Data quality · Index levels
Index levels against a selected dealer.
Indicative index-style levels compared with a selected dealer outside the aggregation. Comparisons use indicative dealer levels. Contract specifications and trading sessions may differ.
DE40 against the reference
One sample per second across the run.
Headline measurements
Measured, not asserted.
8,904 matched samples across 6 names, one per second.
Price fidelity
0.62bp
Median midpoint deviation from the selected CFD-style index level, symbol balanced.
How is this measured?
- Definition
- Median across symbols of each symbol's median |TickerLayer mid − reference mid| / reference mid × 10,000.
- Reference
- selected CFD-style index level
- Aggregation
- Per symbol first, then the median of per-symbol medians so active names cannot dominate.
- Window
- 25 minutes, sampled every 1 s
- Universe
- indices-reference-8 v1, 6 of 8 symbols
- Samples
- 8,904 matched samples
- Method
- indices-reference-v1
Tail deviation p95
1.76bp
Within 2 bp in 75.0% of samples.
How is this measured?
- Definition
- Median across symbols of each symbol's 95th percentile absolute deviation.
- Dispersion
- Median absolute difference between the reference sources at the same sample; the natural floor of any measured deviation.
- Window
- 25 minutes, sampled every 1 s
- Universe
- indices-reference-8 v1, 6 of 8 symbols
- Samples
- 8,904 matched samples
- Method
- indices-reference-v1
Within 5 bp of the reference
94.6%
Within 2 bp in 75.0% of samples, symbol balanced.
How is this measured?
- Definition
- Share of one-second samples per symbol with absolute deviation at or under the threshold, averaged across symbols.
- Window
- 25 minutes, sampled every 1 s
- Universe
- indices-reference-8 v1, 6 of 8 symbols
- Samples
- 8,904 matched samples
- Method
- indices-reference-v1
Quoted spread, TickerLayer vs reference
5.00vs 3.01 bp
Median bid to ask spread over the run, in basis points of the midpoint. A measurement, not a ranking.
How is this measured?
- Definition
- (ask − bid) / mid × 10,000 per sample, median over the run, for TickerLayer and for the reference.
- Window
- 25 minutes, sampled every 1 s
- Universe
- indices-reference-8 v1, 6 of 8 symbols
- Samples
- 8,904 matched samples
- Method
- indices-reference-v1
Distribution
Where the deviation actually lands.
Every matched sample by absolute deviation. Threshold shares are symbol balanced.
Distribution of absolute deviation
Share of all matched samples in basis points.
- under 0.5 bp41.0%
- 0.5 to 1 bp18.7%
- 1 to 2 bp15.3%
- 2 to 5 bp19.5%
- over 5 bp5.4%
Threshold alignment
Symbol balanced share of samples within each band.
- Within 1 bp59.7%
- Within 2 bp75.0%
- Within 5 bp94.6%
- p99 absolute deviation5.63 bp
Universe detail
Every symbol in the universe.
One row per name in the benchmark universe, with the figures a customer would compute from the same feed.
| Symbol | Samples | Median dev bp | p95 bp | Signed bp | Within 1 bp | Within 2 bp | Spread bp |
|---|---|---|---|---|---|---|---|
| DE40 | 1,484 | 0.19 | 0.79 | +0.00 | 97.1% | 100.0% | 5.00 |
| FR40 | 1,484 | 0.32 | 0.94 | -0.17 | 95.9% | 99.7% | 4.99 |
| EU50 | 1,484 | 0.43 | 1.55 | -0.05 | 78.8% | 98.7% | 5.01 |
| JP225 | 1,484 | 0.82 | 1.97 | -0.79 | 59.2% | 96.1% | 5.00 |
| HK33 | 1,484 | 2.09 | 2.70 | -2.09 | 27.0% | 44.5% | 5.00 |
| AU200 | 1,484 | 3.04 | 5.64 | -1.90 | 0.2% | 11.3% | 5.00 |
Freshness and consistency
Delivery, agreement, reproducibility.
Timing is kept apart from price fidelity on purpose.
Update cadence
747ms
Median gap between consecutive frames per symbol on the public stream.
REST vs WebSocket exact
99.2%
131 checks, midpoint gap p95 0.00 bp.
Settled bars reproducible
100.0%
69 one-minute bars re-fetched later; close unchanged in 100.0%.
Reference model
Selected CFD-style index level
Index levels are compared with the indicative index levels of one selected CFD dealer outside the TickerLayer aggregation.
Universe
Eight indices whose dealer reference is outside the aggregation.
Reference
Indicative index levels from a selected dealer. Contract specifications and sessions may differ.
Session
Cash-session hours differ per index; signed deviation shows any basis.
Benchmark run
One run behind every figure.
We keep measuring and publish each run as it lands.
- Benchmark run
- TLQ-INDICES-20260910-1018
- Methodology
- indices-reference-v1
- Universe
- indices-reference-8 v1, 6 of 8 measured
- Observation window
- Sep 10, 2026, 10:18 to 10:43 UTC
- Window length
- 25 min, sampled every 1 s
- Matched samples
- 8,904
- Published
- Sep 10, 2026, 10:43 UTC
Benchmark results describe observed behaviour under the stated methodology during the published window. They are not a guarantee of future performance and not a service level agreement unless separately agreed in writing. These comparisons evaluate TickerLayer quotes against selected market references during the stated observation window. They do not imply exchange affiliation, endorsement, or certification.
Frequently asked questions
How are index levels benchmarked?
Each level is compared once per second with the indicative level of a selected dealer outside the aggregation, and the deviation is published in basis points.
Can an index be compared with its ETF?
Not as a quality measure. The ETF trades with its own spread, premium and session, so the difference is a basis, not an error.
Which indices are in the universe?
Eight benchmarks across Europe, Asia-Pacific and the US small-cap segment, fixed and versioned, each with a selected dealer level available as reference.
Is the reference an official index publisher value?
No. Both sides are indicative index-style levels; the reference is a selected dealer outside the aggregation.
How fresh are index levels?
The observer samples available dealer quotes approximately once per second during the cash session and reports agreement between REST and WebSocket on the same run.
Are the numbers a guarantee?
No. They describe observed behaviour during the published run and are not a service level agreement.