Federico Luduena's Day for Any Date is excellent. However, I've come across one puzzling feature. The crib doesn't seem to work for 00 dates. Take, for example, 1st January 1900. According to FL's system this should be 1 + 5 = 6 = Saturday. But this is not correct ( it was a Monday ), and similar wrong answers occur with all 00 numbers.
In all other respects this effect is a delight, and maybe I am just making a mistake somewhere. I would welcome any informed thoughts on this.

Sorry Mike, I don’t yet have the effect so I can’t comment on the issue you are having.

I have been debating on learning the memory techniques for the day of any date but right now my mental effort is being used for learning other memory routines. With this having a crib, my concern would be having to reference it each time where with an effect like this the appeal is doing it for multiple people in a row, which is how I would like to do it. If that is the case would you still recommend Luduena’s routine?

1900 is the last year of the 19th century (1801 to 1900).
Therefore the calculation should be '1+5+2' which gives Monday.

1800 is the last year of the 18th century (1701 to 1800) which would give Wednesday, which is correct.

However, I think the crib assumes that all years ending in '00' are leap years, which is not true, because it goes wrong after February.

Rik
Yes, I thought of viewing the 00's that way, but, as you say, that still doesn't ensure consistently accurate results.
Thanks for the reply.
Robert
The first thing I did was disguise the crib by changing numbers into symbols, e.g. a square for 4. I then used it openly, saying it was a copy of something found in Dr John Dee's papers. As I could name the day for any date from the 1600's to the 2300's, people couldn't believe so much information could be derived from such a short, enigmatic text and assumed I was joking and really working it all out in my head. Ever the purist, however, I decided to use the letters-as-numbers technique and managed to memorise the crib. So that is how I now do it.
Cheers

Kyle
I tried to pm you but something went wrong so I'll reply here. Thanks for the leap year idea, but, as I said to Rik, that doesn't seem to be the answer.
Thanks anyway.

I don’t have this effect but 00 are only leap years if divisible by 400.

So 1800 and 1900 are NOT leap years. 2000 is a leap year.

Mike

If you use the same offset for 18xx, 19xx etc as the non-zero years then:

1800, 1900 & 2100 are the same as year 90
2000 is almost the same but the first triplet is 623


How do you handle 16xx & 17xx with the differences in the Gregorian calendar?. It was adopted at different times throughout Europe which would be troublesome to deal with.

Hi, Mike. Thanks for your comments! You are right about that glitch. I had not noticed. As Rik and Michael point pout, it has to do with leap and non-leap years. The solution seems to be this: use the "00" codes for all leap years ending in "00". Use the codes just below those (they start "033...") for non-leap years ending in "00". There is no need for a century code for any of the years ending in "00".

I tried the above and it worked, but again, please try it and post your experience if you can. I might be off.

Also, your ideas regarding method and presentation are really good. Thank you for sharing them.

Robert, yes, it is true that this method does not fit a presentation a la Benjamin, but for a more relaxed approach it works well. You can write the entire month for a birthday with almost no work.

Federico
Thanks very much for joining the conversation. As far as I can tell, your suggestion works for 1600, 1900, 2000 and 2021. It doesn't seem to work for 1700 and 1800. For example, according to the (revised) crib, 1st June 1800 should be Friday but it was actually a Sunday. 1st June 1700 looks like Friday but actually a Tuesday. Nevertheless, despite these snags, overall the system works like a dream lncluding, Rik, dealing with the calendar change without the need for extra adjustment. I live in Britain so can only speak for that location, however.
I think on reflection I might ask people to avoid 00 dates as they are too "easy" (!) and I would prefer something more challenging.
Thanks again, Federico, for this elegant prop.

Mike... How does it work for dates before September 14th 1752?

1752-09-01 gives Friday, but it should be Tuesday
1752-09-02 gives Saturday, but it should be Wednesday

The next 11 days were dropped because of the introduction of the Gregorian calendar so the next date is...

