51% Attack and Double Spending: What Majority Hashrate Can Do

51% Attack and Double Spending: What Majority Hashrate Can Do

A 51% attack, or majority attack, is an attack on a proof-of-work blockchain in which one party controls most of the network’s hashrate, the computing power of its miners, and could use it to rewrite recent history. Double spending is spending the same coins twice; in Bitcoin, it means using one output, a piece of coin value, in two conflicting transactions.

What Is Double Spending in Bitcoin?

Double spending is spending the same coins twice. In Bitcoin, a transaction spends outputs, which are pieces of coin value that earlier transactions created. An output that nobody has spent yet is called an unspent transaction output (UTXO), and each output can be used as an input once. Bitcoin double spending is therefore an attempt to use one output in two conflicting transactions.

Nodes are the computers that connect to the Bitcoin network. A full node keeps the chain and validates blocks and transactions against the rules. Each node keeps the unconfirmed transactions it has seen in a waiting pool of its own, called the mempool, so until a block records one of two conflicting transactions, different nodes can hold different ones.

When two transactions try to spend the same output, only one of them ends up recorded on the blockchain. The one that lands in a block on the chain the network follows settles the conflict. With less than half of the hashrate, a double spend is less likely to succeed. A 51% attack shifts the decision about which chain the network follows toward the party that holds the majority.

What Is a 51% Attack, and How Does It Rewrite Recent History?

A 51% attack is an attack in which one party holds more than half of a proof-of-work network’s hashrate. The name “51%” is shorthand for that majority, and the share does not have to be exactly 51%. The same attack is also written as a 51 percent attack, and the party carrying it out is an attacker with a majority of the hashrate.

Hashrate is the combined computing power that miners spend on proof of work, measured in hashes per second. Proof of work asks miners to find a hash below a target value. On average, that can only be done by brute-force computation, so a valid block shows that the work was done.

Nodes are designed to follow the chain with the most work behind it. In the Bitcoin whitepaper’s reasoning, honest miners who hold most of the hashrate build a chain that tends to grow fastest and to outpace competing chains. A party with more than half of the hashrate would be expected to grow a competing chain faster than the rest of the network, and over enough blocks that chain would overtake the main chain.

A chain reorganization, often shortened to reorg, is what happens when a node switches to a competing chain that has more work. The blocks the node leaves behind stop being part of its best chain, and they are called stale blocks. A party with a majority of the hashrate could use a reorganization to replace recent blocks with its own version, which is how it could reverse its own recent transactions and carry out a double spend.

Why Do More Confirmations Make a Reversal Harder?

A confirmation score counts how many blocks on the chain with the most work would have to be modified to remove or change a transaction. A transaction with a score of one or higher is confirmed, and the score is also called the number of confirmations. A transaction that sits in no block yet is unconfirmed.

Changing a block means changing every block after it, and the work to do so grows as more blocks are added on top. For a 51% attack, that means a competing chain would have to redo the block that holds the transaction and each block after it before the network would follow that chain. Each added block is more work to redo, so a reversal gets harder as the number of confirmations rises.

A harder reversal is not a ruled-out reversal. Confirmations are designed to make a reversal cost more work, and no number of confirmations removes the possibility of a reversal entirely.

What Could a Majority of the Hashrate Do, and What Are Nodes That Validate Blocks Designed to Reject?

A party with a majority of the hashrate could change which recent blocks the network follows, and the rules that nodes enforce still limit what else it could do.

A party with a majority of the hashrate could, in principle:

  • Revise recent transaction history by building a competing chain that replaces recent blocks.
  • Reverse its own recent payments and take back coins it spent while it held the majority, which is the double spending case.
  • Leave some or all new transactions out of the blocks it mines (some documentation describes this as preventing them from confirming).

Nodes that validate blocks are designed to reject each of the following, even when a party holds a majority of the hashrate:

  • A block that creates BTC beyond what the rules allow, because the rules limit what a block may claim as its block reward and a node that validates blocks checks that limit itself.
  • A spend of coins that the party has no private key for, because a spend needs a valid signature from the key that controls the output, and a node that validates blocks is designed to reject spends that fail the check.
  • Any other block that breaks the rules, because a node that validates blocks checks the rules itself, so such a block is designed to be rejected even with the majority’s work behind it.

Light clients rely on full nodes and on the chain with the most work instead of checking every block themselves, so the limits above apply to them differently.

By definition, a 51% attack concerns which chain of blocks the network follows. It is a different kind of event from compromising a wallet or the signature scheme. Changes to the rules themselves are a separate topic, covered under soft fork and hard fork.

How Large Is the Network’s Hashrate?

A 51% attack on Bitcoin would need more than half of all the hashrate on the network. As of October 2026, the network’s hashrate is about 1,080 exahashes per second (EH/s), measured as a seven-day average and estimated. One exahash per second is a quintillion hashes per second, so a majority would mean more than half of that figure.

Hashrate moves from day to day, so a majority means more than half of whatever the total is at that moment. The share a party would need follows the total, but the rule stays the same.

Network difficulty is the measure of how hard it is to find a valid block. Bitcoin adjusts it so that blocks keep arriving about every 10 minutes on average.

A Made-Up Example: What Does 60% Versus 40% of the Hashrate Mean?

Take a made-up network in which one party holds 60% of the hashrate and the rest of the network holds 40%. These numbers are invented for illustration and do not describe a real network. On average, 6 of every 10 blocks found would come from the 60% party and 4 from the rest. A competing chain built by the 60% party would therefore grow faster than the main chain, at a ratio of 1.5, which is the pattern a double spend by a majority relies on. Block finding is a matter of chance, so these figures are averages, and any single run of 10 blocks can split differently.

If the roles were reversed, with the party holding 40% and the rest holding 60%, the party’s chain would grow more slowly on average. In the whitepaper’s model, when honest nodes hold the majority, an attacker’s chance of catching up falls quickly as the number of blocks to catch up grows. The model is a calculation, not a guarantee.

November 18, 2026

GoMining News

You might be interested in: