# Leveraging lottery mechanisms to incentivize censorship resistance on Ethereum

*Disclaimer: This post should be viewed as a "fun" idea I had rather than a serious proposal; the focus is on devising mechanisms to prevent economies of scale and centralization vectors from emerging.*

‘‘All you need is 0.001 ETH and a dream’’

‘‘Buy a ticket. Contribute to Ethereum’s censorship resistance. Get a shot at winning life-changing rewards every week.’’

>💡
>
>Some key facts about lotteries:
>
>- **Lotteries are universal**. They have existed for thousands of years (e.g., ancient China in ~205–187 BCE), and enjoy global popularity.
>- Lotteries are purely chance-based, requiring no skill, giving **player equal odds per ticket**. This creates a unique situation in which social status, education, wealth, or skill provides no advantage.
>- Lotteries naturally attract broad, diverse participation, particularly **rewarding unsophisticated players** rather than a few sophisticated actors with greater resources or information.
>- Despite having negative expected value (EV)—meaning most players lose money over time—lotteries still attract **hundreds of millions of participants worldwide** each year.
>- Lottery participation generally rises during young adulthood, peaks in middle age, then declines among seniors. Men and individuals with risk-seeking personalities tend to participate more frequently. This demographic **closely aligns with those interested in blockchain and cryptocurrencies.**

In this post, I argue a lottery might be the perfect mechanism to reward participants across the world for contributing to Ethereum’s censorship resistance.

## Introduction

A known challenge when designing mechanisms for trustless, decentralized settings is rewarding small, unsophisticated, or solo participants without enabling large, resource-rich actors to dominate and capture disproportionate benefits. The thesis is the following: If a mechanism offers positive EV, large, resource-rich actors always have incentives—and often the ability—to capture and dominate rewards. Regardless of the specific activity, if there's value to extract in permissionless settings, sophisticated actors will find ways to exploit economies of scale, either by leveraging operational efficiencies, spreading out fixed costs, or employing Sybil identities. One (h/t Julian) might argue we could theoretically design positive EV mechanisms that do not introduce economies of scale at all, but I had a hard time coming up with an example in practice.

Consider Ethereum staking as an example, and imagine you want to design a mechanism to reward genuine small stakers who own and stake exactly `32 ETH`, rather than larger entities who own `2048 ETH`. In practice, large players can easily split their stake into 64 Sybil identities of `32 ETH` each, benefiting from smoother reward distributions and crowding out authentic small-stake participants. This dynamic undermines Ethereum’s decentralization goals, highlighting the inherent limitations of traditional positive-EV incentive structures.

One potential way out of this fundamental trade-off is to apply insights from behavioral economics to move beyond the assumption of purely rational actors (*homo economicus*) who solely maximize expected value. Instead, I propose transitioning from a system that relies exclusively on altruistic behavior to one that actively incentivizes decentralization and censorship resistance—without introducing centralizing forces.

- **Stage 1: Depend on altruism** or intrinsic motivation, expecting smaller participants to remain engaged and perform duties honestly despite unfavorable incentives.

*(This approach currently describes Ethereum’s solo-staking environment, but it may not be sustainable long-term.)*

- **Stage 2: Boost altruism** using negative-EV mechanisms like lotteries to leverage universal human biases and sustain or increase participation. While each ticket has the same expected value, larger and more sophisticated actors who commit more resources lose more in absolute terms—effectively reversing traditional advantages and dampening economies of scale.

Lotteries are the most simple, well-studied, and popular negative-EV mechanism out there (state lotteries return only 50–70 cents on the dollar in prizes). By drawing on key concepts from economics and game theory, researchers have explored why millions of people willingly play a "losing game," highlighting the roles of risk preferences, probability perception, and utility beyond money. Further studies indicate that lotteries attract a broad cross-section of the public, particularly middle-aged men with risk-seeking personalities. Sounds familiar? It precisely matches the demographic most interested in blockchain and cryptocurrencies. Even more fun, numerous studies highlight key behavioral design techniques and game elements—such as gamification, near-miss effects, and carefully timed draws—that significantly enhance perceived value and sustained participant engagement.

Here, I argue a lottery is particularly suited to reward includers—protocol participants in charge of upholding censorship resistance (CR) on Ethereum. Recent work has focused on mechanisms ([FOCIL EIP-7805](https://eips.ethereum.org/EIPS/eip-7805)) and proposals to unbundle includers from other protocol roles and duties (see [Attester-Includer Separation](https://ethresear.ch/t/towards-attester-includer-separation/21306)). However, existing proposals to reward includers (such as [AUCIL](https://eprint.iacr.org/2025/194)) remain complex and vulnerable to capture by larger, more sophisticated actors due to inherent economies of scale and Sybil-resistance challenges (e.g., crowd out honest includers to control the outcome of the aggregated IL). In the current version of FOCIL, the assumption is that includers will perform their duties honestly without explicit incentives—let’s see if lotteries can improve that by exploiting the asymmetry between expected value and expected utility.

