30+ legacy functions

Compatibility Functions

Legacy versions of functions kept in Excel for backward compatibility. Old NORMDIST, TDIST, RANK, and dozens of other pre-Excel-2010 forms. Still functional — but you should migrate.

💡 Quick answer: If you're on Excel 2010 or newer and building new work — use the modern versions (NORM.DIST, T.DIST, RANK.EQ, etc). If you're maintaining an older workbook or need to share with Excel 2007 or earlier users — the legacy versions still work identically.

In Excel 2010, Microsoft renamed and split many statistical functions to be more accurate and consistent. The old names (called "compatibility functions") stayed in Excel so old workbooks would keep working. The new names are the recommended future. Both produce identical results in most cases, but the new versions have more precise algorithms for edge cases.

Old name → new name migration

Old (Compatibility) New (Recommended) Notes
NORMDIST NORM.DIST Normal distribution — mathematically identical
NORMSDIST NORM.S.DIST Standard normal — new version adds cumulative arg
TDIST T.DIST / T.DIST.2T / T.DIST.RT T distribution — split into three clear variants
CHIDIST CHISQ.DIST.RT Chi-squared — renamed to be explicit about "right tail"
FDIST F.DIST.RT F distribution
RANK RANK.EQ / RANK.AVG Split into "same rank for ties" (EQ) vs "average rank" (AVG)
PERCENTILE PERCENTILE.INC / PERCENTILE.EXC Now includes/excludes endpoints explicitly
QUARTILE QUARTILE.INC / QUARTILE.EXC Same split as PERCENTILE
STDEV STDEV.S / STDEV.P Explicit sample (.S) vs population (.P)
VAR VAR.S / VAR.P Same sample-vs-population split
COVAR COVARIANCE.S / COVARIANCE.P Sample vs population covariance
MODE MODE.SNGL / MODE.MULT Single-value mode vs multi-modal array

Browse by category

Legacy Statistical Distributions

15 functions

Old versions of statistical distribution functions from Excel 2007 and earlier.

Legacy Rank & Percentile

4 functions

Ranking and percentile functions before the .EQ/.AVG and .INC/.EXC splits.

Legacy Variance & Std Dev

6 functions

Old versions that didn't distinguish sample from population.

Legacy Inverse Distributions

6 functions

Inverse of the legacy distributions — get the value at a given probability.

Legacy Tests

3 functions

Hypothesis test functions from pre-2010 Excel.

Other Legacy Functions

5 functions

Assorted older-form functions kept for backward compatibility.

Frequently asked questions

Do old and new versions give the same result?

For typical inputs, yes — they're numerically equivalent. The new versions use improved algorithms in edge cases (very small probabilities, very large numbers) so results may differ slightly at extremes. In practice, the difference is negligible for most business analysis.

Should I rewrite my old workbooks to use the new names?

Only if you have a specific reason — you're editing the workbook anyway, you're improving numerical precision on edge cases, or you're standardizing on modern syntax across your team. If old workbooks are working, "if it isn't broken, don't fix it" applies.

When would I still write NEW code using compatibility functions?

Only when sharing the workbook with users on Excel 2007 or earlier — those versions don't support the new function names. Very rare in 2026, but happens occasionally in enterprises with legacy systems.

Why did Microsoft rename these functions?

Three reasons: (1) The old names were ambiguous — RANK didn't specify tie-breaking behavior, TDIST didn't specify one/two tails. (2) The new algorithms are more numerically stable. (3) The dot notation groups related functions together for autocomplete discovery.

Will compatibility functions ever be removed?

No plans announced. Microsoft's policy has been to keep them indefinitely for backward compatibility with old workbooks. Assume they'll continue to work for the foreseeable future.

Modernize legacy workbooks.

Ask the Add-in "convert all compatibility functions in this workbook to modern versions" — get the mapped-out changes automatically.

Get the Add-in →