitiveness for large liquidations. Historical data indicates that liquidations exceeding $5M represent one of the largest remaining sources of unrealized SVR revenue, and even modest improvements in recapture rates could generate meaningful additional protocol revenue.
Specification
Based on the analysis presented in this report, we recommend enabling SVR for the following assets across the corresponding Aave deployments.
| Deployment | Assets |
|---|---|
| Avalanche V3 | BTC.b, AVAX, sAVAX, USDC, WETH.e, USDT, DAI.e, EURC, LINK.e, AUSD, AAVE.e, WBTC.e, sUSDe, FRAX, USDe |
| Polygon V3 | WBTC, USDT0, USDC, WETH, wstETH, POL, USDC.e, DAI, AAVE, EURS, LINK, MaticX |
| BNB Chain V3 | BNB, BTCB, USDT, USDC, ETH, wstETH, FDUSD, CAKE |
SVROracleSteward
As with all SVR activations on other instances of the Aave Protocol, this expansion will include SVROracleStewards.
To refresh, the SvrOracleSteward allows for the Aave Protocol Guardian to replace any of the newly introduced SVR feeds with the non-SVR feed currently used in production. In the very worst-case scenario where the new SVR is simply not functional or has a major issue, the Aave Protocol Guardian could immediately switch back to the standard price feeds currently used in production.
Next Steps
- Gather feedback from governance participants and relevant service providers.
- Incorporate any necessary revisions based on community discussion and technical feedback.
- If there is broad community support, proceed with an ARFC.
- If the ARFC is approved, proceed with an AIP to replace the existing price oracles with SVR enabled oracles on the approved deployments.
Disclaimer
TokenLogic is an active service provider to the Aave DAO, the beneficiary of stream 100086 and the KPI as outlined in this publication. The scope of this engagement is available via this forum proposal.
TokenLogic supports and maintains an independent delegate voting platform within the Aave community.
TokenLogic and associated entities have no undisclosed material conflicts of interest at the time of submission.
Next Steps
- Gather feedback from the community.
- If consensus is reached on this ARFC, escalate this proposal to the Snapshot stage.
- If Snapshot outcome is YAE, escalate this proposal to the AIP stage.
Copyright
Copyright and related rights waived via CC0.