## In-Protocol Lottery Design

- **Ticket System:** Each slot, participants can buy tickets for the right to become an includer. A per-slot limit is enforced just so that ticket don’t DDOS the protocol, but with a theoretically infinite amount of tickets being sold at a fixed price. Each ticket costs `0.001 ETH` (~$2 at current prices), and each purchase adds to a cumulative weekly prize pool. If not enough tickets are bought, validators will fill the gaps and take on including duties as includer seats are assigned proportionally to the number of tickets bought.
- **Inclusion Duties:** Holding a ticket places you in the includer set, obligating you to contribute to Ethereum’s censorship resistance by building and broadcasting an inclusion list to the network. When it's your turn, being selected as an includer (`16` per slot) automatically qualifies you for the weekly draw. Optionally, adding a validity and timeliness check for inclusion lists could ensure effective participation, though this adds a lot of complexity by requiring ILs to be recorded on-chain.
- **Weekly Draw:** Each week, 100% of the accumulated prize pool from ticket sales is awarded randomly to one eligible includer. Imagine `16` tickets are bought per slot with `50,400` slots weekly, the weekly jackpot—if all tickets sell—would be `806 ETH` (`16 × 50,400 × 0.001 ETH`), over `$1.5 million` at current prices.

*Fun note: The described mechanism is neutral EV but optionally, the lottery could allocate a portion, e.g., 70%, of ticket revenue to the prize pool, with the remainder burned or redistributed.) Another not so fun consideration is ensuring that large actors cannot bias the randomness (e.g., RANDAO) in their favor to increase their odds of receiving prizes.*

## Pros and Cons

### Memetics

**Pros**

**Cultural Resonance:** Lotteries have deep historical and cultural roots, making the mechanism immediately relatable and even fun for participants.

**Cons**

Leaning into lottery memetics might not be healthy for the L1, potentially reducing the focus on robust network security if the narrative becomes too superficial.

Interestingly, one could argue that nation-states have used lotteries [under their own names](https://en.wikipedia.org/wiki/Lotteries_by_country), and embraced the associated memetics for centuries. On the flip side, lotteries are subject to strict regulations in many jurisdictions, and explicit on-chain lotteries could attract unwanted legal attention and impose additional compliance burdens on protocol participants.

### From rational choice theory to behavioral economics

**Pros**

Most blockchain economic models assume that all protocol participants are purely rational. By moving towards behavioral economics, we can overcome the limitations imposed by seemingly unavoidable trade-offs and design mechanisms that effectively leverage inherent human biases, thereby bridging the gap between theoretical models and real-world behavior.

**Cons**

Incorporating behavioral economics introduces additional variables and uncertainty, as human behavior is highly context-dependent compared to the clear-cut assumptions of rational choice models. Moreover, the overall system might become significantly more complex if the underlying assumptions change for each mechanism, resulting in a system that is less coherent as a whole.

### Rewarding honest participation

The simplest form of an in-protocol lottery design automatically makes ticket holders eligible for the prize pool. However, this approach would likely result in low-quality inclusion lists, as there would be no incentive for participants to build and broadcast “good” ILs, or any IL at all. An alternative would be to track the ILs submitted by includers and deem them eligible only if they submit valid, timely ILs when it is their turn to participate. This, however, would require recording individual ILs on-chain, introducing significant complexity and some inefficiencies compared to current approaches like FOCIL, which don’t require on-chain storage (ILs are only broadcast over the CL p2p network).

### **Designing** against economies of scale

The main thesis of this post is that any “positive EV” mechanism rewarding participants in a permissionless setting can be exploited by large, resource-rich actors, who end up crowding out smaller, less sophisticated participants. This outcome appears inevitable in practice, yet in theory, it might be possible to design mechanisms that avoid economies of scale without necessarily resorting to behavioral economics.

### In or out-of-protocol

It is worth noting that lotteries designed to reward participation in improving Ethereum's censorship resistance don't *have* to be enshrined in the protocol. They could be run off-chain without requiring protocol changes, potentially mitigating the risk of negative perceptions associated with an L1-run lottery. However, out-of-protocol lotteries inevitably introduce additional trust assumptions and could lead to reduced participation, which is the very outcome these mechanisms are meant to avoid.