Connext Deep Dive - Josh

Overview

Connext presents the opportunity to bet on a multi-chain, multi-layer world through a capital infrastructure play
At its core, Connext is an L2<>L2 / L1 bridging solution, allowing users to move capital between chains / layers in a trustless and efficient manner
With a strong team and a very competitive product, Connext will be able to capitalize on winner-take-all economics and, through its SDK-based business model, will become the backbone to dApp cross-chain operations. Overall, Connext’s value lies in its future abstraction — it is ‘behind the scenes’ financial plumbing
Finally, Connext presents great opportunities for value add as well as multiple avenues to exit

Protocol

Connext is an interoperability protocol for interacting between L1s and L2s. Without introducing any new trust assumptions or external validators, Connext’s NXTP (the protocol technology) lets users swap assets across L1s and L2s in an efficient and timely manner.

Note that the emphasis is on the word ‘swap’ and not ‘port’ or ‘migrate’. The current version of Connext cannot be used to pass arbitrary event data between chains; therefore, one cannot use the protocol to migrate a token from Chain A to B
Albeit, users will not become frustrated at this inability, as swapping USDC on BSC to USDC on Optimism is basically the same as migrating, except that it is cheaper and faster than using migrating-based bridges
NXTP is a lightweight, simple contract that uses a locking pattern, a network of off-chain routers that participate in pricing auctions and pass ‘calldata’ between chains, and a user-side SDK that finds routes and prompts on-chain transactions

Ecosystem Players:

Users

Interact with a front end that is powered by NXTP. Best example right now is xPollinate
Undergo the transaction process (discussed below) to fulfill the need of swapping one asset from one chain for it on another chain / rollup

Routers

These are the liquidity providers, which set up a pool of assets on two chains / rollups and link up with NXTP to provide liquidity for swaps

dApps

Integrate with NXTP by using the SDK. They then can manage front ends that channel assets to the NXTP contract for swapping to occur

Transaction Process:

The process can be broken down into three phases:

Route action: the user broadcasts a message signaling their desired use of a specific bridging route. Routers respond with sealed bids containing commitments to fulfilling the transaction within a certain time and price range
Preparation of transaction: once a bidder is selected, the user submits a transaction containing the router’s signed bid to the contract on the user’s side. This consequently locks up their funds on the user chain. The router detects an event (sealed chosen bid message) and the contract locks up the assets on their side. This lock up is incorporative of their fees
Fulfillment: the user’s side will then receive a confirmation message of funds locked and it will send a signed message for a relayer to claim. This relayer (typically another router) will submit the transaction and receive a submission fee
Note that transactions can be cancelled unilaterally by the person owed funds on that chain (router for sending chain, user for receiving chain) prior to expiry

Adoption Metrics

Daily Volume

Daily Transaction Count

Implied Average Transaction Value

Team

https://twitter.com/arjunbhuptani

Connext, Co-founder & Project Lead (May 2017 — today)
Moloch DAO, Co-founder (Oct 2018 — today)
Colgate University, B.S. in Physics and B.A. in Philosophy

Notes:

In the space since 2017, great experience and recognition
Known for a strong grasp of state channel mechanics and inter-chain flow of value and information
Previously, Connext used to focus on payments infrastructure within crypto and consequently solving for scalability became key

https://twitter.com/LayneHaber

Connext, Co-founder & Protocol Lead (Jun 2017 — today)
Arctica Health, Founder & CEO (May 2016 — Jun 2017)
UCLA, B.S. in Chemistry and Materials Science

https://twitter.com/RHLSTHRM

Connext, Co-founder & CTO (Jun 2017 — today)
Lone Wolf Partners, Partner (Jan 2016 — today)
ColorTokens, Core Engineer (Jun 2015 — Apr 2017)
Tip’d Off, Co-founder (May 2014 — Jun 2015)
Tesla, Senior Test Engineer (Sep 2010 — May 2014)
Element Energy, Engineering Manager (Jun 2010 — Sep 2010)
UC Santa Cruz, B.S. Electrical Engineering

Notes:

Deep skillset suitable for technical side of Connext. Full-stack developer (NodeJS, React, React Native, RDS, NoSQL, Cloud Infrastructure, DevOps, Serverless) + Ethereum developer (Solidity, Truffle)
Longer history as a developer in crypto relative to others

https://twitter.com/sanchay_mittal

Connext, Project Manager (Nov 2020 — today)
Other: Gitcoin, ConsenSys, HyperLedger

