F.DIST.RT and FDIST in ONLYOFFICE Spreadsheet Editor

• Introduced: Original introduction release is not established by the sources checked.
• Available: Documented in current unversioned ONLYOFFICE function help, checked October 18, 2026.
• Implementation inspected: Official SDKJS v10.0.0.169, commit 3679e0375a758d7dc59d31337a45461be3525ed8. This establishes a source snapshot, not identical behavior in every installed editor.
• Retired or deprecated: Retirement or formal deprecation is not established by these sources. A legacy spelling is not automatically a removed function.
• Testing: Examples are independently calculated or labeled source predictions. No ONLYOFFICE application or calculation-engine execution was performed. No native screenshot or general import/export guarantee is claimed.

Use English names and comma separators as shown; adapt separators outside quoted text if your locale requires semicolons. Keep full precision in calculations and round only the displayed summary.

=F.DIST.RT(x, deg_freedom1, deg_freedom2)

The function accepts exactly 3 arguments. The two degrees of freedom are ordered: exchanging them while leaving x unchanged generally changes the answer. For a ratio of independent sample variances, associate each sample’s n−1 with the variance’s position in the ratio. Other F statistics can arise from different models, so do not assume every F value comes from a two-sample variance calculation.

The output is the right-tail probability at or beyond x. It is a single-tail area. F.TEST uses a two-sided construction and should not be replaced by this result without the necessary tail adjustment.

1. Put 4 in A2 and 2 in each of B2 and C2.
2. Enter =F.DIST.RT(A2,B2,C2) in D2.
3. The independent expected value is .2.
4. Label the cell right-tail probability to preserve the interpretation.

With numerator and denominator df both 2, the distribution has simple rational forms:

left cumulative = x/(1+x)
right tail = 1/(1+x)
density = 1/(1+x)^2

At x=4 these are 4/5=0.8, 1/5=0.2 and 1/25=0.04. The calculation supplies an independent check without calling ONLYOFFICE’s distribution engine. It also makes the density/tail distinction explicit: 0.04 and 0.2 are not alternative rounded versions of the same quantity.

Keep x=4, but use numerator df=2 and denominator df=4. The right tail is (4/(4+2*x))^2, which at x=4 is 1/9 = 0.111111111111. The corresponding left cumulative probability is 8/9 = 0.888888888889.

Now exchange the degrees of freedom to numerator 4 and denominator 2 while keeping x=4. The right tail becomes 17/81 = 0.209876543210, not 1/9. This shows why argument order matters even when the same two df numbers appear in the formula.

If you invert the statistic as well as exchanging the degrees of freedom, the reciprocal relationship connects complementary tails. Specifically, for positive x, the right tail at x with df1/df2 equals the left cumulative probability at 1/x with df2/df1. Changing only one part is not the same transformation.

For df=2 and 2, an upper-tail probability of 0.05 corresponds to x=19, because 1/(1+19)=0.05. F.INV.RT solves this inverse problem. FDIST is the legacy name for the right-tail probability, while FINV is the legacy right-tail inverse; neither denotes the left CDF or density.

A computed tail can be used as part of a model-based significance calculation, but the function itself does not validate normality, independence or how the F statistic was produced. A p-value is not the probability that variances are equal, nor a measure of how large a practical difference is.

The inspected distribution functions reject negative x, either df below 1, or either df at or above 10^10, using the numeric-error category, conventionally #NUM!. The strict upper boundary is worth distinguishing from help wording that may describe an inclusive limit. Values outside the intended domain are not valid inputs simply because another product accepts them.

Unlike the inspected F inverse wrappers, these distribution wrappers do not parse degrees of freedom as integers before passing them to the numerical helper. Do not assume that a fractional df is truncated consistently across F.DIST and F.INV. The examples use integer df, and no general cross-function fractional-df round-trip guarantee is made.

Nonnumeric conversions and unusable helper results can produce other errors. These are source-based boundaries, not captured application displays. Use explicit logical selectors where required and clean scalar inputs; array spills and imported type coercions have not been verified here.

These are independent mathematical expectations or identified source-error predictions, not native application results. Preserve full input precision when reproducing them.

The legacy name FDIST is covered in this guide. FDIST inherits the modern right-tail calculation method. This source relationship does not establish identical metadata, file serialization or every imported-workbook behavior. Legacy does not by itself mean formally deprecated or removed.

Use the exact probability direction, parameter order and data meaning above when moving a workbook between products. Similar names are not a general cross-version or file-format compatibility guarantee.

COUNT helps check numeric sample sizes; SQRT and POWER expose intermediate arithmetic in a statistical model.

• Official F.DIST.RT help, checked October 18, 2026.
• Pinned SDKJS implementation, including parameter validation and numerical helpers.
• Official FDIST help.