Miyamoto Musashi’s Way, applied to markets.
Earth — foundation
Water — stance and adaptation
Fire — engagement and timing
Wind — other styles
Void — emptiness
Everything you need is within you.
Do nothing that is of no use.
DeFi revenue is recovering.
Protocol revenue is up 11% in a month, and the revenue meta is the playbook everyone is quoting.
So I went through the major DeFi tokens: 30-day revenue, how much reaches holders, and what keeps it growing.
S tier
$HYPE: 99% of trading fees buy and burn HYPE. That was $55.4M in 30 days, the most of any token, but at 30x holder revenue the price already reflects it.
The first USDC yield payout (~$14.5M) hits the buyback fund on 3 Oct, and HIP-4 outcome markets will add fees after testing.
A tier
$PUMP: Half of revenue goes to buy and burn, $23.9M last month for a 10.4% yield. It trades under 10x because the income depends on memecoin launches. Custom Pairs and trading on EVM chains broaden that.
$UNI: The fee switch reached v4 and Robinhood Chain in late July, taking the monthly burn from $9.6M to $15.2M. Votes to extend v4 fees to 5 more chains are expected.
$AERO: All swap fees go to lockers. That was $15.1M last month, triple the month before, though emissions offset part of the 23% yield. The Velodrome merger launches as Aero on 21 Oct across 7 chains.
$AAVE: Borrowers paid $36.5M in fees and Aave kept $5.1M. Aavenomics 3.0 buys back ~292 $AAVE a day, Stani has floated burning them, and V4 TVL grew 82% in September.
B tier
$SKY: Holders got $5.2M of $13.5M revenue last month. Spark, Grove and Obex borrow $5.9B of USDS against $17.5B in limits, so borrowing has room to triple.
$JUP: Half of revenue buys back JUP, about $3.1M last month. Unlocks outweighed it until team emissions were paused this year. A community proposal would lift the share to 70% and burn.
$LIT: Trading fees buy and burn LIT, $2.6M last month for a 3.2% yield. Options are next. Against that, Robinhood is routing its US perps elsewhere, and team unlocks start Jan 2027.
$LDO: Lido buys $LDO from its staking fee, capped at $10M a year. Its share of staked ETH fell from 26% to 23% over the year. stVaults is meant to reverse that, and its fee waiver ends 31 Oct.
$ENA: Ethena made $23.5M in fees and ENA captures almost none of it. The fee switch turns on at $7.5B USDe supply, starting at 5% and rising to 20% at $20B. Supply is $4.9B today.
$BP: The exception. Backpack is an exchange, so $BP gets no buyback or fee share, but stakers of 1+ year get a claim on up to 20% of company equity at an IPO or acquisition. With 75% of supply locked, it's a bet on that IPO and Backpack's lead in tokenized stocks on Solana.
C tier
$ETHFI: EtherFi earned $3.8M last month, mostly from the Cash card, where spend went from $54M in January to $129M in September. Buybacks stopped after Q2, and resumed in mid-Sep with over 500k $ETHFI bought back.
$PENDLE: 80% of revenue goes to $sPENDLE, but revenue fell from $2.1M in April to $0.35M in Sep. Boros earns about $35K a month, and its volume has fallen since April. New tokenized assets on Pendle v2 and new markets on Boros will support Pendle's bounceback.
$RAY: A cut of trading fees buys back $RAY. Revenue rose about 5x to $7.5M after LaunchLab opened to any pair, including stock-paired launches. The 10.7% yield holds only if that volume stays.
$CAKE: Every swap burns some $CAKE, and supply has shrunk for nearly 3 years. It yields 7.1% at 9.5x revenue, with the Infinity AMM deploying on Circle's Arc next.
D tier
$MORPHO: $19.8M in fees last month and no fee switch. It runs Robinhood's Earn product, has $5.5B in loans, and Coinbase now offers BTC-backed fixed-rate loans on it. Until a fee vote, the token is priced on an option.
