New Santiment Metrics: Analyze Network & Exchange Activity with Deposit Addresses

Like it or not, exchanges are the backbone of the crypto industry.

According to our recent analysis, crypto exchanges outperformed all other market segments over the last two years – and by a wide margin. The rally was spearheaded by Binance, which hit an all-time high in the BTC market earlier this month (tho it’s still down ~60% in the USD market).

It’s also safe to say that crypto exchanges are the most active part of the digital currency market, generating thousands of daily transactions and interacting with legions of market participants on a daily basis.

As a result, their on-chain behavior presents a treasure trove for any serious crypto analyst.

One of the most insightful aspects of the exchange ecosystem are the so-called deposit addresses. 

What are deposit addresses?

When somebody wants to make a deposit to an exchange, there are two transactions that occur:

1. Coins move from a personal wallet to a temporary wallet, created by an exchange for each separate user.
2. Coins move from that temporary wallet to the main exchange wallet.

We call these temporary wallets deposit addresses, or deposits, and they can supply us with invaluable information about a particular network’s behavior.

Analyzing deposit addresses can prove useful in several ways:

1. The total number of deposit addresses is a fairly reliable proxy of the actual number of people trading a particular coin on the exchanges.

The metric is not 100% accurate in this regard as a single person may have a variety of different deposit addresses, but it does give us a clear upper limit on the total number of users, and works pretty well dynamically.

2. Relying on network-level metrics (for example, total transaction volume) doesn’t really tell us what goes on in different sections of network activity.

Being able to distinguish which part of total network activity was generated by deposit addresses lets us get a lot more granular when exploring a specific market event.

3. Tracking the activity of deposit addresses may help us examine and foresee market movement with a new level of awareness.

For instance, a sudden spike in active deposit addresses for a certain coin may indicate an upcoming dump (more on this in a moment!)

So after promising quite a bit, let’s dive into the actual metrics:

Our New Deposit Metrics

We’ve created 4 new deposit-related metrics: each looks at exchange’s deposit addresses and their impact on the overall network from a different, unique angle.

These are:

• Number of Daily Active Deposits,
• Share of daily active deposits in daily active addresses
• Deposit-related transactions
• Share of deposit-related transactions in total network transactions

Let’s cover them in more detail:

1. Daily Active Deposits (DAD)

Daily active deposits (DAD) shows the number of unique deposit addresses that have been active on a particular day.

As an example, let’s look at the DAD metric for REP, Augur’s native coin. REP’s price is in green, while Daily Active Deposits are colored blue.

The first thing we can see is a more-less stable DAD spread over the past 6 month. On average, 43 deposit addresses were active on the Augur network each day.

Compared to other ERC-20 tokens, that’s not a bad tally. For example, over the same time period, SONM recorded an average of 10 DAD, while SNT (Status) had 27, despite it being one of the most actively developed ERC-20 projects.

In contrast, OMG – one of the most active tokens on the Ethereum network – boasts an average of 110 DAD.

We can also see a notable spike in DAD in the middle of January, which just preceded REP’s downward trend. Dump preparation or a (improbably) lucky coincidence? You decide.

2. Share of Daily Active Deposits in Total Daily Active Addresses (SDD)

Built atop of DAD, SDD calculates the share of daily active deposits (DAD) in the total number of daily active addresses (DAA), by dividing the two (DAD/DAA). The results are shown in percentages for simplicity.

The idea behind this metric is to serve as a quick-glance indicator of the speculative interest in a particular coin. The bigger the SDD, the more people are using the coin for trading rather than whatever its proposed utility, and vice versa.

It’s worth keeping in mind, however, that SDD does not give us the full picture in terms of an absolute speculative interest on a network.

For example, an SDD of 20% does not mean that only 20% of the network activity on a given day is trade-related.

Our DAA metric, which tracks a total number of Daily Active Addresses, consists of several types of addresses interacting with a network daily:

1. Deposit Addresses
2. Main Exchange wallets
3. Withdrawal Addresses
4. Addresses that are actually using the coins for its utility

Obviously, 3 out of 4 of these (types of) addresses are trade related. SDD, by itself, accounts only for the share of Deposit Addresses in total Daily Active Addresses.

