As used today across the crypto-world, market capitalization (market cap) can be an extremely misleading, even deceptive, metric that’s nonetheless used nearly exclusively to compare cryptocurrency valuations.
This is a brief overview of market cap and its inherent flaws as a cryptocurrency comparison metric, and an outline of some alternative metrics with promise to provide improved comparison between cryptocurrencies.
Derivation 1. market capitalization = total coin supply * daily trading price
Dv. 1 units: $USD = COIN * ($USD/COIN)
Traditional assets are often valued according to their market capitalization. However, owing to the peculiarities of cryptocurrencies relative to traditional assets, a number of rather serious issues arise when applying an analogous metric to cryptocurrencies.
The cryptocurrency market cap metric is defined by Derivation 1: where the COIN unit substitutes for the particular coin in question, e.g. BTC, BCH, ETH.
It is a simplistic metric that can be profoundly misleading (and often intentionally deceptive) when applied to cryptocurrency valuations.
Basically, there are two primary problems with cryptocurrency market cap:
- All new forks initially inherit their parent blockchain’s coin supply, regardless of the actual initial uptake, adoption, and use of the airdropped coins.
2. Many centralized shitcoins can trivially manipulate their own market cap due to limited liquidity or by simply expanding their supply of coins.
Essentially, the first fault allows anyone who can trivially fork Bitcoin to create an alt-coin with apparently equal historical context to that of Bitcoin. However, an examination of the underlying metrics of forked blockchains reveals pronounced deficits, relative to Bitcoin, which are all but ignored by the market cap metric.
For example, only a small fraction of newly airdropped coins are ever even activated (spent on new fork), i.e. there is extremely low uptake and adoption of airdrops. Nonetheless all the historical coins present at the time of the fork count towards the new altcoin’s market cap.
One of the primary measures of how valuable a currency is is how widely it is used. It doesn’t take a great amount of research to discover that the vast majority of forked blockchains have no real use case and, as such, are not being used. The smoking gun: negligible activity of their coins, indicated by underlying activity metrics such as active addresses, bitcoin days destroyed (BDD), transaction and payment counts, etc.
As a direct result of the first flaw, for those coins with negligible real-world usage, the second flaw allows for the direct manipulation of their market cap because of their inherent lack of liquidity.
Additionally, centralized altcoins may manipulate their market cap by simply removing their coin cap (if any) and inflating the supply of their scamcoin, at will, ad infinitum.
Because of these two inherent flaws, others researchers have decided to stop using the term ‘market capitalization’ altogether (coinmetrics.io uses the term network value instead). I toyed with the idea of calling it ‘market crap’ throughout; but upon review, decided this would likely introduce needless bias.
BCash, a relatively recent fork of Bitcoin, has been able to piggy-back off Bitcoin’s blockchain history for over a year now. By no means is it the only forked shitcoin that owes its seeming success (and continued existence) almost exclusively to the misuse of market cap. Similarly illustrative comparisons may be applied to the shitcoin of your choosing.
I’m picking on BCH here because I recently published an article that serves as a good case study to illustrate this point: Bye bye BCash: good riddance to bad actors. I apologize if it’s a little over the top; but honestly BTrash, you guys started it by continually trying to hijack the Bitcoin brand for personal gain.
As for the main BCH chain, the following [metrics] tell a story of plummeting prices… falling volumes… decimated transaction and payment counts… sputtering active addresses… negligible block sizes… dust-level fees… and nose-diving mining difficulty.
Yet, despite all the above BCH metrics clearly signalling the floundering state of the BCash blockchain, it remains in the top five leader-board of all major exchanges, by market cap. All major exchanges continue to use the flawed metric, ubiquitously and nearly exclusively, to compare valuations between cryptocurrency markets.
Partners in crime
By continuing to promote the mainstream use of market cap, I would argue that these exchanges are not only enabling this deception but are often directly responsible for the present plague of utterly useless altcoins, scamcoins, and shitcoins. I view them as a hoard of hitchhiking monkeys on the back of Bitcoin, see Figure 0.
