Mileage Board/Year Report v. Actual Miles

[Jeff Linger]

Any idea why my mileage report by year, as well as my miles on the mileage board for the year are coming up 7.5 miles short of the total I get when I add each week up manually? I've checked through the entire year to make sure there were no days that were edited and saved, but not published to calendar. I found 1, but that only brought it up from 13.5 short to 7.5 short.

[Jon Allen]

Are you adding the weekly mileage for the first and last weeks?  Or are you only counting the mileage for the days of those weeks actually in the year?  For example, the last week of the year only had Dec 30 and 31 in the year, but Dec 30-Jan 5 would appear on the weekly mileage.  Not sure if that is the problem, just a guess.

[Sasha Pachev]

Could be a bug. How do you do it "manually"? I assume you use a spreadsheet, right? If so, could you e-mail me the spreadsheet?

[Jeff Linger]

How do you do it "manually"? I assume you use a spreadsheet, right?

LOL

By manually, I mean good old fashion manually ... I added it by hand. I took every week for the year and wrote the totals down on a sheet of paper and then added the weekly totals with a calculator. I checked the calendar twice and my math twice, but got the same result ... however, what I missed was that December 31st was part of the mileage for week 1 total. Take a guess what I ran on December 31st?

[Jon Allen]

I'm guessing it wasn't a 100-mile ultra...

Problem solved.  If it makes you feel better, I've done the same thing on monthly mileage- I just add the weeks, forgetting the account for the start and end days.

[Sasha Pachev]

Wisdom from a programmer. When debugging, never use large numbers or large quantities of numbers. The computer is never wrong just because a number is large (unless you overflow, in which case it is ridiculously wrong), or because there are a lot of numbers (unless you run out of memory, but you need really a lot of numbers to do that on a modern computer, you will run out of memory yourself much quicker). If it is going to be wrong, 9999 times out of 10000 it will be wrong when you have only two or three numbers less than 10 in absolute value. Computers err a lot, but they err like computers, not like people.