Jake Kidd

Connext, Full-stack Developer (Feb 2021 — today)
Other: Neato Robotics, Akruta

Cap Table

First raise when company was an Ethereum payments solution (2017): ~$1m
Bridge round (2019): $0.5m from ConsenSys Ventures
Seed (03/22/21): Raised $2.2m from Polychain, 1kx, Huobi Capital
Ecosystem round (07/13/21): Raised $12m from 1kx, ConsenSys Mesh, Coinbase Ventures, MetaCartel Ventures, Scalar, and others. Also included investments from staking service providers (i.e. Blockdaemon, Figment, Stakefish) and ecosystem players (i.e. Edge and Node, Loopring, Arbitrum founders)

Competitive Analysis

There are three ways to approach interoperability between chains:

Narrow Competition

Hop Protocol

A locally verified mechanism for sending tokens across rollups. Hop leverages existing arbitrary messaging bridges (“AMBs”) to send funds between chains, with a bonder (liquidity provider) fronting the capital to make the process fast. To incentivize rebalancing, the protocol also utilizes AMMs on both sides to swap between the “canonical” asset for a chain and hTokens, a representative asset used by the bonder
Since the proof of transfer is passed between chains using an AMB, Hop doesn’t need users to run off-chain code like Connext

Presents the following tradeoffs:

Hop is less economically secure than Connext. It depends on AMBs, which means that if chains don’t have them, they need to be created. This introduces new security risks. Hop’s design also cuts corners like optimistic rollups’ seven day exit periods for settlement
Hop is less capital efficient. It requires both entry and exit liquidity given its AMM model on both sender and receiver chains. In contrast, Connext LPs only need exit liquidity and incur no liquidity lockup, “achieving 10x+ better capital usage”
Limitations to bonding. Bonders must pay gas costs based on Ethereum mainnet. Connext can have transactions go directly from L2 to L2 without touching the base layer
Bonders are exposed to MEV and transaction submission races, according to the Connext team. Transactions cannot be front run on Connext as the users negotiate route and pricing off-chain

Synapse

Synapse conducts integral blockchain activities such as asset transfers, swaps, and generalized messaging with cross-chain functionality. The Synapse network is secured by cross-chain multi-party computation (“MPC”) validators operating with threshold signature schemes (“TSS”). The network is leaderless and maintains security by each validator running the same process upon receiving on-chain events. Once two-thirds of all validators have collectively signed the same transaction using their own individual key, the network achieves consensus and issues a transaction to the destination chain
Currently, validators are selected based on community governance and consensus. Upon reaching a certain project milestone, validators on the Synapse Network will have to stake $SYN to secure the network

Presents the following tradeoffs:

While it can enable fast and efficient swaps, it introduces significant new trust primitives that can lead to increased security risk. With a whole new group of validators (currently selected by the community, which can be worrying) and game theory, Synapse adds complexity, whereas Connext bypasses it. Both can do the same thing when it comes to swaps, except that right now, Synapse can use its protocol ($SYN) to increase yields for pool deposits, creating increased liquidity in the short term
When compared to Connext, certain swapping routes are relatively complicated and involve several more swaps than needed. This may introduce issues pertaining to higher fees in the long run, as well as increased risks along the way

Representative screenshot from Synapse Docs

Wide Competition

Ren

Not necessarily trustless
Lacks scalable way to integrate with other chains / layers given its security configurations
Great for porting over assets, not necessarily moving value fast

Cosmos

Works when all chains run on Tendermint / are a parachain
Expensive in terms of time and fees. Like Ren, the focus isn’t on moving value fast and efficiently

THORChain

Positioned more as an L1<>L1 exchange that seeks to establish permissionless swapping between L1 native assets, using RUNE as a medium of exchange as well as for security collateral. There is a future where THORChain releases pools for rollup swaps, however, this is not within the project’s current core focus
Leads to increased security risk, whereas Connext inherits the security of the base chains it works on top off
If THORChain’s decentralized swapping capabilities prove successful and scalable, this could be a threat to Connext’s offerings pertaining to L1<>L1 swaps

Bridges

Examples: Optimistic bridge, Near Rainbow bridge, Avalanche bridge
These lack the time and cost benefits that Connext offers, and in many ways, are the worst choices presented to users. They are crucial in the infrastructure they provide, but their use should be minimized to significant capital movements (in order to set up swapping pools and liquidity on the other side) as well as batch-settlement capital flows

Potential Investment Theses

#1 If one underwrites the thesis of a multi-chain, multi-layer world, Connext offers a ‘project-neutral’ way to bet on it. At its essence, it is an infrastructure play and wins no matter which L1s and L2s prevail.

