At this stage in the (very limited) maturity of cryptocurrencies, I usually look at exchange volume. Obviously, a CC isn't much use unless it's liquid. One thing that makes me uncomfortable about this approach is that I don't really understand why people are buying the CC they are (in the sense that every femto-satoshi exchanged has both a seller and a buyer.) Market Cap doesn't have much functional meaning, and is distorted by pre-mined CC.

The trade-volume approach leads one to think Bitcoin and probably Litecoin (about 1/4 of Btc volume) are viable, but Dogecoin is way down (usually around 1/10 of Ltc) , and past that it's all noise.

Until the regulatory environment is figured out, all CC are going to be volatile.

Since blockchains are public, there are some interesting *data* mining opportunities, though it doesn't seem like there's been much sustained or academic interest in this so far.

Actually, there are a few works doing "data mining" of the blockchain (under different terminology) and some interesting results so far:

https://eprint.iacr.org/2012/584.pdf‎

https://eprint.iacr.org/2013/734

https://eprint.iacr.org/2013/782

On a slightly different angle, there is a ton of fascinating work to be done on the metrics of *building* a cryptocurrency and the corresponding delicate balance between mining/spending efficiency (choice of proof-of-work, alternate designs), and resulting security (resilience against selfish mining, race conditions, etc.).

These traditional criteria for "currency-hood" are unuseful. Crypto-currencies don't care whether they are a store of value, for instance: it's easy to imagine a future scenario where no one holds a Bitcoin for more than a moment.

It's perplexing why crypto-currencies get people so excited (either positively or not). They are a mechanism, and will succeed if they provide a better way for people to go about their business. Fundamentally, there is *great* need for a fast, cheap, secure, and trust-free way to make purchases, and crypto-currencies will at least be part of that solution. This is in contrast to old-fashioned systems like credit cards and wire transfers, simply because those mechanisms are based on a set of pre-digital, pre-internet premises. The fundamental metrics are whether customers find Bitcoin (et al) easy to use, and how much they save, relative to the insecure traditional approaches. (Credit card companies work hard to hide their inherent insecurity, and make problems like the Target breach disappear. But ultimately the customer pays for these flaws, as well as the last-century infrastructure...)

Thanks for your thoughts Alan. My question was not so much about the Aristotelian axioms of money-ness, as much as a question about viable metrics to determine the health of a currency. Many types of data can be faked (a number of coins appear to have a large market cap by having an absurd number of coins at whatever value they can muster) but some information remains useful. Days destroyed seems to have occasional application. The number of transactions per day or per block might have some utility. I guess there's a larger question here regarding looking at the distribution and 'velocity' of the currency.

I think that you can see a pretty stellar possible direction here: http://statisticaleconomics.org/2014/03/13/quantifying-the-value-of-bitcoin-2/

To draw on Mark's point, I don't think that there are good enough reasons to justify good metrics in cryptocurrencies (at least not in their capacity as 'currency') . Don't get me wrong; this is not to dismiss the value of current and previous studies and, of course, we have yet to reach technological maturity. For sure, this is a highly challenging new area for quantitative analysis, but my view is that performance is irrelevant and true value comes solely from algorithm robustness and popularity. The only 'intrinsic' value of CCs is their speculative value, and as it happens with all commodities, the Markets have rushed-in to capitalise on this by developing relevant derivative products.

Possibly we should look to Innovation Studies for more insights on future directions. True, in future, CCs may only be used instantly as a means of exchange rather than any form of store of value. The 'honesty of nodes' principle can be challenged (as we have seen in recent developments with BTC) when, say, a powerful group of miners can hide behind anonymity, take advantage of the 'democratic' 51% and manipulate or double-spend.

Technical Report How many bitcoins should you buy and what is an estimate of ...

Bitcoin (not asset backed cryptocurrencies) is nothing more and nothing less than unlicensed gambling. Instead, central bank cryptocurrencies -- digital regular money issued by government -- will lead change.

Details are in my book: Bitcoin: The Mother of all Scams. https://www.amazon.com/gp/product/B095NMLM2F