# How Can We Improve Leap Years to Better Match the Seasons?

By James O Malley on at

Stand-up Mathematician Matt Parker has posted a new video just in time for February 29th looking at some of the maths behind Leap Years. It turns out, our seemingly fool-proof system of adding a day every so often isn't so great, when viewed over the (very, very) long term.

The problem can be measured, according to Matt, by how long it takes for the calendar to fall exactly one day out of alignment with the seasons.

In the video, which you can see above, he explains that the Julian Calendar which first introduced leap years during the reign of Julius Caesar goes out of sync by one day every 128 years. Which isn't great. This is why we made the switch to the Gregorian calendar in the 17th century - this still has leap years, but skips February 29th on years that are multiples of 100, but not 400 (hence why 1900 was a leap year, and 2000 wasn't). Under this system, the calendar only falls out of sync once every 3216 years. Better.

But this still isn't good enough for Matt, who goes on to propose his own upgrades for the system - one that would take out an addition 3 leap days for every 10,000 years. Check out the video for the full explanation but here's the bit: The easiest way to calculate it is using binary.

And if you're still looking for something to do on February 29th, perhaps another of Matt's videos could provide some inspiration: Why not mess about with spreadsheets?