🤫 Selfish Mining Selfish Mining
A cheating strategy on Proof-of-Work blockchains where a miner secretly keeps the blocks it finds instead of sharing them right away. It builds a hidden private chain, then reveals it at the perfect moment to grab more than its fair share of rewards.
🏃 The simple version — a hidden head start
Picture a relay race where one runner secretly sprints ahead on a hidden track. The honest team keeps running fair and square. Just as they're about to reach the same point, the cheater steps out and says "I was already further ahead." Now the honest team's progress counts for nothing, and only the cheater gets credit. That's selfish mining: a miner finds a new block but doesn't announce it, quietly builds a longer private chain, and reveals it only when honest miners are about to catch up.
⛏️ How it works, step by step
- 🔍 Find, but stay quiet — The selfish miner solves the puzzle and finds a valid block, then keeps it secret instead of broadcasting it
- 🛤️ Mine in private — It keeps building on its secret block, growing a hidden chain that's ahead of the public one
- 🎉 Release at the right moment — When honest miners are about to catch up, it publishes its longer private chain
- 🗑️ Orphan the honest blocks — Because Proof-of-Work follows the longest chain, the honest blocks get discarded and that work is wasted
📌 No code is broken here. The cheater simply follows the longest-chain rule and controls when it shares its blocks. That's why selfish mining is a strategy, not a hack.
📉 Why 51% is not the magic number
For a long time, people assumed you needed more than half of all mining power to attack Bitcoin. Selfish mining changed that picture. The research found it can beat honest mining once an attacker controls roughly 25% to 33% of the network's total hash rate — far below a majority. The exact threshold depends on how fast the attacker can spread its blocks to other nodes: the better connected it is, the lower the bar.
| Assumption | What selfish mining showed |
|---|---|
| 🛡️ "Mining is safe below 51%" | A miner with only ~25%–33% can already earn an unfair share |
| 🐞 "Any attack must be a bug" | It abuses normal rules and timing — nothing is technically broken |
📜 Where it came from
The idea was introduced in a 2013 paper titled "Majority is not Enough: Bitcoin Mining is Vulnerable" by Ittay Eyal and Emin Gün Sirer at Cornell University. It later appeared in Communications of the ACM (Vol 61, No 7, 2018). The paper's headline point: Bitcoin's mining rewards are not "incentive-compatible" — meaning the system doesn't always make honest behavior the most profitable choice.
🧲 Why it matters
The real danger is centralization pressure. If a selfish pool earns more than its fair share, rational miners have a reason to join it to grab those higher rewards. As the pool grows, it gathers more power and pushes the network away from being decentralized — which is the whole point of Proof-of-Work. To date, no large-scale selfish-mining attack has been confirmed on Bitcoin; it remains mostly a research result that exposes a weakness rather than a proven real-world event.
❓ FAQ
- Is selfish mining a hack or a bug in the code?
- No. Nothing is technically broken. The cheater follows the normal rules (including the 'longest chain wins' rule) and just times when it shares its blocks. That is what makes it tricky: it abuses the rules rather than breaking them.
- Don't you need 51% of the network to attack mining?
- That is the common assumption, and selfish mining challenged it. The 2013 research showed a miner with only about 25% to 33% of the network's hash power can already earn an unfair share — well below a 51% majority.
- Has selfish mining actually happened on Bitcoin?
- No large-scale selfish-mining attack has been confirmed on Bitcoin. It is mainly a research result that shows a weakness. It still matters because it reveals why concentrated mining power is risky.