Introductory Analysis of Finality Times in Crypto Networks

Key Insights

• Finality times in crypto networks are an important consideration for future applications, and high-value applications may require the fastest possible "true finality" time.
• Networks that support fast finality may have an advantage over slower networks, contrasting the Ethereum rollup-centric consensus in the crypto space today.
• Avalanche and Aptos are the clear winners when it comes to finality times among live networks, with Solana having the fastest finality paired with a large node count.
• Use cases for fast finality include on-chain games with high activity and complex interactions, high-frequency flash loan (hf-fl) leverage, and a SWIFT-like system executed via on-chain credentials and smart contract checks.

Time to finality (TTF) is often overlooked when compared to transactions per second (TPS). However, in future crypto applications, there may be an "elastic" demand for quick finality. Some apps won't require it at all, while others may demand the fastest possible "true finality" time.

If high-value applications end up prioritizing fast finality times, networks that support fast finality will have an advantage.

On the other hand, if high-value applications don’t end up valuing the difference in finality times, then slower networks may still be able to compete based on superior characteristics such as decentralization, developer support, liquidity, and others.

Defining Finality

Broadly speaking, crypto network consensus protocols can achieve two different kinds of finality — probabilistic and deterministic.

Probabilistic Finality

As the name suggests, systems that use probabilistic finality finalize transactions, well, probabilistically. As blocks are completed, the probability that a transaction has been finalized asymptotically approaches 100%.

There are several reasons why transactions never reach 100% certainty. Typically, the cause is competing blocks being proposed at the same time, causing validators to have differing views on which block (and therefore, which transactions) forms the head of the chain.

In that case, validators use a “fork-choice” algorithm that chooses the correct chain to follow. If a transaction happens to land in the incorrect chain, it is rejected and thus not finalized.

Probabilistic-finality networks include Ethereum, Bitcoin, Solana, and others.

Deterministic Finality

On the other hand, systems with deterministic finality “guarantee” that once a transaction is included in a block, it cannot be undone. These systems typically use consensus mechanisms like Tendermint (Cosmos and others) and Lachesis (Fantom) to ensure that blocks are never proposed simultaneously.

For example, in Tendermint consensus, blocks filled with transactions go through “pre-vote” and “pre-commit” stages. Both stages require two thirds of the validator set to come to a consensus on which block should be voted through and subsequently committed onto the chain. Blocks that make it through this “pre-consensus” will never face the problem of competing with another block to be added to the chain.

Source: Tendermint Docs

Examples of networks with deterministic finality include Cosmos, Fantom, and Polkadot.

Note: In-Protocol Versus Out-of-Protocol Consensus

The distinction between probabilistic and deterministic consensus is, as Jump puts it, an in-protocol distinction. This distinction holds so long as consensus is found within the parameters of the protocol.

Outside any given protocol exists a social layer. A protocol’s social layer can always choose to fork and rollback a chain for any number of reasons, regardless of consensus being deterministic or probabilistic. Hence, the social layer of a protocol enforces an “out-of-protocol” consensus.

For example, Ethereum infamously hard-forked on June 20, 2016 in response to the so-called DAO hack which happened 3 days prior on June 17.

The people who transacted between June 17 and June 20 that year probably thought their transactions were final.

They were not.

Social forks and rollbacks (forkbacks?) are rare historically, and hopefully they remain that way. Our analysis moving forward will focus solely on finality derived from in-protocol consensus.

Comparing Finality Times in Crypto Networks

Avalanche and Aptos are the clear winners when it comes to finality times among live networks. Avalanche is the top choice with transactions and blocks almost always confirmed in less than one and two seconds, respectively.

However, Solana has the fastest finality when paired with a large node count. Although Solana has near-identical block finality (or “slots” as called in Solana) as Avalanche, it has nearly triple the node count.

However, Solana suffers from a high finality time variance at extremes. The cause of this problem is not well understood. If the teams contributing to Solana’s core can solve it, they would boost the network’s value proposition significantly, as Solana maintains advantages in several other areas versus its competitors.

Given the current technical roadmap, Ethereum is unlikely to catch up with respect to finality times. Single-slot finality is the most ambitious finality time boost on the roadmap, but it does not bring Ethereum within striking range of Avalanche, Solana, and the like. Moreover, the Ethereum roadmap focuses heavily on execution happening on rollups rather than L1, further lengthening finality times.

A full list of sources may be found at the end of this report.

Are There Use Cases for Fast Finality?