Competition in the L1 and L2 space can intensify, developing commodity-like market traits
The cross-chain bridging space, on the other hand, is indicative of winner-take-all economics. With L1s or L2s, it makes sense for a user / dApp to use multiple ones (as they have different offerings); however, for a bridge aggregator technology, it makes more sense to use the best one that covers all ecosystems, rather than multiple ones
Location within the value chain can emerge as a value sap: Connext is a group of capital superhighways and generates revenue through tolls

#2 Value lies in the SDK. Connext is a B2B protocol that will behave like abstracted infrastructure within the logic of dApps. It strategically makes sense for dApps to have inter-chain functionality built into their ops as dApps will want to maintain users on their site for the whole cross-chain process, otherwise they risk losing the user to another competing dApp. Therefore, Connext’s capital plumbing will be of value.

This is not a long-term prediction; rather, dApp integration with cross-chain liquidity infrastructure is becoming a priority. Several takeaways from ETH Lisbon have discussed dApp teams needing to focus on solving for cross-chain capabilities.

#3 Sizable TAM. The value of transactions going through bridges will be in the billions, and eventually trillions, assuming the multi-chain, multi-layer world develops. With some tweak-able conservative assumptions, Fee TAM is in the low billion single digits:

#4 Great team with deep sub-sector expertise

Known amongst the right circles for their leadership in cross-chain bridging
Rich dev experience suitable for the complexity of the team’s goal
Past track record highlights ability to execute. Specifically, recent releases have been well received, and in circumstances of bugs, they were corrected and updated in a timely manner

#5 Opportunities for value add combined with multiple exit strategies

Leverage network as integration opportunities for the Connext team
Assist with team hiring and community growth. Currently, the community is somewhat small as evidenced by experience on Discord. Achieving a community like Index Coop, where several BD strategies can be led by members, could formulate part of the go-to-market
Offer expertise on token economics design

Value add is complemented by multiple paths to exit:

M&A: Can be acquired by another bridging solution, wallet provider (i.e. Metamask), or infrastructure company
Token (much more likely): Can go liquid via a token drop, with lockup expiry probably lying between 1–4 years

Catalysts

#1 Growth of L2 ecosystems

Once more dApps onboard and bring transaction-intensive ops to L2s, ecosystem growth will really hit full tilt
Progress with development at Arbitrum, Optimism, and other chains / rollups point to the imminence of this catalyst

#2 High gas environment on Ethereum

Drives capital to other ‘cheap’ chains: Avalanche, Binance, and Fantom

#3 Connext’s integrations with other frontends and applications

This will drive transaction flow to Connext and will lead to recognition of its cheap and fast cross-chain swapping mechanism. Progress with integrations is going well according to the team

#4 Progress with liquidity expansion

Team recently selected the first cohort of routers

Risks

#1 Smart contract risk

Cross-chain swapping attracts smart contract and game theory risks across the board. Part of the allure of being able to pull this infrastructure design off is matched by the higher burden of risk. Given that state channel-based bridging at scale hasn’t been done before, several unknown unknowns are present. Additionally, it is still early, and while the Connext team has made respectable progress to date, scalability and growth can introduce several issues that may not exist at this point in time

#2 Execution risk

While the team has great experience, significant execution risk exists. It has yet to be seen whether this team can adequately scale a platform and turn it into an infrastructure layer provider like Chainlink or the Graph. Team hires and community growth will be key, a possible area of value add

#3 PMF risk

Assuming PoS / MPC-based cross-chain swapping applications work at scale (in light of the aforementioned security tradeoffs), and dApps desire the ability to do more than swap cross-chain (i.e. general information delivery), then Connext will be at a disadvantage to protocols like Synapse and THORChain. These protocols would be able to bundle swapping with cross-chain data delivery in one offering

#4 Aggregators win out over protocols

Several great minds in the space see value accrual power moving over to app-based aggregators. These will plug in and provide multiple services to users. If fee compression were to start hitting DeFi (some signs of this already), then aggregators would sit on top and find different revenue opportunities while bridge protocols could be commoditized

Potential Model

Drivers:

Transaction Count
Average Transaction Value

Assumptions:

Projections:

Valuation:

Disclosure: This blog series is strictly personal/ educational and is not investment advice nor a solicitation to buy or sell any assets. Please always do your own research.Disclosure: This blog series is strictly personal/ educational and is not investment advice nor a solicitation to buy or sell any assets. It does not represent any views from where the author is working — all views, opinions, and arguments are the author’s. Please always do your own research.