Cheat Sheet: Messari Crypto Thesis 2023

The pon(z)ies.

Daniel Lanciana
10 min readJan 17, 2023

--

Things I found interesting from the 147-page thesis. My comments in non-quoted italic.

Don’t understand the copy-paste investing section. Starts professional, but a bit sloppy at the end (e.g. “Don’t. Fuck. Up. The. Money.” and “get in loser, we’re going shopping”). Littered with plugs and ads for Messari (fair enough). Couple of overblown statements (“our physical survival depends on the decentralisation of hardware…hang in the balance between an open internet and a global police state”). Overall an interesting read.

TL;DR

  • Prolonged crypto winter (bear market) for the foreseeable future, with no V-shaped recovery. Market sentiment is a recession in 2023. A classic credit boom and bust cycle.
  • 2023 will be a bloodbath for crypto startups
  • Time to build!
  • Embrace personal wallets (“not your keys, not your coins”) and decentralised finance. Frame desktop wallet. Brave browser.
  • Ethereum will stay over 70% dominant during the crypto winter

Action Items

  • Market Value to Realised Value (MVRV) has been a good indicator for how hot/cold the market is. MVRV below 1 is buy; above 3 is sell. Chart.
  • After the Ethereum Shanghai upgrade scheduled for September 2023, no reason not to stake (if you have 32 ETH)
  • Lido and Rocket Pool tokens could be long-term blue chip DeFi assets
  • Bullish on Zcash (especially) and Monero tokens
  • NFTs as investments is largely over. NFTs as consumables is just getting started. Overall the NFT market size will 100x in 10 years.

Quotes

“If you paid taxes, didn’t get greedy day trading, and parked some cash for a rainy day…you’re probably doing ok.”

“As an investor, your choice is to either pick winners or buy the basket.”

“Twitter is rarely a drone strike…it’s often a suicide vest.”

“By far the most significant event in finance during the past decade has been the extraordinary development and expansion of financial derivatives.” — Fed Reserve Chairman, Alan Greenspan (1999)

“Crypto is embedded not just with cypherpunk values, but with the values of the US nation’s founders.”

“You don’t compete on fees, you compete on value.”

2022

“Yield-hungry investors were forced further out onto the risk curve. If you don’t understand the yield, you are the yield.”

“A lot of people farmed tokens and played the pon(z)ies.”

  • Last year was We’re All Going to Make It (WAGMI), but that didn’t really pan out
  • At the end of 2022, a memecoin (Dogecoin) was top three by market cap
  • $3 billion in on-chain hacks in 2022
  • Terra depeg (wiped $60 billion). FTX collapse (wiped $32 billion). GBTC investors billions underwater.
  • Grayscale Trust allowed investors to use their 401k to buy GBTC through OTC-traded securities — but they weren’t ETFs. Became toxic collateral. SEC screwed up.
  • Crypto (tech in general) layoffs
  • Tornado Cash that uses zero-knowledge proofs for fully private Ethereum transactions has been accused of laundering over $1 billion in crypto. It’s Dutch co-founded has been jailed without charge since August. It was placed on the SDN blocked persons list — including etherium addresses for the smart contract!
  • AAM Curve was hijacked by mercinary board members. Balance is dealing with hostage negotiations with a shareholder. Sushi had a nightmare reorganising leadership.

The lessons are fairly straightforward: don’t co-mingle customer assets, cut losses early on bad traded, don’t double-down and pray; maintain internal controls, split assets across custodians, and only keep on exchange what you can afford to lose. In other words, don’t get too greedy.

Bitcoin

  • Proven durability, but it’s role as a usable currency is questionable
  • Only 39% of mining energy is renewable; it produces about the same e-waste (e.g. hard drives) as the Netherlands — but costs are capped by market forces. The carbon footprint is much less than traditional banking. Mining might strengthen renewable grids by using excess energy during off-peak times.
  • Gold parity would mean $500,000 per coin

