Litecoin.watch · Research & practical tools
Observed Litecoin confirmation times
A rolling study from our own Litecoin Core node: first mempool observation, first observed block inclusion, and every incomplete outcome.
Collecting observations
The observer is reading the node every 15 seconds. Existing mempool transactions at startup or after a gap are excluded from new timing measurements.
- Latest collector update
- 5 Oct 2026, 11:35 UTC
- Validated height / known headers
- 3,189,972 / 3,189,972
- Core version
- /LitecoinCore:0.21.5.8/
- Connected peers
- 10
Node values last read successfully: 5 Oct 2026, 11:35 UTC.
This page is a snapshot; reload to get a new update. Initial synchronization can take hours or days. A node's header count is its current view, not independent proof of the network's highest height.
Every sampled outcome counts
Rolling seven-day cohort, selected by first observation time. A transaction qualifies when the first eight hexadecimal characters of its txid, interpreted as an integer, are divisible by 16. This fixed rule selects roughly one in sixteen eligible txids.
| Outcome | Transactions | Meaning |
|---|---|---|
| Observed in a block | 40,640 | An eligible mempool observation was matched to the next observed block. |
| Still in the mempool | 9 | No block inclusion has been observed yet. The waiting time remains open. |
| Left the mempool without a matched block | 38 | Could be eviction, replacement, a conflict or a missed inclusion. The cause is not inferred. |
| Interrupted by a collection gap | 3,068 | Continuity was lost before an eligible confirmation observation. |
| Block membership awaiting recheck | 0 | Prior results are withheld while their confirming blocks are checked after an interruption or chain change. |
| Confirming block no longer canonical | 21 | The original observed confirmation is excluded from timing summaries. |
| Total sampled transactions | 43,776 | Unconfirmed and interrupted outcomes remain in the denominator. |
The CSV refreshes about once a minute and may lag this table. UTC timestamps are Unix seconds. It retains original observed inclusion details even when a record is later excluded; use the state column when analysing it.
Time to the first observed block inclusion
This distribution includes only 40,640 currently matched confirmations. It excludes pending and interrupted outcomes listed above, so it cannot estimate how long every payment will take. Very fast transactions that enter a block between polls may never enter the cohort.
| Observed duration | Confirmed sample count |
|---|---|
| 0–60 seconds | 10429 |
| 61–150 seconds | 13847 |
| 151–300 seconds | 10790 |
| 301–600 seconds | 4872 |
| Over 600 seconds | 702 |
Among these confirmed observations: median 120.0 seconds; 90th percentile 360.0 seconds. These conditional figures omit open and censored waits.
Collection coverage
Only consecutive valid snapshots at most 30 seconds apart contribute observed time. Syncing, disconnected periods, restarts, inconsistent reads and jumps over more than one block create gaps. The table includes the current partial UTC day; the oldest day can also be partial at the rolling window boundary.
| UTC date | Recorded polls | Covered hours | Excluded baseline entries |
|---|---|---|---|
| 2026-09-28 | 2,931 | 0.00 | 0 |
| 2026-09-29 | 5,604 | 0.00 | 0 |
| 2026-09-30 | 5,598 | 0.00 | 0 |
| 2026-10-01 | 5,753 | 18.89 | 728 |
| 2026-10-02 | 5,758 | 23.78 | 1,154 |
| 2026-10-03 | 5,759 | 23.81 | 923 |
| 2026-10-04 | 5,758 | 23.81 | 1,364 |
| 2026-10-05 | 2,782 | 11.55 | 199 |
Baseline exclusions count mempool entries at each new baseline, not unique transactions. A full day has 24 hours. Intervals are assigned to the day in which their ending poll occurs, with up to 30 seconds of boundary imprecision.
The first 30-day study
Fixed cohort period: 2 Oct 2026 through 31 Oct 2026, UTC. 3 of 30 calendar days have ended. A day with gaps stays in the period; it is not replaced with a better day.
The report is not available yet. Its calculation runs only after the fixed period ends and all 30 daily files exist. No forecast or synthetic data fills the waiting period.
The live table covers seven days; the working database retains 90 days. Closed UTC days receive immutable ZIP exports containing raw polls, a retained-cohort snapshot, the actual export time and SHA-256 checksums. A daily snapshot can repeat earlier transactions. Use the last snapshot for outcomes and each day's polls for coverage; do not add the cohort files together.
| UTC day | Covered hours | Export |
|---|---|---|
| 2026-10-04 | 23.81 | Download ZIP |
| 2026-10-03 | 23.81 | Download ZIP |
| 2026-10-02 | 23.78 | Download ZIP |
| 2026-10-01 | 18.89 | Download ZIP |
| 2026-09-30 | 0.00 | Download ZIP |
| 2026-09-29 | 0.00 | Download ZIP |
| 2026-09-28 | 0.00 | Download ZIP |
| 2026-09-27 | 0.00 | Download ZIP |
An export records states when it was written, not a reconstruction of midnight. Later reorganizations can change confirmations; frozen reports are historical observations. Files are held on this server, not described as an independent off-site backup.
How to read this study
Our private, pruned mainnet node has no wallet. Polling measures when this observer detects a transaction and a block, not when a sender broadcast it or a miner found the block. Each endpoint can lag its event by a polling interval; relay delays and the node's mempool policy also affect the sample. This is one observer, with no guarantee that another node sees the same mempool.
Only transactions exposed by Core's public mempool and block transaction lists are eligible. Private MWEB transfers are not measured individually. Reappearing sampled txids keep their first record within the 90-day retention period; a lost observation is not restarted as a shorter wait. No conclusions about fee recommendations, exchange crediting, attack probability or network-wide payment speed follow from this chart alone.
Confirmation study analysis
Final analysis is waiting for the fixed 30-day observation period. The clock has started; missing collection time does not extend or restart the period. No timing result is inferred from synchronization progress.
Current rolling cohort: 43,776 sampled transactions, 40,640 matched confirmations. Confirmed-only quantiles are withheld below 20 matched confirmations. The final report uses its own fixed cohort, rather than adding overlapping rolling summaries.
What we will and will not infer
The measured delay begins when this node first sees a sampled transaction and ends at its first matched block observation. It is not the sender’s broadcast time. Fifteen-second polling adds observation uncertainty. A merchant’s confirmation threshold and receiving-service credit are separate events.
Pending, disappeared, gap-censored, reorg-censored and recheck-required outcomes stay in the denominator. Gaps and synchronization are not replaced by the 150-second target interval. A confirmed-only median can look favorable when difficult waits remain unresolved; the coverage and full outcome table must be read together. The data do not justify a network-wide fee recommendation or a promise of finality.