The number of main exchange wallets is more-less constant and almost entirely inconsequential compared to the number of deposit and withdrawal addresses.

However, when using SDD to gauge the relative speculative interest in a coin, keep in mind that withdrawal addresses are not included in the calculation, so the actual % of trade-related network activity will usually be even higher.

That said, a spike in SDD will still undoubtedly indicate that the speculative activity on the network is growing. On the other hand, a drop in SDD may suggest network maturation, as less people use the coin for trading and turn to its (other) purported benefits.

P.S. We’re currently working on several new metrics (including those related to withdrawal addresses), that will make it even easier to evaluate both the relative and absolute speculative interest in a project. Stay tuned!

Compared to DAD, which can be used to examine and explain momentary market events, Share of Daily Active Deposits is designed as more of a long-term indicator. So far, we found it to be much more useful for a wall-to-wall retrospective or comparative analysis.

Let’s compare the SDD for a couple of different projects:

1. 0x (ZRX)

2. Bancor (BNT)

Save for a few notable spikes, 0x has exhibited a relatively stable share of Daily Active Deposits in DAA for the last 18 months. On average, the network’s SDD stands at 15.5% for the observed time frame.

Bancor’s SDD, on the other hand, is a fair bit more volatile, ranging from an average of 20% at the end of 2017 to just ~5% for the last 30 days. This tells us that the coin’s being traded less than it used to be, at least relative to the number of total daily active addresses on the network.

This should be good news for the project, as the decline in relative speculative interest indicates a growing maturity of the network: nowadays, Bancor seems to be used more often for its primary (utility) purpose.

3. Deposit-related Transactions

Deposit-related Transactions shows the total number of all incoming and outcoming transactions involving deposit addresses on a particular day.

As such, this metric accounts for both user-to-exchange transactions:

1. from a personal wallet to a deposit address, and
2. from a deposit address to the main exchange wallet.

Deposit-related Transactions is often valuable when used in conjunction with Exchange Inflow – another one of our metrics – which shows the total amount of coins moving to exchange wallets on a particular day.

So, for example, an increase in exchange inflow but a plateauing number of deposit-related transactions means that more coins are entering the exchanges, although the number of transactions to the exchanges stayed the same. In other words, the average deposit size has increased.

As another use-case of this metric, let’s take a look what Deposit-related Transactions can tell us about MKR’s popularity over time:

The Transaction graph clearly shows a steady increase in the total number of deposit-related transactions – especially in the past few weeks. In other words, the speculative interest in Maker seems to be blooming.

It’s also noteworthy that the Transaction Number metric really started breaking out in Autumn, coinciding with Maker’s surge in popularity.

4. Share of Deposit-related Transactions in Total Transactions (SDT)

Finally, this metric shows the share of deposit-related transactions in the total number of transactions on a network.

The main goal remains to measure the level of speculative interest in a coin – but from another perspective.

SDT is particularly useful when combined with Daily Active Deposits and their share in total daily active addresses (SDD).

For example, it could be the case that a relatively small number of deposit addresses generates a disproportionately large amount of transactions on the network. Or vice versa – a large number of deposit addresses generate a relatively small amount of network transactions, indicating that a token is used predominantly for its utility.

The former seems to be the case with REP. If we look at Augur over the last 6 months, we’d find just 30% of deposit addresses (on average) responsible for 50% of all network transactions:

 

We see another peculiar case in the case of OmiseGo, where the number of daily active deposits over the last 6 months stayed the same, but the share of deposit-related transactions significantly increased. This means that non-deposit addresses now account for comparatively less transactions, which may indicate a steady rise of speculative interest in the OMG network:

 

Moving Forward

The above is not an exhaustive list of things we could build on top of our database of deposit addresses.

For one, it’s also possible to distinguish between deposit addresses of different exchanges; this could, in turn, help evaluate how many active deposit addresses – or what part of total deposit volume – belong to a particular exchange.

If you have any questions about these metrics or ideas for what metrics you’d like to see Santiment build in the future, feel free to get in touch with the Santiment team on our Discord or Telegram!