Responsible team: Reclaim / MarketWhale · Reviewed:
Solana Rent Estimates in an 85-Wallet Archived Leaderboard Cohort
On 9 September 2026, Reclaim attempted read-only previews for every unique address in a saved 85-wallet leaderboard cohort. 75 previews completed; 10 failed. Among completed previews, the median estimated net was 0.050522828 SOL before network fees. These are estimates of account rent, not completed recoveries.
Results: 75 completed previews out of 85 attempts
72 wallets were provisionally eligible: 96% of completed previews, or 84.7% of the full attempted cohort. Three completed previews had no candidate accounts. Ten wallets have unknown results; they are excluded from amount statistics, not counted as zero.
| Statistic | Gross rent (SOL) | Estimated net (SOL) |
|---|---|---|
| Median | 0.056136474 | 0.050522828 |
| 75th percentile | 0.098660640 | 0.088794577 |
| 90th percentile | 0.281935680 | 0.253742113 |
| Total | 8.827311159 | 7.944580111 |
For the 72 provisionally eligible wallets alone, the median gross was 0.0571710285 SOL and median net was 0.0514539265 SOL. The downloads contain both populations, with exact lamport values.
| Amount (SOL) | Gross: wallets | Net: wallets |
|---|---|---|
| 0 | 3 | 3 |
| >0 to <0.01 | 11 | 11 |
| 0.01 to <0.1 | 43 | 46 |
| 0.1 to <1 | 16 | 14 |
| >=1 | 2 | 1 |
The totals describe only the completed previews. They are not estimates for the full cohort or for Solana wallets generally.
Cohort, collection window and failed requests
We used all 85 addresses from the project’s retained leaderboard snapshot, whose embedded run date is 25 January 2026. There were 85 rows, 85 unique addresses and zero duplicates. Addresses were sorted lexicographically before collection. No wallet was selected or removed based on its current result.
The original article attributed its source to the KOLScan leaderboard, but the retained snapshot does not establish the original selection rules or collection window. We therefore describe this as an archived leaderboard convenience cohort. Current leaderboard membership, activity and ownership were not verified. Names and social profiles were not used in this analysis or included in the downloads.
The new scan ran on Solana mainnet-beta from to on 9 September 2026. It used the version 1.0 preview operation on a local Reclaim server backed by the configured Helius RPC. Reads used confirmed commitment; successful observations span slots 445645271–445645788. Reads are not an atomic snapshot across wallets, or necessarily across calls for one wallet.
Collection made 103 HTTP requests for 85 wallets, including one retry for each of 18 wallets. Requests were spaced by 1.1 seconds. Transport failures, HTTP 429 and server errors could be retried once; account/response-limit errors were not retried. All attempts are retained in the observation JSON.
- 7 wallets: WALLET_TOO_LARGE. The preview enforces a 5,000-account limit and a 6 MiB response budget. Their amounts are unknown; no partial account list was used.
- 3 wallets: RPC_UNAVAILABLE after retry. These did not yield usable previews. The error does not establish a zero balance or absence of empty accounts.
Excluding oversized wallets can bias the completed sample toward wallets with fewer accounts and potentially lower rent amounts. We did not impute failed results, extrapolate totals or calculate population confidence intervals.
What the estimates include
Both the legacy SPL Token program and Token-2022 were queried with getTokenAccountsByOwner. The preview selects accounts with a zero parsed token amount, preserves one empty wrapped-SOL account when present, and checks reported close authority. Gross is the sum of candidate account lamports, rather than a fixed rent amount multiplied by the account count.
Net subtracts the preview’s 6% service and 4% referral portions; both go to the service in this no-referral preview. Each portion is rounded down to integer lamports. At a wallet balance of 0.0001 SOL or less, the model adds 100,000 lamports per candidate and can deduct rent for a service account. The result is floored at zero. This run recorded no subsidy or service-account-rent deductions.
Network fees were unknown in every preview and are excluded. No transaction was built, simulated, signed or submitted for this study. “Eligible” is provisional: extension rules, changed account state, signing authority, transaction validation and batch rounding can affect execution. 68 completed previews carried the Token-2022 validation warning.
Solana’s Close Account documentation explains the authority and balance requirements; its getTokenAccountsByOwner reference describes the account lookup. The study’s stricter candidate filters and deductions belong to Reclaim. See the preview contract and fee reference.
Download and reproduce the calculations
Amounts use exact integer lamport strings: 1 SOL = 1,000,000,000 lamports. Median means the middle observation, or the arithmetic mean of the two middle observations for even samples. P75 and P90 use the nearest-rank convention: sort ascending and select one-based rank ceil(p × N). Half-lamport medians are preserved in the downloads.
- Aggregate results: CSV · JSON, including distributions and failures
- Frozen observations with direct identifiers removed: CSV · JSON with provenance and request attempts
- Dependency-free calculation script · Read-only collection script · Method and data dictionary
Download observations.json and analyze.mjs into the same directory. With Node.js 20 or later, run:
node analyze.mjs observations.json reproducedThis writes reproduced.json and reproduced.csv from the frozen observations without network access. Public rows omit wallet addresses, wallet balances, names and social handles; this reduces direct identification but does not guarantee anonymity. The source and raw-run SHA-256 hashes are recorded for provenance. The public files reproduce the aggregate arithmetic; they cannot independently verify which wallet produced a row. A new live scan will observe different state.
How to cite this result
Reclaim / MarketWhale. (2026, September 9). Solana Rent Estimates in an 85-Wallet Archived Leaderboard Cohort. Study version 1.0: 85 attempted wallets, 75 completed previews. Permanent results link.
This is a descriptive study of a fixed historical cohort. Use a current public-address preview for your own wallet, and review its warnings before considering a separate signing flow.