Data as of
A 3–5 year thesis, scored in public
The claim: NEAR's value engine has moved from blockspace to Intents, and captured Intents revenue can grow faster than token dilution. We split that into six plain claims, each with a frozen pass/fail rule, and re-grade them on public data every day. Failing claims stay on the page.
+207% YoY (90d window)
+202% vs prior 90d
Of Intents swap volume, trailing 90d
Trailing 90d vs a year earlier
Needs ≥ 1.00× to offset issuance
Of total supply minted per year
Earlier points are backfilled with the frozen rules, so descriptive only. Out-of-sample points so far: 1.
Live pulse
Updated 00:47:07 UTC
Snapshot model uses a 7-day average price of $3.61. Live values are shown for context and do not change any test result until the snapshot is rebuilt.
Falsification scorecard
Each test has a numeric threshold in one constant in the pipeline (scripts/fetch_public.py). A unit test pins its hash, so the rules cannot be tuned without failing CI. Every rebuild is graded with the same rules, so a failing test can flip, and so can a passing one.
Disclosure: the thresholds were chosen after a first look at the 2026-11-12 snapshot, so results on that snapshot are in-sample. They were frozen on 2026-11-13; only results from later snapshots are out-of-sample. This snapshot: out-of-sample · sha256:b9548c7a9ca0…
T1 · Demand
Holds205% of the pass linefail < 1.00×
Intents volume, trailing 90d vs same 90d one year earlier
pass ≥ 1.50x · fail < 1.00x · otherwise inconclusive
T2 · Monetisation
Holds275% of the pass linefail < 0.90×
NEAR holders revenue (buyback-directed), trailing 90d vs prior 90d
pass ≥ 1.10x · fail < 0.90x · otherwise inconclusive
T3 · Adoption
Fails28% of the pass linefail < 0.75×
Avg daily active accounts, trailing 90d vs same 90d one year earlier
pass ≥ 0.90x · fail < 0.75x · otherwise inconclusive
T4 · Activity quality
Holds198% of the pass linefail < 0.80×
Gas fees per active account, trailing 90d vs one year earlier
pass ≥ 1.00x · fail < 0.80x · otherwise inconclusive
T5 · Value accrual
Fails19% of the pass linefail < 0.25×
(Annualised Intents holders revenue + gas burn) / annual gross issuance value
pass ≥ 1.00x · fail < 0.25x · otherwise inconclusive
T6 · Market signal
Fails32% of the pass lineno inconclusive band
Max weekly lead correlation (1–4w) vs Bonferroni critical r
pass if best leading r > critical r (0.2513) over 99 weeks · otherwise fail
Track record
Every column re-grades all six claims as of that week's data, using exactly the thresholds at thresholds-v1. From May 15, 2026 to Nov 13, 2026 the score moved from 1/6 to 3/6.
Read this as descriptive. These weeks are backfilled with rules chosen in November 2026, after the data was seen, so they show how the claims would have read, not a forecast that came true. Columns after Nov 12, 2026 are out-of-sample: 1 so far. The weekly grid is anchored to the latest snapshot, so the count grows by one each week; every committed snapshot stays in git history.
| Claim | May 15, 2026Nov 13, 2026 | Flips |
|---|---|---|
| T1 More money is swapping through NEAR Intents than a year ago | 0 | |
| T2 NEAR is capturing more of that flow as revenue | 2 | |
| T3 People still use the NEAR base layer | 0 | |
| T4 Each remaining user pays more in fees | 2 | |
| T5 Buybacks and burn outweigh new token issuance | 0 | |
| T6 Intents activity moves before the NEAR price | 0 | |
| Claims holding | 3/6 |
Pillar 1 · Demand & monetisation · T1, T2
NEAR Intents lets users and agents swap assets across chains, with settlement coordinated on NEAR. Weekly swap volume is the leading indicator of demand. Captured revenue is the share NEAR sweeps to its revenue wallets. Since 23 Feb 2026, that revenue has funded open-market NEAR buybacks.
Weekly Intents swap volume vs captured revenue
Complete ISO weeks · USD · DefiLlama
+207% YoY · +87% QoQ
Protocol + frontend affiliate fees
+202% vs prior 90d
4.23 bps of volume reaches holders
Pillar 2 · Base-layer adoption · T3, T4
A thesis that ignored this chart would not survive review. Daily active accounts and transactions are down sharply year over year. Gas fees per active account nearly doubled over the same period, which is consistent with incentive and bot traffic leaving and higher-value usage staying. The data supports that reading but cannot prove it.
Average daily active accounts and transactions, weekly
Distinct signer accounts, not humans · NearBlocks
-75% vs 288.3K a year ago
-84% vs 4.6M a year ago
+98% YoY · daily average
Pillar 3 · Value accrual · T5
NEAR mints new supply every year. Holders are compensated by two flows: 70% of gas fees are burned, and captured Intents revenue is spent on buybacks. Coverage compares those flows with the dollar value of new issuance. Below 1.0×, a non-staking holder is being diluted.
Monthly holder value vs value of new issuance
Issuance = Δ total supply + burned NEAR, at monthly average price · complete months only
2.51% of supply · $115.9M at today's price
Buybacks $21.2M + burn $1.1M
Holder value must grow 5.2× to offset dilution
Market cap $4.5B
Market signal · T6
The first draft of this thesis listed this question as untested because it had no price history. We added a price series and fixed the method in advance: weekly log changes, lags of −4 to +4 weeks, and a Bonferroni-adjusted threshold across the four leading lags. No lag clears the bar.
Correlation by lag (positive = Intents leads price)
Dashed lines: critical |r| = 0.251 over 99 weeks
n=95 weekly pairs
Scenario model · re-run it yourself
Year 0 is the measured baseline from the snapshot. Coverage each year is (Intents volume × take rate + gas burn) ÷ (supply × issuance rate × price). Price is held constant on purpose: this is a sensitivity analysis of the token's economics, not a price target.
Measured trailing YoY: 207%
Measured 90d: 4.25 bps
Measured 12m: 2.5%
Gas burn is small either way
Holder value overtakes dilution in year 5.
| Year | Intents volume | Buybacks + burn | Issuance value | Coverage |
|---|---|---|---|---|
| Y0 | $50.2B | $22.4M | $118.8M | 0.19× |
| Y1 | $75.3B | $33.2M | $121.8M | 0.27× |
| Y2 | $113B | $49.3M | $124.8M | 0.40× |
| Y3 | $169.5B | $73.5M | $128M | 0.57× |
| Y4 | $254.2B | $109.6M | $131.2M | 0.84× |
| Y5 | $381.3B | $163.8M | $134.5M | 1.22× |
Deep dive · raw chain data
Aggregators can be wrong. The second pipeline queries raw transactions and receipts from bigquery-public-data.crypto_near_mainnet_us using versioned SQL, so anyone can audit signer counts and contract usage without relying on NearBlocks.
Aug 15, 2026 – Nov 12, 2026
Weeks where signers and txns equal NearBlocks
Raw-chain counts reproduce the NearBlocks figures behind T3 to the unit in every comparable week, so the base-layer numbers above are not an aggregator artifact. This confirms NearBlocks; it is not a second, differently-defined measurement.
| Contract | Receipts | Originating txns | Signer-days |
|---|---|---|---|
| token.sweat | 17.8M | 14M | 4.3K |
| v2.ref-finance.near | 11.7M | 10M | 81.7K |
| intents.near | 11.5M | 8.8M | 33.2K |
| cooldogsclub.near | 10.8M | 503.5K | 51 |
| game.hot.tg | 9.5M | 9.4M | 5.2M |
| v2.jars.sweat | 8.6M | 6.7M | 112 |
| wrap.near | 8.5M | 5.2M | 151K |
| dclv2.ref-labs.near | 7.1M | 4.4M | 99.2K |
| claim.sweat | 6.6M | 4.8M | 91 |
| 17208628f84f5d6ad33f0da3bbbeb27ffcb398eac501a31bd6ad2011e36133a1 | 5.7M | 3.3M | 50.2K |
Imported from BigQuery console exports of sql/daily_usage.sql and sql/contract_usage.sql. Raw exports are committed under data/bigquery/.
Methodology & reproducibility
Window: Nov 14, 2024 – Nov 13, 2026 (730 days). Snapshot generated Sat, 14 Nov 2026 04:26:27 GMT. No API keys, no hand-entered numbers, no sample data.
git clone https://github.com/zecvilan/falsifiable-near && cd falsifiable-near
make data # python3 scripts/fetch_public.py (stdlib only)
make test # unit tests for the statistics + aggregation
make dev # dashboard at http://localhost:3000| Endpoint | Feeds |
|---|---|
| NearBlocks charts API | Transactions, active/new accounts, price, total supply |
| DefiLlama · NEAR chain fees | Daily gas fees (USD) |
| DefiLlama · NEAR chain revenue | Burned share of gas fees (70%) |
| DefiLlama · NEAR Intents volume | Daily swap input volume (USD) |
| DefiLlama · NEAR Intents fees | Gross protocol + affiliate fees |
| DefiLlama · NEAR Intents holders revenue | Captured revenue directed to buybacks |