Undoubtedly, most have been used by exchanges and thieves to confuse and scam the public, tarnish the Bitcoin brand (along with a meager handful of other useful coins), and generally mire adoption of cryptocurrency.
This bad state of affairs need not continue; what follows is a brief overview of a few proposed alternative metrics that I believe allow for a more grounded, informative, and transparent comparison between cryptocurrencies.
Note: usefulness is certainly subjective; for example, shitcoins are very useful to their lead scam-heads. I don’t wish to become mired in the perpetually raging shitcoin-wars but instead to highlight a few alternative crypto-valuation metrics.
One potentially useful thing that all public blockchains are good at generating is data. So without further ado, let’s get researchy!
The researchers at coinmetrics.io have put together a comprehensive plotting tool for quickly comparing a wide range of metrics and cryptocurrencies. Owing to its ease-of-use, I’ve opted to primarily use figures generated with their plotting tool, rather than compile data and generate figures myself. This allows me to spend more time on interpretation, less on coding; thanks coinmetrics.io, you da real MVP!
The metrics discussed below have been conceived by others (see Acknowledgments section). Here, I’ve simply tried to further clarify their derivation formulas (Dv. 1, 2, 3, 4), respective units (which can often add significant clarity and intuition to equations), original purposes, and potentially novel applications.
Derivation 2. NVT = 90 day forward/backward average of (market cap / total daily transaction value)
Dv. 2 units: None = $USD / $USD
Network value to transactions ratio (NVT; Dv. 2; Fig. 1) was first proposed by Willy Woo, who in Sep. 2017 published an article in Forbes explaining NVT. The idea behind NVT is to be semi-analogous to the price-earnings ratio (PE Ratio) used to valuate traditional assets. It can also be thought of as a measure of “inverse monetary velocity.”
Additionally, NVT Signal (NVTS; Fig. 2) was proposed as a modified version of NVT with enhanced smoothing and top/bottom signalling effects. It is calculated as follows: market cap / 90 day moving average of total daily transaction value; as such, it is also a unitless ratio. Those interested can learn more about NVTS from its champions Dmity Kalichkin and Willy Woo.
Of interest to investors and traders, is the potential use of NVTS as a signaler of large-scale market tops and bottoms. Certain Bitcoin NVTS thresholds have historically been associated with the most severe oversold (<1) and overbought (>45) market states (Fig. 3) — though these thresholds do not appear to hold for other cryptos.
The idea is that when NVTS crosses over these thresholds it’s a good indication that a prevalent market trend is likely to reverse course — owing to the proximity of market under- or over-valuation extremes.
Also of note in Figure 3 is BCash NVTS falling off a cliff, as of their latest forking fiasco, while Bitcoin and Litecoin NVTS remain fairly stable, despite the most recent crypto-wide drop in trading prices.
Problems with NVT(S)
Firstly, the derivation of NVT(S) relies directly upon market cap (Dv. 2), which alone should signal alarm bells, and may deter its usage by the cautious researcher or investor.
Secondly, they are unitless ratios, contrary to the usual valuation metrics ($USD), which may introduce confusion if applied outside their intended scope of usage. It’s important to keep in mind that NVT(S) are inverse monetary velocities and not direct valuation metrics.
Thirdly, the emergence of sidechains, such as the Lightning Network and Blockstream’s Liquid Network, means that NVT(S) may need re-calibration — as significant quantities of inter-exchange transactions may eventually shift off-chain. Although this re-calibration by no means renders the historical data useless, it may introduce certain ‘noise’ to the metric.
Though NVT(S) surely have important applications, the above flaws must be kept in mind when using NVT(S) as a tool for valuating cryptocurrencies.
As an aside, in a future article I’d like to explore how the above re-calibration may prove useful in measuring just how much BTC transaction volume has moved off-chain and onto sidechains such as Liquid and Lightning.
Main-chain transaction volumes could be reduced as exchanges move high volume transactions off-chain; hence, Willy Woo’s new oversold threshold should increase proportionally from its previous level (45). The resulting re-calibration (delta change) should, in theory, be equal to the inverse of the total transaction volume moved off-chain to sidechains such as Lightning and Liquid. We’ll have to wait and see.
