Blockchain indicator analysis trap: Why are active addresses and TVS unreliable?

Original title: The fallacy of daily active addresses

Original author: Donovan Choy, Blockworks

Original translation: TechFlow

Better use of blockchain metrics

Blockchains generate a lot of public data. On Crypto Twitter, people are constantly comparing blockchain A to blockchain B, and investors, researchers, and opinion leaders (KOLs) have many metrics to refer to when defending their views. However, the wrong use of these numbers often blurs people's understanding of this field.

In today's 0xResearch article, we will explore three metrics and their problems: active addresses, blockchain "profitability", and total value protection.

Active addresses

"Active addresses" refers to how many active, paying users there are on a certain protocol.

“Facebook has three billion monthly active users” is a useful piece of information that tells us something about the social network. Since there aren’t enough profitable opportunities for spammers to flood Facebook, active addresses are a good way to assess the true value of the platform to consumers.

But for blockchains, active addresses are less valuable due to how easy it is to create new wallets and the obvious opportunities to profit through airdrops or protocol incentives.

For example, the chart below shows a clear case: Solana has the most daily active addresses over the past month, so Solana looks very active.

Source: TokenTerminal

Most Solana users trade on decentralized exchanges (DEXs), so we need to look closely at activity on DEXs. When we drilled down into Solana’s active addresses on its DEX, we found that the majority of addresses — about 3.4 million, 4.4 million total — had traded less than $10 in lifetime volume over the past day.

This suggests that there may be a lot of spam or bot activity due to Solana’s low transaction fees, rather than a large number of “quality” users.

Source: Blockworks Research

Here’s another example I mentioned earlier: Celo L1 (now L2) saw a surge in daily active addresses sending stablecoins to 646,000 in September. This number surpassed Tron, and therefore attracted the attention of Vitalik Buterin and CoinDesk.

After further analysis, Jack Hackworth, a data analyst at Variant Fund, found that 77% of Celo addresses transferred less than two cents, mainly due to the thousands of users receiving tiny amounts of funds through a universal basic income protocol called GoodDollar. In both cases, the active addresses showed high usage, but upon closer analysis, this statement did not hold true.

For more information, see Dan Smith’s research, which focuses on the misuse of daily active addresses.

Blockchain Profitability

Rather than focusing on active addresses to study blockchain activity, it is better to look at the network fee metric. Fees reflect the total gas consumption of using the protocol, without considering the issue of "quality" users.

Fees are often used by analysts and investors to judge which blockchains generate the most "revenue". We then consider the issuance of tokens that the blockchain pays to validators as a cost. The result is the "profitability" of the blockchain.

This is how Token Terminal generates "financial statements" for crypto protocols. For example, the chart below shows that Ethereum L1 has accumulated millions of dollars in losses over the past two months.

Source: Token Terminal

The only problem is that this calculation does not take into account a key factor: unlike PoW chains (such as Bitcoin), users on PoS chains can also easily obtain token issuance rewards.

After all, if I can earn 5% of my ETH/SOL staking yield from a liquid staking platform like Lido or Jito, why should I care if the network is “unprofitable”? Therefore, it is problematic to conclude that “Ethereum is unprofitable” by considering token issuance as a cost.

In the real world, inflation is harmful because when central banks print a lot of money, the increased money supply reaches different participants in the economy at different times, and those who get the new money first benefit before the “real” price adjusts. This is known as the Cantillon effect.

This is not the case in a PoS blockchain economy because inflation (i.e. token issuance) is received by everyone at the same time. Therefore, no one becomes richer or poorer as a result - everyone’s wealth remains the same.

Instead, we can consider using an alternative metric called Real Economic Value (REV). REV combines network fees and MEV tips to validators, but does not consider token issuance as a cost.

Based on this, we can see that Ethereum has actually been profitable over the past two months:

Source: Blockworks Research

REV is arguably a better metric for assessing true demand for the network, and a more comparable revenue metric to traditional finance (TradFi).

In summary, traditional PL accounting is not easily and directly applicable to blockchains.

For more on this complex topic, listen to the recent Bell Curve podcast with Jon Charbonneau.

Total Transaction Value (TTV), not Total Value Secured (TVS)

Oracles are critical infrastructure for blockchains to access off-chain data. Without oracles like Chainlink, the blockchain economy cannot reliably reflect real-world prices.

A common way to compare the market share of Oracle providers is to use the "Total Value Secured" (TVS) metric, which aggregates all TVL secured by Oracles. This is how DefiLlama explicitly calculates it:

Source: DefiLlama

The problem with TVS is that it obscures the activity that Oracles are actually securing.

For example, Oracles that power high-frequency trading products like perpetual contract exchanges continually “pull” price updates from off-chain data sources with sub-second latency.

This contrasts with “push” Oracles used for lending protocols, which only need to update prices on-chain a few times a day because they don’t need to be updated as frequently.

TVS focuses on the total value managed by Oracles, but ignores the performance strength of the Oracle providers.

In other words, it’s like saying that a fine steak and a salad are both priced at $50 on the menu, so they are of the same value to the diner. But obviously, it takes far more work to make a steak than a simple salad, and that’s a factor worth considering.

An alternative metric is total transaction value (TTV), which takes into account the periodic volume of transactions that use Oracles to update pricing.

TTV excludes low-frequency applications such as lending, CDPs, and re-collateralization, but as Ryan Connor explains, "Only 2-9% of Oracle price updates come from these low-frequency protocols, which is a small percentage in the crypto space because fundamentals are so volatile."

When Oracles are evaluated against TTV, market share changes significantly.

For more information, see Blockworks Research's report on how TTV can better reflect Oracle fundamentals.

Original link