OK, going back to my example - I want to spend 1 BTC on a widget. I could spend my Bitcoin I received five minutes ago, which would only slightly increase the Days Destroyed statistic. Or I could spend a Bitcoin I received two years ago, which would much more greatly increase Days Destroyed. But there's no difference in my savings between these two actions.
Your 200,000 BTC from back in 2010 are unlikely to be in your daily wallet. Even if they were the coin picking algorithm is more likely to select newer coins that still avoid mandatory fees.
While you could spend 200,000 BTC to pay for 1 BTC trivial transaction it is unlikely you would. On the other hand say you decided you want to get a $100,000. Moving 1 BTC to the exchange isn't going to do that. Moving 10,000 BTC from 3 years ago is going to result in a major spike in BitcoinDays destroyed.
Like most metrics it doesn't matter what the individual is doing it matters what the aggegate is doing. Generally speaking in aggregate giant, old coins aren't being moved for trivial purposes.
To op: the cumulative chart is totally useless. It is just a pretty upward sloping line.
You should be looking at daily chart.
http://blockchain.info/charts/bitcoin-days-destroyed?timespan=30days&showDataPoints=false&daysAverageString=1&show_header=true&scale=0&address=
