Satoshi Nakamoto introduced Bitcoin as A peer-to-peer electronic cash system. The essence of Bitcoin is a shared record of all transactions that have occurred on the Blockchain, maintained through proof of work by Bitcoin miners.
Bitcoin mining is the spending of resources in the form of electricity and computing power to find a predetermined nonce (Number used once). This value allows a miner to propose a block of transactions and receive Bitcoin Tokens (BTC) and the transaction fees of those transactions as a reward.
The longest chain rule is a rule of the Bitcoin network where miners are required to follow the chain that has the most proof of work (the chain that has had the most mining work done).
In case two Bitcoin miners broadcast different versions of the block, other miners will prioritize the block they received first and start working on that one. The tie will be broken on the next block, as a new block is suggested, making one of the 2 chain versions the longest.
Difficulty adjustment is a routine adjustment of the mining difficulty done roughly every 2 weeks to make sure that blocks are produced by the miners around 10 minutes apart.
Selfish mining is a theory that challenges the common Bitcoin mining practices. It is a strategy on how a miner, or a group of miners, can manipulate the Bitcoin mining rewards and increase their share.
The concept was first introduced in a paper by Ittay Eyal and Emin Gün Sirer titled Majority is not Enough: Bitcoin Mining is Vulnerable. The Key idea in the paper is that a miner can make their competitors waste their work by refraining from broadcasting solved blocks, and cleverly releasing them to maximize their rewards.
To understand selfish mining, we need to think about it as a strategic game where the selfish miner makes decisions based on how far ahead or behind they are.
When a selfish miner finds a block, they don’t immediately broadcast it to the network. Instead, they keep it private and continue mining on top of their secret chain. The decision of when to reveal blocks depends on what happens next:
Scenario 1: Selfish miner is 1 block ahead
The selfish pool might mine block 100, but keeps it secret. They start working on block 101 while honest miners waste computational power trying to solve block 100.
If the selfish miner finds block 101 first, they now have a 2-block lead. They can reveal block 100, causing honest miners to abandon their work and switch to it, while the selfish miner continues mining on their private block to find block 102.
If an honest miner finds their version of block 100 first, the selfish miner immediately broadcasts their own block 100, creating a tie. If the selfish miner has good network connectivity, their block might reach more nodes first, and the tie would be broken in their favor.
Scenario 2: Selfish miner is 2+ blocks ahead
This is the comfortable position. The selfish miner can reveal just enough blocks to stay ahead of the public chain while keeping the rest private. If honest miners catch up by one block, the selfish miner reveals another block to maintain their lead.
The key insight: By keeping blocks private and revealing them strategically, selfish miners force honest miners to waste work on blocks that will eventually be orphaned (abandoned). This wasted work means the honest miners’ effective hashpower is reduced, while the selfish miner’s remains fully productive.
In honest mining, if you control 30% of the network’s hashpower, you’d expect to earn 30% of the block rewards over time. This is your “fair share.”
With selfish mining, that same 30% hashpower can earn you 31%, 32%, or even more of the total rewards. This is done by making other network participants waste their hash power, and thus increasing the ratio of blocks solved by the selfish Miner.
The original Eyal-Sirer paper shows that selfish mining becomes profitable above ~33% hashrate (the miner has no network advantage); lower (~25-30%) with better propagation (better connectivity in the network for faster block propagation).
Let’s trace through the economics with a concrete example. Imagine there are 100 blocks to be mined in a certain time:
Honest mining with 30% hashpower:
You mine 30 blocks, earn 30% of the rewards.
Other miners mine 70 blocks, earn 70% the rewards.
Selfish mining with 30% hashpower:
You mine 30 blocks, but through strategic withholding, you cause honest miners to waste work
Perhaps 5 of the honest miners’ blocks get orphaned because you reveal your private chain
Final tally: You earn 30% of rewards, others earn only 65% of rewards
Your share: 30/95 = 31.6% instead of 30%
The honest miners still did the computational work for those 5 orphaned blocks—they burned electricity and time—but they received nothing for it. That wasted effort is your gain.
While selfish mining is theoretically profitable, several factors have prevented it from becoming a widespread problem in Bitcoin.
Why Haven’t We Seen Major Attacks?
Detection Risk: Selfish mining creates detectable patterns. A sudden increase in orphaned blocks, especially if they’re consistently losing to a certain mining pool, would raise red flags in the community.
Community Response: The Bitcoin community has demonstrated a willingness to respond to threats. If selfish mining became prevalent, the community could implement protocol changes, coordinate to exclude bad actors, or even consider a hard fork.
Profitability Margins: The paper demonstrates that profitability exists, but the actual gains are relatively small—perhaps a few percentage points. Given the operational complexity and risks involved, the risk-adjusted return might not be worth it.
Market Impact: If a successful selfish mining attack were publicized, it could undermine confidence in Bitcoin, potentially crashing the price. A mining pool earning 32% of blocks instead of 30% doesn’t help much if the value of each block drops by 20%.
Bitcoin’s security model is more complex and nuanced than initially understood. The assumption that honest mining is always the most profitable strategy turned out to be false.
The selfish mining attack works by forcing honest miners to waste computational work on blocks that get abandoned by taking advantage of the longest chain rule.
Despite its theoretical profitability, we haven’t seen real-world selfish mining due to detection risk, reputational concerns, and practical challenges.
The Bitcoin network has naturally developed some defenses through better connectivity and block propagation.
All in all, this shows that no matter how robust a network is in security and incentives, there is always an imperfection. This has prompted multiple collective efforts to keep researching ways to improve the Bitcoin network, and also build guardrails to potential issues before they might even occur.
To better understand selfish mining, I suggest you read the original paper and additional work done in the area.
Till Next time.