That's good stuff, but I don't think the actual data is particularly useful. The year over year convention is used because you can tease out all other numbers easily therefrom: For example, suppose you have 3 consecutive drops of 10% according to the year over year convention. Your beginning to end real percentage decrease is determined by: decrease = (((.9)*(.9)*(.9)) - 1) * 100 = 27.1%. The real decrease, therefore, is not determined by simply adding the consecutive decreases, but, rather, by multiplying the previous yearly decrease by the subsequent yearly decrease.

It is important to express the proper decimal. If a yearly drop is 7%, your first factor is .93. If your second year over year decrease is 11%, your second factor is .89. If your third year over year decrease is 8%, your third factor is .92: Resulting in the following: (((.93)(.89)(.92))-1)*100) = 23.85% year 1 to year 3 real percentage decrease.

So why bother actually figuring out from year 5 to year 1 drops when that information is baked into the year over year statistics already.