I won’t dwell longer here on what else NVT(S) is good for, as I think coinmetrics.io already killed it.
Briefly, I’ll mention the buy support metric, as featured by Samson Mow on Magical Crypto Friends, which coinmarketbook.cc describes as the “sum of buy orders at 10% distance from highest bid.”
Essentially, it’s a compilation of buying support (buy orders within 10% of highest bid) averaged across the order books of various exchanges. Unfortunately, the proposing researchers have yet to publish the exact derivation of their buy support metric; however, units are cited in $USD.
Additionally, their related buy support rating metric highlights the amount of buy support for different coins relative to BTC (compares any coin buy support to that of BTC). As such, it may turn out to be instrumental in bringing more transparency to the scale of liquidity of various cryptos.
Owing to their current lack of transparency for the exact derivation of buy support, I cannot further unpack it here. For more information, you may have to sign up with coinmarketbook.cc (or better yet, ask them to publicly disclose further information).
Derivation 3. realized cap = UTXO * trading price at date of UTXO creation
Dv. 3 units: $USD= COIN * ($USD / COIN)
Perhaps the most promising of the proposed cryptocurrency valuation metrics is realized capitalization(Dv. 3; Fig. 4, 5). This month, coinmetrics.io published an article detailing their realized cap metric and added it to their plotting tool.
Realized cap is specifically designed to “reduce the contemporary impact of long-lost coins.” However, as I hope to illustrate, it accomplishes a great deal more than merely marginalizing ‘lost’ coins — many of which are not lost at all, but simply haven’t been activated for a long period of time.
Many keys to very old UTXOs are most certainly lost forever (an estimated 15% of bitcoins); however, some are just in very deep storage or, in the case of forked coins, have never been activated or spent (no uptake) on the upstart fork.
Figures 4, 5, and 6 reveal realized cap to be an approximate step-wise function that remains generally stable in the long term, with pronounced step-ups associated closely with the largest crypto bull runs. However, these step-ups are not only caused by price increases but by contemporary movement of old coins out of deep storage (when price was much lower).
Note: Coins in Figure 6 were specifically chosen based on two characteristics: 1) realized cap derivation is readily available; 2) relative longevity — i.e. all coins on coinmetrics.io amenable to realized cap that are relatively long-lived. I neither endorse these coins nor know much about them.
Of course realized cap comes with its own unique problems, namely: it’s not directly applicable to all coins, and it can be somewhat difficult to calculate, relative to more simplistic metrics.
Yet despite these issues, Figures 5 and 6 demonstrate well the general tendency of realized cap to provide a much stabilized valuation across cryptocurrencies, relative to their more volatile market caps.
If major exchanges began placing more emphasis on realized cap valuations, rather than market cap (where applicable), what could be the results?
Much of the extreme volatility seemingly inherent in market cap could be largely eliminated — excepting the relatively brief step-up, step-down periods. And the extremes of the market hype cycle — both overvaluation and undervaluation — could be dramatically reduced.
The relatively low volatility (both short- and long-term) of realized cap could prove invaluable in furthering the use of Bitcoin as a practical medium of exchange, i.e. genuinely usable currency.
Over the course of the 2018 bear market, from their respective all time highs, Bitcoin valuation dropped by ~13%, Litecoin by ~36%, and BCash by ~49%, by realized cap (Fig. 5). Whereas by market cap Bitcoin valuation dropped by ~82%, Litecoin by ~93%, and BCash by ~98%, over the same time period.
Clearly realized cap presents a much more stable metric for arriving at a more grounded valuation of applicable cryptocurrencies. Notice also that, by realized cap, Bitcoin has been a greatly superior store of value than all other coins; whereas this is not readily apparent from looking only at market cap.
Why is this the case? Simply put, Bitcoin has much stronger hodler support: despite large drops in trading price, realized cap remains stable, so long as long-lived UTXOs remain unspent.
