Litecoin.watch · Research & practical tools
LTC/EUR liquidity and execution costs
Compare public order-book spreads and visible-depth execution estimates for €100, €1,000 and €10,000.
Latest captured books
Loading recorded observations…
Scheduled captures every 15 minutes; this is a saved observation, not a streaming quote. Venue responses overlap in time but are not synchronized.
| Venue / observation | Bid / ask · EUR | Spread · bps | Depth within 1% · EUR | LTC / EUR replay | VWAP · EUR/LTC | Depth impact · bps | Cost vs midpoint · bps |
|---|
Trading costs over time
Use the order size and direction above for both the latest books and this history. The chart shows saved book replays; it does not show orders that were actually submitted.
Loading execution history…
What was captured
| Venue | Valid / recorded | Full / valid | Partial / valid | Unavailable / recorded | Valid / expected 15-min slots |
|---|
A slot is a UTC quarter-hour touched by the available history window. Boundary slots may be partial. One valid replay is enough to cover a slot; extra requests do not increase coverage. The 15-minute schedule is approximate: this is capture coverage, not venue uptime or a promise that a quote lasted 15 minutes. Missing slots and unavailable responses remain missing.
Costs among complete visible-book replays
| Venue | Full replays | Midpoint cost median · bps | Midpoint cost P95 · bps | Impact median · bps | Impact P95 · bps | Fill median · % |
|---|
Quantiles use linear interpolation at (n − 1) × p, with one equal observation per valid capture. Cost and impact omit partial replays; fill includes full and partial replays. P95 is a descriptive quantile, not a confidence interval. Responses overlap in time but are not synchronized.
Each sell replay uses the selected EUR budget divided by that venue's best bid. The LTC amount changes by venue and capture. These nominal EUR cases are not a comparison of selling one fixed quantity, and the table is not a venue ranking.
The selected CSV includes every retained capture in the chosen window and direction, with full, partial and unavailable outcomes. Missing cost values stay blank. It does not insert rows for uncaptured slots.
Retention is limited to 30 days, beginning with the first collector capture. A 30-day selection cannot create history before the archive began. Gaps over 30 minutes stay open; failed responses and partial-only costs never turn into zeroes.
How to read the numbers
- Spread: (best ask − best bid) ÷ midpoint × 10,000. One basis point is 0.01%.
- Buy replay: spend the selected EUR budget through successive asks. VWAP is total EUR divided by obtained LTC.
- Sell replay: sell budget ÷ that venue’s best bid LTC. A €1,000 selection means nominal €1,000 at its best bid; the exact LTC amount differs by venue.
- Depth within 1%: summed EUR value of asks up to best ask × 1.01, or bids down to best bid × 0.99. The captured-level count describes the entire returned side, not just that band.
- Depth impact: buy VWAP ÷ best ask − 1, or 1 − sell VWAP ÷ best bid, in basis points.
- Cost vs midpoint: buy VWAP ÷ midpoint − 1, or 1 − sell VWAP ÷ midpoint. Includes the distance from midpoint to the quoted side and visible depth.
- Insufficient depth: partial amounts remain visible; whole-order VWAP/impact are withheld. An unavailable response is not zero liquidity.
We sort and aggregate positive levels and reject crossed books. Kraken and OKX request 100 levels; Coinbase level 2 returns its available aggregated book. Different depth limits can change coverage. Coinbase auction books and captures with unknown auction state are withheld from continuous-market replay statistics. Orders can disappear; hidden liquidity, queue position, latency, tick/lot rounding and price movement are excluded. These are observations, not executable offers or a venue recommendation.
Kraken level timestamps are not treated as a book-wide event time. Coinbase sequence and OKX book timestamp are retained when supplied. Request and response times are collector times. Raw downloads expose a SHA-256 response-body digest. HTTP errors, old exchange timestamps and invalid books remain failed captures.
API formats: Kraken, Coinbase Exchange, OKX.