Miscellaneous desk
< November 24	<< Oct \| November \| Dec >>	Current desk >

Welcome to the Wikipedia Miscellaneous Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

November 25

Thanksgiving and the last 2 digits of the year

This year, the United States is currently celebrating Thanksgiving on November 24. However, November 24 does not match 2022. But next year, Thanksgiving will be on November 23, and November 23 does match 2023. Likewise, November 26, 2026, and November 27, 2127, are also matches.

So, is it true that the date of Thanksgiving coincides with the last 2 digits of the year only 3 times in 400 years (e.g., 2023, 2026, and 2127 within the 400-year period 2000–2399)? If so, then after 2127, the next match would be in 2423, followed by 2426, 2527, 2823, 2826, and 2927 in the 3rd millennium. GeoffreyT2000 (talk) 04:53, 25 November 2022 (UTC)[reply]

Yes.

Proof (if cal on your UNIX or Linux system works the same way as mine):

   for i in 20 21 22 23                  # First digits of year
   do
      for j in 22 23 24 25 26 27 28      # Possible dates in November of US Tthanksgiving
         do
         t=`cal 11 $i$j |                # Calendar for November
                 grep -v '[a-z]' |       # Remove heading lines, leaving just the numbers
                 sed -n 's/^.. .. .. .. \([ 0-9][0-9]\).*/\1/p' | # pick out Thursdays
                 sed -n 4p`              # 4th Thursday
         test $j = $t && echo $i$j       # Does it match?
      done
   done

--174.89.144.126 (talk) 05:29, 25 November 2022 (UTC)[reply]

This a contrived "coincidence" of zero sigificance. The US Thanksgiving holiday, by definition, falls between November 22 and 28. That reduces possible final two digit year matches in a century to seven. And since the day varies within that seven year window, with seven possible outcomes, it is completely trivial that the match happens only a once or twice in some centuries, and not ar all in other centuries. Over many centuries, it should happen once a century on average, but the current US Thanksgiving schedule has only been official for 80 years. Cullen328 (talk) 05:54, 25 November 2022 (UTC)[reply]

I didn't think the poster was asking about or positing any significance of the fact. --174.89.144.126 (talk) 08:51, 25 November 2022 (UTC)[reply]

Why discuss insignificant things? Cullen328 (talk) 00:15, 28 November 2022 (UTC)[reply]

Why not? --174.89.144.126 (talk) 23:11, 28 November 2022 (UTC)[reply]

The pattern repeats every 400 years:

        Thu Nov 23, 2023   Thu Nov 23, 2423   Thu Nov 23, 2823
        Thu Nov 26, 2026   Thu Nov 26, 2426   Thu Nov 26, 2826
        Thu Nov 27, 2127   Thu Nov 27, 2527   Thu Nov 27, 2927

This is not a coincidence. --Lambiam 10:40, 25 November 2022 (UTC)[reply]

Correct! It’s a statistical pattern arising from an artificial construct. DOR (HK) (talk) 22:58, 25 November 2022 (UTC)[reply]

Fucking everything repeats every 400 years. So long as you only deal with one or more of 'YY/MM/DD/weekdays you can reuse your 1600s ~~calendar~~ Catholic calendar collection, just cross out the century and write 20: 201622. Cause 400*365.2425 is 146,097 which is 0 modulo 7. I don't really need calendars, even without knowing most of the shortcuts I can find the date of Thanksgiving 98,747,503 AD in my head. Sagittarian Milky Way (talk) 05:06, 27 November 2022 (UTC)[reply]

Uh, no, we weren't all on the Gregorian calendar in the 1600s. In the Julian calender, the repeat pattern was every 28 years. --174.89.144.126 (talk) 08:41, 27 November 2022 (UTC)[reply]

Right, you can reuse your collection of accurate 1600s calendars. Maybe Ireland or Maryland had some? Otherwise it's slim pickings for ones in English. Maryland was America's only Catholic/Catholic tolerant colony settled March 1634. Sagittarian Milky Way (talk) 14:02, 27 November 2022 (UTC)[reply]

There was in the 1970s, and may still be, a prayerbook in daily use in King's College Cambridge with a table for calculating the date of Easter right up to 1800! And another one with a table for calculating the date of Easter forever! ColinFine (talk) 15:16, 27 November 2022 (UTC)[reply]

That book (the Anglican Book of Common Prayer?) seems cooler than the ones in Catholic pews. Do Episcopal pews have them? Imagine if the forever table didn't require calculation till the 1st cycle ends, it'd be 5.7 million entries long! Just a wall of text of the form 2019A21,2020A12,2021A4,2022A17,2023A9,2024M31,2025...5649876MDD etc would be the size of like 7 encyclopedia volumes. Sagittarian Milky Way (talk) 20:58, 27 November 2022 (UTC)[reply]

True believers would know it by heart. -- Jack of Oz ^{[pleasantries]} 17:57, 28 November 2022 (UTC)[reply]

This BCP of 1762 has A TABLE to find EASTER-DAY from the Present Time till the Year 1899 and a couple of pages later, TABLE to find EASTER-DAY from the Year 1900, to 2199 inclusive. It requires some skill in arithmatic. Why anyone in the 18th century would need to know the date of Easter in 2199 is something of a mystery. The BCP has been largely superceded by more modern liturgies for the main services, but it is still in use in some parish churches and cathedrals for Evensong, which newer texts fail to match. Alansplodge (talk) 12:39, 29 November 2022 (UTC)[reply]