3-Year ETH/SOL Holding Experience Recap: Market Data and Ecosystem Development Outlook
I've held ETH/SOL for 3 years, and the market doesn't lie — after a year and a half, users have voted with their actions. The takeaway: * Ethereum still has more capital and developers than Solana. * Ethereum mainnet users have migrated to L2s, but L1's ability to capture L2 value remains weak. Solana's user base is significantly larger than Ethereum L1. * ETH-denominated assets have had fewer cycles in this bull run compared to SOL-denominated assets. * Ethereum's ecosystem development has gone off course.
I've held ETH/SOL for 3 years, and the market doesn't lie — after a year and a half, users have voted with their actions. The takeaway:
- Ethereum still has more capital and developers than Solana.
- Ethereum mainnet users have migrated to L2s, but L1's ability to capture L2 value remains weak. Solana's user base is significantly larger than Ethereum L1.
- ETH-denominated assets have had fewer cycles in this bull run compared to SOL-denominated assets.
- Ethereum's ecosystem development has gone off course.
In the past, we evaluated public chains primarily on: 1. TVL capital; 2. Developer count; 3. Users; 4. Ecosystem; (5. Whale presence). We'll go deeper on these five dimensions to analyze the current state of the Ethereum ecosystem (L1 + L2) and Solana.
1. Capital
Ethereum mainnet has outperformed Solana by a significant margin. This trend hasn't changed in 3 years. Since 2021, Ethereum's TVL has consistently held at 50%–60%. While Ethereum's TVL dominance has slipped recently, it still sits around 60% when you include Base/Arbitrum TVL — on par with the 2021 bull run. That said, Solana's TVL has now reached 8%, up sharply from the 1%–3% range it occupied in 2021.
2. Developers
The Ethereum ecosystem's total developer count still leads. Ethereum mainnet developers declined slightly over the past year, likely due to some migration to L2s like Base. From 2022 to 2023, Ethereum builders treated "credible neutrality" as the highest badge of honor, and Ethereum investors fixated on it. During that period, I constantly heard projects hype how ZK would change everything. It stung, but we have to acknowledge that ZK is gradually gaining real traction.
That said, I've watched Solana's developer ecosystem surge over the past year. This was especially visible at this year's Breakpoint 2049 conference — it had been a long time since I'd seen a major public chain pull off such a successful ecosystem event. On top of that, the Solana Foundation has been running hackathon houses nonstop, and many teams I know are making the move to Solana. This aligns with the 20% year-over-year growth in Solana developers.
3. Users
Solana leads by a wide margin. Solana's daily active users (58%) outnumber Ethereum and all L2s combined (42%). On top of that, Solana's transaction volume (84%) dwarfs the entire Ethereum ecosystem (16%). However, core asset transaction volume remains concentrated on Ethereum and L2s — a reflection of Solana's current positioning as a "shitcoin launchpad." Also worth watching: Base, which is growing fast.
4. Ecosystem "Shitcoin" Asset Valuations
Shitcoin asset valuations are a byproduct of public chain ecosystem development. In 2020–2021, Ethereum's DeFi/NFT ecosystem led to all DeFi and NFT assets being priced in ETH. People talked about DeFi TVL in terms of thousands or tens of thousands of ETH, or an NFT costing a few ETH. Whether it was real buying pressure or psychological anchoring, the native coin became the ultimate pricing standard — even if those coins ultimately got dumped through a game of hot potato.
Looking back at BSC's GameFi era in 2021, users also used BNB as the primary price anchor for ecosystem assets, which drove BNB's explosion during the GameFi boom.
More recently, I believe Solana's biggest win has been @pumpdotfun's relentless use of their three-part playbook to launch new shitcoins — using various methods to slice tokens and manufacture hype — alongside get-rich mythology for a small slice of users, which has attracted significant buying pressure across the entire market. Over time, users have developed a mental model of pricing shitcoin assets in SOL. By the way, http://pump.fun is one of the three most perfectly executed products I've seen in the past three years (the top two will be discussed separately).
5. Has the Ethereum Ecosystem Underperformed Over the Past Two to Three Years?
Quite the opposite — I think it's developed well. Over the past three years, I've considered @ethena_labs and @Polymarket to be the 1st and 2nd most perfectly executed products. Ethena brought funding rate arbitrage yields — previously accessible only to a select few — to decentralized retail investors. In 251 days, the protocol generated $100 million in revenue, benefiting the project, investors, and users alike, though the token utility design has some issues.
Polymarket brought prediction markets onto the blockchain, leveraging crypto to solve the settlement and regulatory headaches of traditional prediction markets. Capitalizing on the election cycle, it has become a go-to data source for global media platforms covering election results and polling data.
Ethena delivers a real-world yield scenario for return-seeking users, while Polymarket solves genuine problems in the web2 prediction market space and has successfully crossed into mainstream awareness. Both have added new wins to the Ethereum ecosystem. But what have they actually done for ETH's price? These two successful trailblazers have struggled to recreate the magic of the DeFi/NFT era, when Ethereum served as the ecosystem's pricing anchor.
6. The Ecosystem's Future Outlook
I believe future ecosystems will split into two distinct lanes. One lane covers real-world use cases — like @ethena_labs and @Polymarket — which will continue to lean on Ethereum. We need more PayFi initiatives on Solana, since that's the genuine long-term direction for blockchain development. At the same time, we need projects like @pumpdotfun that use these mechanisms to drive core asset prices higher. Both are necessary.