If we cannot show any use cases which strictly demand faster finality times, the value of fast finality diminishes significantly. We looked across several sources — both traditional and crypto-native — to generate a prospective list of applications likely optimized with fast finality times.

Tradfi Markets

Finality times across traditional finance are surprisingly slow. Equities and bond markets take two days to settle. SWIFT transfers can take anywhere up to 4 days). The reasoning for this is twofold. First, especially in the case of SWIFT, a system of checks and balances is in place to make sure banks aren’t laundering money for one of the many various axes of evil. Those checks and balances take time, especially because this system was originally designed back in the 1970s.

We argue that, should a SWIFT-like system ever be implemented on chain, those checks and balances could be executed automagically via on-chain credentials and smart contract checks. In this scenario, banks would likely prefer to have their money finalized in milliseconds rather than in a week.

The second reason finality times are slow in TradFi is due to leverage dynamics. In both equities and bonds, highly professional players trade on amounts of leverage that would make a GMX ape proud. Rather than settling each of the thousands of trades as they happen, brokers allow these large players to settle the net result at the end of it all. In effect, the two-day settlement period is like the “deterministic finality” in the system, whereas brokers facilitating each trade as they happen within that two-day period is the “probabilistic finality.”

It would seem that the professional players would not prefer extremely fast finality times, as it would reduce the wiggle room used for leverage. However, crypto does allow for a solution to that — at least in theory.

High-Frequency Flash Loans (HF-FL)

One solution to the lack of leverage problem could be flash loans. Flash loans allow degenerates to take out almost infinite leverage, provided said degenerates can pay the leverage back within the same block. For flash loans to be effective for the high-frequency traders and market makers that dominate traditional finance, they would need to occur at a similarly fast pace. In order to do that, blocks would have to be confirmed as close to the speed of light as possible.

There’s already an indication that such a strategy would work wonders in crypto. Uniswap V3 LPs on Ethereum L1 can already provide much tighter liquidity by using just-in-time (JIT) LP practices. This would only improve with faster block times and the resulting access to flash loan leverage.

Quick On-chain games

In the context of on-chain games, faster finality times will be necessary for enabling any games that require fast-paced, complex interactions.

It is possible, and perhaps likely, that such games may not require the security of being directly on an L1. Instead, they could be content with instant L2 or even L3 finality. In those cases,a sort of probabilistic confirmation would instantly occur at the higher layers, allowing gameplay to move smoothly, while a more “deterministic” confirmation would occur later on. An example of this approach is Lattice’s OPCraft.

Final Thoughts

The analysis of finality times in crypto networks is an important consideration for future applications. Some high-value applications may demand the fastest possible "true finality" time. For these types of apps, networks that support fast finality would have an advantage over slower networks, contrasting the Ethereum rollup-centric consensus in the crypto space today.

Time will tell if the application use cases listed or others still uncovered will serve as a forcing function for the adoption of fast finality chains.

Finality Time Source List

Ethereum(current +single slot): Single Slot Finality

Zk Rollups: Optimistic vs. Zk Rollups

Optimistic Rollups: Norswap, Optimism Explorer

Cosmos Ecosystem: Tendermint Docs, Mintscan

Polkadot: Polkadot Wiki, Subscan

Sui: Sui Github Docs

Solana: Solana Compass, Validators.app

Fantom: ftmscan, Fantom Docs

Avalanche: Avascan, Avalanche Documentation

Aptos: Aptos Labs Explorer

Sei: Sei Blog

Let us know what you loved about the report, what may be missing, or share any other feedback by filling out this short form. All responses are subject to our Privacy Policy and Terms of Service.

All content was produced independently by the author(s) and does not necessarily reflect the opinions of Messari, Inc. Author(s) may hold cryptocurrencies named in this report. This report is meant for informational purposes only. It is not meant to serve as investment advice. You should conduct your own research, and consult an independent financial, tax, or legal advisor before making any investment decisions. Nothing contained in this report is a recommendation or suggestion, directly or indirectly, to buy, sell, make, or hold any investment, loan, commodity, or security, or to undertake any investment or trading strategy with respect to any investment, loan, commodity, security, or any issuer. This report should not be construed as an offer to sell or the solicitation of an offer to buy any security or commodity. Messari does not guarantee the sequence, accuracy, completeness, or timeliness of any information provided in this report. Please see our Terms of Service for more information.

No part of this report may be (a) copied, photocopied, duplicated in any form by any means or (b) redistributed without the prior written consent of Messari®.

‏Source messari.vercel.app