Forced Lightning Closures Cost Median 925 Satoshis

A preprint by Saurabh Kumar of Bitcoin Data Labs measures the cost and delay of unilateral Bitcoin Lightning channel closures.

Updated on Nov. 15, 2026 in Data Centers

Forced Lightning Closures Cost Median 925 Satoshis

A preprint published November 12 by Saurabh Kumar of Bitcoin Data Labs reports that forced Lightning channel closures in 2026 cost a median 925 satoshis per channel across both parties. The party initiating a forced closure waited a median 2.5 days before it could reclaim its own balance.

Why it matters

Forced closures cost roughly 4.7 times more than cooperative closes, which bears directly on whether routing node operators can profitably keep channels open. The findings quantify the economics of a settlement path that users depend on when a cooperative close is not possible.

The study is a preprint analysis of on-chain channel closure transactions, with a 2026 forced-close median of 925 satoshis against a cooperative-close median of 197 satoshis, a 4.7-to-one ratio. The break-even annual return was 1.2% in 2026 and 16.7% over the full study period from 2018 onward, and the sample excludes private channels.

The players

Saurabh Kumar

Saurabh Kumar is the author of the preprint on forced Lightning channel closures, working with Bitcoin Data Labs on on-chain analysis of Bitcoin payment channels.

Bitcoin Data Labs

Bitcoin Data Labs is the organization with which Saurabh Kumar published the preprint, focused on data analysis of the Bitcoin network.

The details

A unilateral close settles a Lightning channel using a previously signed commitment transaction rather than a negotiated agreement between both parties. Under the BOLT 3 specification, the output returning the broadcaster's own balance carries a to_self_delay, which gives the peer time to penalize a broadcaster that submits a revoked channel state. The reported median wait across the whole study period is 1.7 days.

Timeline

  1. January 2018: Start of the study period covering 973,581 channel closes.

  2. October 4, 2026: The detailed 2026 cost cohort of public force closes ends.

  3. October 31, 2026: The study period covering channel closes ends.

  4. November 12, 2026: Saurabh Kumar publishes the preprint on forced Lightning closures.

Roadmap

The BOLT 3 specification's to_self_delay gives a peer time to punish a broadcaster that publishes a revoked state, and this study measures what that protection costs in practice. The 2026 figures show that the delay mechanism and forced-close fees together make unilateral settlement several times more expensive than cooperative closes.

Users who open Lightning channels should expect that a forced close, if one becomes necessary, will typically cost around 925 satoshis and take a median of 2.5 days before the initiating party can reclaim its balance. Those figures matter most for small payments, where the fee can be a large share of the amount settled.

The takeaway

Forced closures are a costly fallback, so channel operators should weigh the measured 4.7-to-one cost ratio when deciding how to manage open channels. Treat cooperative closes as the default wherever possible.

Further reading

For more on how the underlying network infrastructure is run, see Data Centers coverage.