Every claim below was read from Robinhood Chain directly, not from a data aggregator. Where we used an aggregator we say so, and where we got it wrong last week we say that too.
The author holds 500,000 PONS as a leveraged long on Hyperliquid, blended entry 0.5503, and is selling into a published exit ladder. This memo argues for a price we benefit from. Read it as a marked book, weigh the method rather than the conclusion, and check the sources at the bottom yourself. Nothing here is investment advice.
Pons documents a buyback: 80% of protocol fees are used to buy PONS on the open market via an automated TWAP and send it to a burn address. Most tokens make a claim like that. Almost none of them can be checked.
This one can. Here is the last 24 hours, read from the chain:
One burn roughly every 85 seconds, all day, from a single address
(0x5795d2…c324). That is not a marketing claim, it is a TWAP running in public.
The daily shape confirms it is tracking revenue rather than running on a fixed schedule:
| Window | Burn events | PONS burned |
|---|---|---|
| Last 24h | 1,069 | 1,441,821 |
| Day −1 | 492 | 1,015,754 |
| Day −2 | 82 | 152,656 |
For four days we argued the opposite case. DefiLlama's Pons V2 adapter does not implement a holders-revenue field, and we read that missing field as evidence the buyback had stopped. It was an adapter gap, not a policy gap. DefiLlama's entire 30-day holders-revenue figure is beaten by a single day on-chain. We publish that because a record you can only read when it flatters us is not a record.
PONS earns like an incumbent and is priced like a startup. Both halves of that are measurable.
| Metric, 4 Sep | Pons | pump.fun | Ratio |
|---|---|---|---|
| DEX volume | $153,877,521 | $54,764,845 | 2.81× |
| Day-on-day volume | +20.0% | −27.2% | — |
| Market cap | $603M | $1,654M | 0.36× |
| Fully diluted value | $603M | $3,495M | 0.17× |
| Circulating share | 100% | 47.3% | — |
| Revenue to holders | ~80% | 74% | — |
PONS did 2.81× pump.fun's volume on a day pump.fun's fell 27%, and it is valued at 17% of pump.fun's fully diluted value. The circulating line is why FDV is the fair comparison: PONS has no unlock schedule, the 295.8m burned tokens are gone permanently, and its market cap is its FDV. pump.fun still has roughly 440 billion tokens to issue.
On revenue, pump.fun trades at 3.74× annualised on market cap and 7.91× on FDV. PONS trades at 1.18×.
Float is 704,187,579 and falling. Taking a deliberately harsh 4×, roughly half the peer multiple, as a discount for a seven-week-old protocol:
| If daily revenue… | Annualised | at 4× | at 8× |
|---|---|---|---|
| reverts to the 30-day average | $110M | $0.64 | $1.28 |
| holds at the 7-day average | $383M | $2.22 | $4.44 |
| holds at the current rate | $511M | $2.96 | $5.92 |
Our number is $2.20 to $3.00. Not because the market is stupid, but because the market is pricing durability, and durability is the one thing seven weeks of history cannot prove either way.
One figure frames the whole thing. At roughly $1.0m a day of buying against a $603m cap, PONS carries a buyback of about 60% of its market cap per year. At the current token rate the entire float retires in about sixteen months. That cannot persist as a yield. It resolves through the price, or it resolves by the revenue falling.
Chain reads were made against the Robinhood Chain mainnet RPC at block 55,044,278 on
5 September 2026. Burns are Transfer events from the PONS contract to
0x…dEaD. Dollar values of burns use an estimated volume-weighted price across a
window that ranged 0.628 to 0.856, so token counts are exact and dollar figures carry roughly
±10%. The 4 September Pons revenue figure is derived, not reported:
DefiLlama's V2 fee adapter has not posted for that date, so we applied the observed 4.65% take
rate to reported volume. It is an estimate and we have labelled it as one throughout.