When examining the ecosystems developed around these two chains, it's important to consider both the number of projects and protocols launched on each chain, as well as the number of additional chains launched within their respective stacks. Arbitrum Orbit highlights the potential for launching L3 chains settling on Arbitrum, whereas the OP Stack enables projects to launch L2s united in the so-called ‘Superchain’ and settle directly on Ethereum.

Number of projects

As of March 23rd, 2024, DefiLlama reported 566 protocols built on Arbitrum, 223 on Optimism and 238 on Base. Even summing up Optimism and Base, Arbitrum still leads in the number of protocols and apps built on it.

Arbitrum Orbit vs OP Stack

A major winning point for Optimism (actually, the OP Stack) is the number and quality of projects launching an L2 chain, which drives additional monetary and branding value to the OP token. While Base and other L2 chains do not utilize OP as their native gas token, the revenue generated by their sequencers is distributed to the Optimism Collective to support OP RPG Funds. Notably, Base has committed to allocating 15% of its revenue to the Optimism Collective. It’s interesting to note that while Arbitrum hosts more projects in terms of apps and protocols, the OP Stack has attracted more projects that are willing to launch a semi-independent chain.

As this table from L2beat shows, most of the top optimistic rollups are built on the OP Stack. Source: https://l2beat.com/scaling/summary

Looking at the revenue generated by the OP Stack (Optimism and Base), it closely mirrors that of Arbitrum. While Base may have diverted a portion of transactions from Optimism (as evidenced by the Daily Transactions Volume section), it's reasonable to claim that Base had a positive overall impact on the revenue of the OP Stack.

Prior to the launch of Dencun, Arbitrum maintained a significant margin advantage, implying that the costs associated with Optimism and Base were higher than those of Arbitrum (Optimism even experienced several days of negative margins). However, following the deployment of Dencun and ArbOS (both of which had the effect of lowering fees), this trend changed: Arbitrum's L2 fees (and subsequently its revenue) notably decreased, thereby reducing the margin (calculated as L2 fees minus L1 costs) and margin percentage (calculated as (L2 fees minus L1 costs) divided by L2 fees).