$FLUID: Revenue is $0.6M a month, with no buybacks since Q1 after $4.75M over the year. DEX v2 and Jupiter Lend's growth could change that.
$JTO: Staking and MEV fees go to the DAO treasury, and the token is down 67% on the year. Since July, 80% of JTX terminal fees go to buying and burning $JTO, the first direct link between product and token.
holy moly, we won the 1st prize on the hackathon with a 3D tea shop using three.js and mostly Opus 5.5, Cursor and with Nansen API data under the hood.
kuddos to my team and h/t to @nansen_ai for organizing. had a lot of fun building this. more cool stuff about to come to our cozy teashop!
also h/t @MengTo for inspiration and three js elements used inside the app and @tripoai for building a great product which helped us create a nice 3d model of uncle inside the teashop.
150+ builders entered the Nansen Meridian Buildathon.
These are the ones that took it home.
🏆 1st: @0x_iroh — $10,000 USDC
🥈 2nd: @th3Wh1t3Rabbit — AirPods Max + Ledger Stax + Keychron Q Pro
🥉 3rd: @iAteUrSOL — Sony WH-1000XM5 + Ledger Stax
Plus 5 honorable mentions earning 100,000 Nansen API credits each.
Here’s what they built ↓
In this day and age you don’t see many project as transparent as @NetNetCap.
The fact that they are shipping new experiences every week and still haven’t launched their blue chip product is insane to comprehend if you ever tried to build sth.
$NET cannot go to zero. The immutable core contracts set backing at $173 right now, including the RWA sleeve its over $210.
At these levels we are aggressively buying $NET . uncapped upside with a worse case 50% drawdown.
The treasury continues to climb, V3 RWBs turn off dilution, we have $1M in stables coming to the credit desk today which we will use to buy more $NET.
Overall, we've grown our treasury to nearly $26M in under 3 months, we have basically built all of 2021-22 DeFi in under 3 months, We've delivered brand new on-chain experiences including on no-loss lotteries, no-loss sportsbooks, an advance rebase desk, and more. We have the largest morpho blue market on RH chain, and we have not even released our bluechip (frontier) product yet.
We've also done all of this without any public acknowledgement of any kind from RobinHood yet. Worth mentioning.
A bet on $NET is a bet that our team will be able to grow the treasury past the pace of emissions. So far that bet has been correct, but it is clear that additional revenue streams outside of bonds are required for the next leg up.
I am betting that we can deliver.
This argument that someone had to be a good trader to get and retain copy traders is so incorrect its not even funny.
For one, you only need to be popular to get copy traders.
Secondly, you can make one lucky trade then get copy traders and from there you switch to low caps where you are able to move price enough to build up stats/pnls and show up on pnl leaderboards which then leads to increased popularity and more copy traders. Self fulfilling prophecy.
And lastly, you can fake / manipulate stats to get on leaderboards which leads to popularity and copy traders. Then its once again a self fulfilling prophecy.
"But wouldnt the people just stop copying?"
When you have enough copies and buy low enough mc tokens you are able to artificially make anything a winner to some extent. And at a certain level its to the point that the volume from your copies moves every pair you buy enough that it reaches volume trackers for at least a short period of time giving it the chance to reach outside of your bubble.
Obviously not to this is the case for everyone. Frank for example can move 8 figure market cap coins with his buys the same way most of these guys move $4k mc shitters lol...
We built a 3D tea shop with three.js for the Nansen Meridian Buildathon 🍵 Hope you guys enjoy it!
What we have live now:
• Read Uncle's 3 crypto theses scored by smart money
• Chat with Uncle, an AI host (Nansen Research agent under the hood)
• Meet the month's top HL traders, smart money, whales and memecoin traders.
This has been so fun and exciting to build!
Many, many AI tokens were used, a lot of polishing and testing but the result ended up being so nice, interactive and hopefully useful.
Team: @tldde@0x_Takezo@david_grii@PandaCoderexe
Built with @nansen_ai
Try it live: https://t.co/fjz0T6TASi
Github: https://t.co/cKu8vNZJAX