Hubble constant — published H0 estimates and their errors, 1929–2019

The Hubble constant H0 is the expansion rate of the universe. Published estimates have come down from above 500 in 1929 to about 70, and their error bars shortened a few hundredfold on the way. This chapter counts how often the errors quoted in two compilations covered the distance to a later reference value: the question of the physical-constants chapter, asked of individual measurements instead of editions.

Takeaway

In Huchra's list (1929–2010, 548 rows) the yearly median relative error halved every 58.7 years; in Table A2 (1976–2019, 163 rows) every 13.9 years.
Against Planck 2018, 25.9% of Huchra's list lies beyond twice the quoted error (normal: 4.6%); against Riess et al. 2022, 31.9%. Table A2: 25.9% and 23.9%. Under either reference the quoted errors miss more often than a normal distribution predicts.
A Student-t fitted to the z has ν ≈ 2.30–1.97 (Huchra's list): a heavy tail. The edition-pair ν of the physical-constants chapter is 1.00.
The chapter does not decide which side of the tension is right; it counts whether quoted errors covered the distance to a later reference.

At a glance

58.7 yearsH1 precision half-life, Huchra's list 1929–2010, yearly median (95% interval 38.0–128.5); Table A2: 13.9 years 25.9%normal 4.6%H3: share of Huchra's list beyond twice the quoted error against Planck 2018 (95% interval 20–33%); against Riess et al. 2022 31.9% 20.4%normal 2%H3 surprise index (|z| > 2.326) against Planck; against Riess 26.5% ν = 2.30H4: Student-t degrees of freedom fitted to the z of Huchra's list, against Planck (95% interval 1.91–2.98); against Riess 1.97

Table A2 (rows)Huchra (decade aggregates)Planck 2018Riess et al. 2022

Blue dots: rows of Table A2 (bars: ± quoted error, the mean if asymmetric). Gray bands: quartile range and median of Huchra's list per decade (with counts). Dashed lines: the two references. Log vertical axis. The events along the top are documented on the chronology page.

What is measured

The Hubble constant H0 is the recession velocity of galaxies divided by their distance, in km s⁻¹ Mpc⁻¹. Unlike a CODATA edition, an H0 estimate has no "next edition": several teams publish values by different methods in the same year, and today early-universe (cosmic microwave background) and late-universe (distance ladder) measurements give different values. The chapter therefore sets every estimate against two reference values and counts, for both, whether the quoted error covered the distance to the reference. No analogue of the edition-to-edition move (M2) is published.

The two sources and their terms

Huchra's list — J. Huchra, "Estimates of the Hubble Constant" (hubble.plot.dat, final update 2010-10-07): a list of 553 published H0 estimates from 1927 to 2010 (550 with an error), assembled as part of the NASA/HST Key Project. The file carries a copyright line and a request to "reference the author and the Key Project" and grants no right to republish. This site therefore shows aggregates only: no row values, no chart that plots individual estimates, no row-level CSV. https://www.cfa.harvard.edu/~huchra/hubble.plot.dat

Table A2 — T. Faerber and M. López-Corredoira, "A Chi-Squared Analysis of the Measurements of Two Cosmological Parameters over Time", Universe 6, 114 (2020), Appendix Table A2: 163 H0 values from 1976 to 2019. CC BY 4.0; published here row by row. The paper states that all 120 values of 1990–2010 come from the compilation of Croft and Dailey, whose original papers the table does not name. The table was read from the arXiv preprint (2008.04035v1) and has not yet been checked against the published version. https://doi.org/10.3390/universe6080114

H2, not produced. The exceedance rate of consecutive pairs changes its verdict with how the pairs are grouped, so no number is published. Why.

Metrics

H1 · Precision half-lifeYears for the relative error of published H0 estimates to halve H2 · Not producedThe exceedance rate of consecutive pairs changes its verdict with the grouping, so it is not published H3 · Surprise against a referenceEach estimate's distance from a reference value divided by its own error: Planck 2018 and Riess et al. 2022 side by side H4 · Student-t degrees of freedomDegrees of freedom of a Student-t fitted to the z of H3: how heavy the tail is

Chronology and method

ChronologyH0 values and errors from 1929 to 2022, the changing mix of methods, events Method and limitsSources, terms of use, error levels, decisions, code tables, limits

Sources: J. Huchra, Estimates of the Hubble Constant (hubble.plot.dat, last updated 2010-10-07), Harvard-Smithsonian Center for Astrophysics; assembled as part of the NASA/HST Key Project on the Extragalactic Distance Scale. Aggregates only. · Faerber, T.; López-Corredoira, M. A Chi-Squared Analysis of the Measurements of Two Cosmological Parameters over Time. Universe 2020, 6, 114, doi:10.3390/universe6080114, Appendix Table A2 (read from arXiv:2008.04035v1). CC BY 4.0.