Friday Passovers, 26-36 CE
GradeComputed
The Gospels' details can be tested against the sky. Passover depends on the moon, and the moon's past phases can be calculated. When the Gospels' Friday is set against the Passovers of Pilate's term, only a handful of dates survive, and the two best ones, 7 April 30 and 3 April 33, are the same dates that published reconstructions have reported, by Fotheringham (1934) and by Humphreys and Waddington (1983). The method also passes an independent check: the full moon it gives for 3 April 33 falls on the date of a known partial eclipse of the moon. That independent astronomy and the Gospels' own data converge on so narrow a window is a mark in favour of the Gospels as sources for the setting of Jesus' death.
Method: The spring new moons for 26-36 CE were calculated from the standard astronomical formulas of Meeus, converted to ordinary clock time with Morrison and Stephenson's correction, and then to Jerusalem local time. A Jewish month began at the first sunset when the new crescent could be seen. Because the exact visibility rule is unknown, three rules were tried: a moon at least 20, 24 or 30 hours old at sunset. Nisan was taken as the month whose Passover fell on or after the spring equinox, and the following month was also checked, in case an extra month had been added that year. Dates are given in the Julian calendar.
Computed. If Jesus died on Nisan 14 (John), the Friday dates in the usual month are 7 April 30 and 3 April 33, with 11 April 27 possible on a stricter visibility rule and 22 March 26 barely possible. If he died on Nisan 15 (the first three Gospels), a Friday comes only on the most lenient rule, with a crescent seen at under a day old (11 April 27 or 7 April 30), or in a year with an added month (30 April 28, 27 April 31, 23 April 34).
Explore the counterarguments
The astronomy narrows the year to a few candidates but cannot choose between them, because the visibility rule actually used and whether an extra month was added in a given year cannot now be known. The calculation reconstructs a calendar based on observation, not the later fixed Hebrew calendar, so its results are reliable only within those assumptions. The Synoptic date fits only under looser assumptions, which counts somewhat against the first three Gospels' timing of the Passover.