Ethereum

  • Merge a huge milestone for crypto six years in the making. A $200 billion network porting to a new ledger and security model, in real time, with zero hiccups. A spectacular technical feat.
  • Proof-of-Stake reduces the environmental footprint by 99%, reduces new token issuance by 90%, and eliminates nearly $500 million in monthly sell pressure from miners. It created a net deflationary (burns ~85% of all transaction fees) asset with real yield (5–7%). No other crypto project has achieves this supply dynamic. It could provide 1–2% steady deflation per year.
  • Requires 32 ETH to stake, with 6% returns. Once the Shanghai upgrade is released in 2023, the minimum staking duration will be 27 hours (current stakers can’t withdraw). Little to no reason to avoid staking — expect holders to stake en masse.
  • Staking services such as Lido and Rocket Pool provide access (e.g. stake less than 32 ETH) and liquidity (trade ETH for other tokens)
  • Maximum Extractable Value (MEV) — validators picking the most profitable transactions, their order, or injecting (i.e. front-running) their own transactions. The biggest concern is censorship by large validators who voluntarily enforce OFAC compliance to not anger the Government.
  • MEV is one of the biggest technical challenges remaining. If done well, it can arguably make protocols more liquid, efficient, and antifragile
  • Next step of the roadmap is to reach 100,000 transactions per second in “blobs” in EIP-4844, or Proto-Danksharding (!). Later come SNARKs for easier verifying and Verkle trees — cryptographic proofs paving the way for stateless clients.
  • Interoperability bridging controls are an important emerging area, but have led to numerous hacks (Axie Ronin for $600 million, Wormhole $230 million, Nomad $200 million)

Layers

  • Layer 0 (L0) is the “moral stack.” Good governance and social primitive values at the community level are lacking.
  • Layer 1 (L1) is the core blockchain (e.g. Ethereum).
  • Alt-L1s (i.e. alternative blockchains) such as Aptos and Sui, which use an open-source Rust-derived smart contract language called Move. Both claim to be high-speed and throughput (finalise transactions in under a second) by decoupling consensus and parallelising transaction processing. Also Avalanche and Polkadot.
  • An appchain is a L1 application-specific blockchain designed to only operate a single application.
  • Layer 2 (L2) increase throughput by moving computation and state-storage off-chain. Ecosystems such as Optimism and ZK-rollup have been rapidly evolving. L2 competition will look similar to the browser wars.
  • L2 Optimistic rollups (Arbitrum, Optimism) rely on fraud proofs (i.e. assumed valid unless challenged within a time period).
  • L2 Zero-knowledge rollups (ZK-rollups) rely on validity proofs are are coming “soon.”
  • L2 Sovereign rollups are private blockchains that just use the base later to store data.

Monolithic vs Modular

  • Blockchains with a single settlement layer are “monolithic,” while those with many are “modular” — similar to microservice architecture
  • Modular rollups such as Celestia allow developers to mix and match protocols. It only orders and publishes transactions, but does not execute them. Allows users to deploy their own blockchains with minimal overhead.

Stablecoins

  • The backbone of the crypto economy
  • You can optimise for censorship resistance, price stability, or reserve auditability — but not all three.
  • Purely algorithmic stablecoins like Terra work well…until they depeg. Still experimenting. Is it really healthy to give these programmatic black swans oxygen? Is the juice worth the squeeze?
  • MakerDAO has shown resilience
  • Frax Protocol is pioneering fractional reserve stablecoins — a sweet spot between overcollateralised and pure algorithmic.
  • Predicting an inflation-resistant stablecoin within a few decades

NFTs

“You better like the jpeg enough to hold it forever.”

  • Market cap is between $7.5–10 billion, with 2/3 from ProFile Picture (PFP) collections like Crypto Punks
  • Will be like ICOs during the 2013–15 Bitcoin bubble. Essentially unlimited potential to “become a ubiquitous standard for wrapping financial assets in the same way they currently wrap monkey JPEGs.”
  • Art is a Veblen good (i.e. a luxury good for which the demand increases as the price increases)
  • Bored Ape Yacht Club (BAYC) made an ApeCoin and BAYC metaverse!
  • Art Blocks is the fourth-largest NFT project by volume. It sells generative art, like pictures of squiggles, for thousands.
  • Gucci sold a digital bag for $800 more than the real thing
  • OpenSea has the same brand advantage as Coainbase. It also defaults enables creator royalties. Most low-fee marketplace (i.e. other than OpenSea) activity is wash trading.

dApps

  • The Axie Infinity blockchain game brought in $1.35 billion
  • Decentral Games is a DAO with crypto poker rooms and takes in $2 million per year in fees
  • Decentralised Social (DeSo) split into three layers: front end, content storage (IPFS, Arweave), and social graph (connections, content). Any developer can build a front end using the shared social graph. Transparency, user control, portability, revenue shares, and 8 billion users.
  • Ethereum Name Service (ENS) are human-readable wrappers (i.e. domain names) for wallets
  • Urbit is an enigmatic, open-source, privacy-focused version of WeChat

DAOs

“A long-term hassle to figure out is how DAOs actually work in the real world from tax, contract law, and compliance standpoint.”