1752-09-14 which correctly gives Thursday.

For all dates after that if you stick with adding the century codes and use the same value for 18xx, 19xx etc as the non-zero years then you can just add two extra lines to the crib as I indicated above.

Maybe my thinking is wrong, my mind is not as sharp as it once was.

O.K. Rik, you have a point.
In order to check my calculations I was using the website dayoftheweek.org, which uses the Gregorian calendar from 1600's onwards. I was aware of the Julian/Gregorian change but I was happy to ignore it as this particular website was confirming that my calculations for all dates were correct. Another website, I have now discovered - timeanddate.com- displays the Julian calendar up to September 1752 and then the Gregorian, which of course is appropriate for Britain.
So, perhaps I'll just stick to the 1800's onwards, or, maybe, direct people to the fully Gregorian website: dayoftheweek.org when they check my calculations.

Incidentally, when working with non- leap year 00 numbers, to return to my original query, wlould it not be feasible to discover the relevant day for the next year, and then move back one day, e.g.1st January 1901 wasTuesday, so 1st Jan 1900 would be Monday.

Hello Friends,

I can't comment on Luduena's approach, having not seen it. But I have indeed spent a fair amount of time on the "day of the week" problem over the years, and would like to share the following, as possibly shedding some general light.

In particular, a chapter of a book I've been cranking on for some time now explores what's going on beneath the surface of the calendar. This has been written to entice newcomers to various interesting questions from mathematics and the liberal arts. The entire book is intended to serve as a supplement in the class room, either at the high school level or in an introductory college class for liberal arts students. The only prerequisites are high school algebra, and of course a curiosity about how things work.

I'll offer this chapter free of charge, with no strings attached. But I do ask you keep it to yourself and not share it around. It is copyright, and as mentioned will one day appear in a book I've been writing. Other than that, you're welcome to download it and see if it helps explain how the various "day of the week" methods work.

Here's the link to the PDF download:

https://drive.google.com/file/d/1DGADzND......=sharing

Of course, any comments would also be welcome.

Hope this helps answer some questions,

Thomas Henry

Tricky things calendars

I cannot immediately think of a reason why your non-leap year solution would not work.

For pre-Gregorian dates in the UK and USA you can just add 11 to your normal calculations, but it all starts to get messy and makes the crib harder to use with all the exceptions you need to take care of. Sticking with significant dates for living people has a natural tendency to limit the range to the Gregorian. The crib is a marvel and must have taken a lot of work to put together.

Thomas, thank you for sharing that erudite article - lots to digest there.
I've found this discussion very helpful, so thanks, everyone.
And, of course, thanks to Federico L. I've been involved with magic for many years, and, despite the (very minor) issues aired above, I regard this inexpensive gem - from Lybrary.com - as one of the best purchases I ever made.

Thanks for those words, Mike! The glitch intrigues me, because there must be a rule (a simple rule of thumb) that explains, if not solves, the issue.

Thanks.

I don't know Federico Luduena's method but I've looked at several calculation based "day for any date" methods in the past and found that they either
- contain glitches, often due to a division which disregards the remainder.
- are too long/elaborate to be performed fast/effortlessly.

Since then I've changed over to a memory based system, which is much faster.
A system like this is taught by Sal Piacente on his Penguin lecture and allows you to instantly name the day.

I've now had the opportunity to view Sal Piacente's material. It's very good and well worth purchasing. If I hadn't already memorised Federico Luduena's day for any date system, I might have considered memorising SP's method. Even so, I think I'm right in saying that the latter requires leap year adjustments, whereas, with a few easily remembered exceptions, FL's doesn't. To me, this is an important consideration.

Yes with the method that Sal uses you still need to account for leap years by subtracting 1 if the date is before the 29th of February.

Even so the method is very fast: you just add together 4 numbers from memory ( all single digit) and you know the day. If it happens to be a date in January or February of a leap year ( except for the 29th of February) it will be the day before.