Tumbling addresses as I understand are not part of the final count as they end up empty.
I am assuming that
this graph counts addresses that appear in the blockchain (as output addresses) for the first time on each day, and does not subtract addresses that have become empty on that day. Is this second assumption incorrect?
(Sigh, blockchain.info should have explained more precisely what is being shown in each chart...)
In the extreme tumbling scenario I just described, there would be 150'000 new addresses created every day, and all would be non-empty at the end of the day.
deposits at an exchange are most often pooled often in identified address, and wouldn't add a significant count. (I've done a few purchase lately with bitpay and found they even reuse address that end up empty so that doesn't add either. I think exchange withdrawals are a legitimate addition to the total as those are the final unique address holding coins that are counted.
Exchange deposits and withdrawals may contribute little to the new address count, but they may be a substantial part of the and plots. After MtGOX, exchange clients are advised to keep their balances to a minimum.
It does not seem right to count those BTC transfers as real use: although the input and output addresses have distinct owners, the coins in the exchange logically belong to the clients, so they are not "payment" for anything. It would be like counting cash deposits and withdrawals at banks as part of the GDP.
The blockchain traffic from Chinese bitcoiners, for example, must be almost 100% exchange deposits and withdrawals; its use for e-payments is probably very small.
While the measurements may not be accurate I feel confident they are magnitudes more accurate than the tools to manage existing monetary policies.
Banks and businesses are required to report their activity to their government, for tax and other purposes, and there are many thousands of people whose full-fime job is to extract useful statistics from that raw data. For example, the published numbers on credit card usage are probably accurate to 2 digits, at least; and we know the subtotals by country and purpose. Ditto for total salaries, total company income, etc.
There is no such data for bitcoin. We don't know how much of the traffic shown in the blockchain.info plots are payments for goods and services. It could be 10% or 90% -- and the fraction could vary substantially from month to month.
In fact, the shape of those plots over the last couple of years suggests that most of that traffic is NOT commercial payments.
manipulation doesn't prove anything or give anyone a quantitative competitive advantage just yet.
Inflation of the blockchain traffic would obviously benefit all bitcoin enterprises, especially bitcoin invstment funds. For example, I have seen a couple of articles recently claiming that bitcoin has already passed PayPal in volume of payments --- using the blockchain.info as the bitcoin number.