A stronger emphasis on realized capitalization could go a long way towards increasing user confidence in Bitcoin as a store of value and medium of exchange alike.
Of particular interest to investors and traders is how realized cap has historically dipped below market cap during the greatest buying opportunities (Fig. 4, 5, 6, 7, 8). Enter MVRV.
Derivation 4. MVRV = market value / realized value
Dv. 4 units: None = $USD/$USD
Another way to view realized cap (a.k.a. realized value), relative to market cap (a.k.a. market value), is market value to realized value (MVRV; Dv. 4; Fig. 7, 8), introduced by David Puell in Oct. 2018. MVRV is a unit-less ratio.
A few major trends that are immediately obvious from Figures 7 and 8 are as follows:
- Bitcoin’s MVRV has predominantly remained >1, except proximal to its longterm pricing bottoms.
- Litecoin’s MVRV has spent much of its time with fractional (<1) MVRV, except proximal to pricing tops.
- BCash’s opening MVRV has steadily plummeted, to a greater degree than Bitcoin and Litecoin.
Over the next few years, it should be revealing how BCash MVRV responds, as the crypto-markets are likely to transition from bear back to bull. If the BCash scam-heads are right, about it being in the same league as Bitcoin, its MVRV should quickly follow Bitcoin’s back to >1 levels; if not, it should languish at fractional MVRV levels (similar to Litecoin’s), until we approach the next FOMO market peak. Based upon the general collapse across most BCash blockchain metrics, I predict the latter.
So too do others: as indicated by the most recent, dramatic BCH sell-off event (Fig. 4, 5). By no means is this sell-off event the first of its kind. The first mass BCH sell-off apparently occurred right out of the gate, as indicated by the initial extreme overvaluation of BCH MVRV (Fig. 7, 8).
A likely cause of the extreme overvaluation of BCH MVRV at its inception was masses of BTC holders ‘cashing out’ their newly airdropped BCH UTXOs. Why was there not a corresponding spike in BTC MVRV? Because the original BTC UTXOs (airdropped to BCH) needn’t be moved, and thus, BTC realized cap remained largely unaffected.
Most other shitcoin MVRVs also contain similar airdrop sell-off events and spend much more time <1 than does that of Bitcoin.
Looking only at logarithmic plots, tends to minimize the gap between Bitcoin and its would-be rival coins. To put everything in perspective, we can turn to the linear comparison of the coins (Fig. 4, 7).
I don’t know about you, but I like to see it as Bitcoin pressing its thumb down on those young upstarts who got a little big for their britches. Some of them will surely be all-but suppressed on logarithmic scales for the foreseeable future.
Historically, owing to its often drastic volatility, reliance upon market cap valuations by exchanges has no doubt inhibited mass adoption of cryptocurrency. Perhaps such volatility will be reduced along with increased adoption, however this . Perhaps instead of waiting for increased volumes to decrease crypto-market volatility, we should seek improved valuation metrics.
To arrive at a firm valuation of any asset, its necessary to understand as much about it as one may. Traditional assets are generally rather stingy with public disclosures of information. Fortunately, decentralized blockchains, by their very nature (privacy coins aside), require 100% transparency of their internals.
When valuating any cryptocurrency — Bitcoin is certainly no exception — it’s best to draw upon a wide range of metrics, each according to its unique insight.
The proposed metrics above are but a taste of some new and improved analytical tools promising to bring greater transparency and understanding to the fledgling crypto-sphere.
There is much insight yet unfound, hiding in plain view, waiting to be teased to the surface by existing metrics — not to mention those metrics yet undiscovered.
In my next article, I intend to propose new metrics derived from a combination of these and other metrics, which may provide even further insight into this great experiment that we call the crypto-sphere.
If you slogged all the way here, thank yourself for finishing.
Thanks to all those researchers referenced within — Dmity Kalichkin, Willy Woo, Nic Carter, and David Puell — along with others I may have missed. Keep up the great work!
Special thanks to Nic Carter and Giulio Prisco for reviewing this article and to ChainRift Research for sponsoring it