  • ConstitutionDAO was an attempt to buy a copy of the US constitution. SPAC model. Could it be used to rescue a sports club? How does it stand up in court?
  • Gitcoin pioneered quadratic funding (i.e. mathematically optimal way to fund public goods) and funded over $70 million in grants to projects such as Optimism and Uniswap
  • Network States are (hypothetical) online communities that crowdfund and buy physical territory, perform collective action, and gain diplomatic recognition (i.e. sovereignty). Do they fund armies?

Tokens

  • Good token design solves the “cold start” problem by looking more like growth capital than seed funding. It creates an early user base and improves the product. They only work in markets where network effects exist.

CeFi (Centralised Finance)

  • Binance accounts for nearly 75% of global spot volume (i.e. immediate exchange). In hot water with various policymakers.
  • Coinbase is the top custodian of crypto assets with nearly 10% of the total market cap. Has $3.5 billion debt yield (i.e. return that a lender would receive if the borrower defaulted on the loan) of 15%. Stock fell 80%.
  • Over half of all staked Ether is controlled by three entities
  • Over half of Bitcoin’s hash rate is supplied by three mining pools
  • Most on-chain data is read using a few APIs (Infura, Alchemy)
  • Nearly 70% of hosted Ethereum nodes are hosted on three cloud providers (AWS, Hetzner, OVH)
  • Around 60% of Ethereum validators are located in Germany or the US
  • After FTX, movement for companies to provide “proof-of-reserves”
  • Need better access to secure, verified, real-world data (i.e. an oracle such as decentralised Chainlink) and on-chain state

DeFi (Decentralised Finance)

  • Flash loans allow suers to borrow large amounts of crypto for transactions within the same block without providing upfront collateral — provided the loan is repaid within the same block. Allows any developer to take advantage of arbitrage opportunities.
  • Uniswap v3 is a 10x improvement and the market leader for Decentralised Exchange (DEX) using Automated Market Makers (AMMs). Tokens are staked in pools for a return (equivalent to the staked percentage within a pool) taken from fees of each trade. AAMs adjust the price of an asset based on the size of the pool — and relies on arbitrage from users to keep prices aligned.
  • 0x is the backbone of several NFTs, but on-chain order books require a constant stream of updates — which is expensive

Tax

“Tax reporting for crypto is pure nightmare fuel.”

“Authorities will come for their pound of flesh.”

“Taxes due from a bull run are still due in full — even if your bag gets slashed the next year.”

  • The IRS added 87,000 new agents and will create hell for crypto investors
  • Under civil asset forfeiture law, which is essentially state-sanctioned theft, it may become easier for authorities to seize crypto assets without filing charges. Between 2018 and 2021, the IRS seized $3.8 billion.

Regulation

  • In the US, monetary code is speech (!) and not subject to unreasonable search and seizure.
  • Markets in Crypto-to-Assets (MiCA) and Transfer Funds Regulation (TFR) will go into effect in Europe in 2024. It provides a path to a digital euro stablecoin, but algorithmic stablecoins are essentially banned. Sustainability standards might also provide a backdoor to ban Proof-of-Work mining.
  • FTX was pushing through the Digital Commodities and Consumer Protection Act (DCCPA), which favoured FTX and hurt other crypto firms — including Binance. The bill was leaked. Binance’s Changpeng Zhao (CZ) didn’t bail out FTX in large part because of this lobbying.

Innovation

  • Need to solve security issues at scale (AI monitoring, algorithmic security breakers)
  • Solana Saga (Android phone for crypto)
  • Light clients to allow mobile users to connect directly to networks
  • Prediction markets are “always one year away” and not attractive
  • Crypto is notoriously easy to track with forensic tools like Chainalysis
  • The trend towards de-globalisation will push non-US/EU nations to look to “outside money” (gold, commodities, crypto)

AI

“The lawyers always win.”

“Crypto-powered identity is the future.”

  • Will OpenAI/ChatGPT upgrade or downgrade our cognitive abilities, or simply rewire our brains? GPS made us worse at directions and spatial awareness, but made us better off. Social media made us better at connectivity, but less mentally healthy.
  • One of the biggest boons to productivity in the next decade — similar to early Google and Wikipedia. The imminent demise of Google has been greatly exaggerated.
  • AI is neither good or bad, but the data sets might be
  • Horrifying potential for public (or hacked) data (tweets, emails, posts) to be uploaded to AI and queried. Blockchain required to siphon data from soon-to-be AI overlords.
  • See Ocean Protocol data tokens, which can be shared or monetised (i.e. sell your data)

Links

--

--