Forex Spread Statistics Calculator
Summarize 2–500 user-recorded spread observations in one consistent pips or points unit. The calculator reports mean, median, range, sample dispersion and explicit R7 percentiles, then compares the sample with your own reference threshold without grading a broker or claiming live conditions.
Enter one consistent spread sample
Use one symbol, account, price basis and observation method. Convert every row to the same selected unit before entry.
Observed spread summary
Entered Trading Infrastructure Observations 1.0.0
On smaller screens, scroll horizontally to inspect the audit notes.
| Audit item | Entered or derived basis |
|---|
How the spread sample is summarized
Sample SD = square root of [sum of squared deviations ÷ (n − 1)]
R7 percentile location = p × (n − 1), with linear interpolation between ordered observations
The page accepts 2–500 nonnegative values separated by commas, spaces, semicolons or new lines. Duplicate observations are retained because repeated measurements are part of the entered sample rather than data-cleaning errors.
Minimum and maximum show the observed range, while the arithmetic mean uses every value. The median is the middle R7 percentile and can remain less sensitive than the mean to a small number of unusually wide observations.
The sample standard deviation uses denominator n minus one. It describes dispersion in the values supplied; it does not create a confidence interval or prove that the observations were independently or randomly sampled.
The 90th and 95th percentiles use the disclosed R7 linear-interpolation convention. NIST notes that several percentile definitions are used in practice, so another platform can give a slightly different small-sample value.
The user-entered reference divides observations into at-or-below and strictly-above counts. The page applies no universal good-spread threshold because instrument, session, account, volume, venue and market conditions can differ.
Worked example from the audited fixture
The audited fixture enters ten EUR/USD observations in pips: 0.8, 1.1, 0.9, 1.4, 2.2, 0.7, 1.0, 1.6, 0.8 and 1.2, with a user reference of 1.2 pips.
- The sample mean is 1.170 pips, median is 1.050, minimum is 0.700, maximum is 2.200 and sample standard deviation is approximately 0.460 pips.
- R7 interpolation produces a 90th percentile of 1.660 and 95th percentile of 1.930 pips. Seven observations are at or below the entered 1.2-pip reference and three are strictly above it.
Reproduce it: select “Load audited example” above. The immutable fixture is recomputed from first-principles arithmetic rather than copied from a provider result.
How to interpret the result
- Read mean and median together. A mean materially above the median can reveal that wider entered observations are pulling the average upward.
- Use the maximum only as the widest value actually recorded in this sample. It is not a worst-case spread or a cap on future quotes.
- Use R7 percentiles as order-statistic summaries, not guarantees that exactly the same proportion of future spreads will remain below them.
- State whether values are pips or platform points. A five-digit FX quote often uses a smaller point than the conventional pip, but broker symbols can vary.
- Keep bid and ask observations synchronized. Subtracting prices from different timestamps can create a number that is not the contemporaneous spread.
- Segment materially different sessions, symbols, account types or news windows instead of averaging incompatible contexts into one convenient figure.
Assumptions and limits
- No broker, quote stream, Market Watch, symbol specification, historical feed or account is connected.
- The calculator does not derive spread from Bid and Ask or convert point observations to pips; normalization must occur before entry.
- Commission, markup attribution, volume-weighted spread, depth, slippage, swaps and total monetary trading cost are excluded.
- Timestamp order is not used, so the page does not plot changes, detect spikes, label sessions or claim to monitor a live market.
- Selection bias, missing rows, irregular sampling, rounded platform displays and stale quotes can affect the summary.
- No broker, account, symbol, session, threshold, order, strategy or trade is recommended or ranked.
Which infrastructure calculator answers which question?
Batch 49 keeps cost, spread and time in separate models because one number cannot safely stand in for another. The VPS calculator normalizes entered expense, the spread calculator summarizes a same-unit price-width sample, and the latency calculator summarizes elapsed milliseconds. None converts its output into a broker, host or trading verdict.
| Tool | Primary input | Primary output | Hard boundary |
|---|---|---|---|
| VPS Cost | Entered subscription and related costs | Monthly, term, annual and allocation views | No provider or suitability verdict |
| Spread Statistics | One consistent pips or points sample | Centre, range, dispersion and R7 tails | No live quote or broker grade |
| Execution Latency | One consistent millisecond sample | Centre, range, dispersion and R7 tails | No fill, slippage or quality prediction |
Frequently asked questions
- It reports count, minimum, maximum, arithmetic mean, median, sample standard deviation and R7 90th and 95th percentiles for one entered same-unit sample.
- No. Every observation must already use the one selected unit. The page does not convert between pips, points and money.
- Version 1.0.0 uses R7 linear interpolation at zero-based location p times n minus one across ordered observations.
- Several percentile interpolation conventions are used in practice. Small samples can produce different values, so this page discloses R7 explicitly.
- It is a user-defined comparison line. The page counts observations at or below it and strictly above it without assigning a good, bad or acceptable label.
- No. It reads only values pasted into the browser and has no broker, quote, Market Watch, symbol or historical-feed connection.
- No. Spread width remains separate from commission, slippage, swaps, depth and total monetary transaction cost.
- No. Sampling method, symbol, account, session, volume and market conditions can differ, and the calculator provides no broker grade or recommendation.
Sources and methodology
- MQL5 — SymbolInfoInteger — Official example for spread in points and Ask-minus-Bid divided by symbol point size.
- Investor.gov — Foreign Currency Exchange — Official explanation of bid-ask spread as an inherent retail-forex trading cost.
- NIST — Percentiles — Primary statistical reference for ordered observations, interpolation choices and the R7 convention.
Version 1.0.0 performs deterministic local arithmetic and uploads no entered value. Sources define platform fields, financial concepts and statistical conventions; they do not verify the user’s sample or endorse a displayed result.
Move from spread description to distinct cost questions
Compare Top Forex Brokers
Infrastructure statistics do not replace current broker disclosures. Verify the exact entity, account, symbol, pricing, execution and platform terms available in your jurisdiction before opening or funding an account.
Risk and affiliate disclosure: Leveraged forex and CFD trading can result in substantial losses. Historical spread or latency observations and infrastructure cost do not predict execution or returns. These are affiliate links, so ForexMT4Indicators.com may receive compensation if you register or trade through them, at no additional cost to you. Availability and terms vary by jurisdiction and entity.