The question:

“Build a list of rankings to help us find the address with the most FLOW tokens. Identify any trends or outliers you see in this collection of whales. Bonus points if you can identify the wallet holders.”

• Find pools and contracts and remove them from the list. These are outliers in a sense. We are looking for actual participants in the system.

• We will look at transaction data to get a sense of what type of addresses we are dealing with. Some addresses are just storing value, while others have large transaction volumes in and out.

• We will look at the age of the wallets and the distribution of age across the top 25 wallets, with the contracts and pools removed.

• We will examine the type of behavior that the “whales” engage in.

  
  \

  \

\n

On the left, we see the “first run”. To get balances, we simply take the sum of all the transfers in to the wallet and subtract the transfers out. Sadly, using the ez_token_transfers table was a fail. The table was not catching all transfers. Instead, I used the flow.core.fact_events table, honing in on the TokensWithdrawn and TokensDeposited event types. This yielded results with a satisfactory, but not perfect accuracy. These values were checked using Flowscan, the block explorer for Flow.

Next, I utilized the `flow.core.dim_swap_pool_labels and `flow.core.dim_contract_labels` tables to filter out swap pools and contracts.

This yielded a small list of addresses which were either contracts or swap pools (See ‘Special Whales’ table below.

There are only a handful of addresses to be concerned about, with the swap pools, in which $FLOW is one of the pairs. You can see that these are low enough in the rankings that we won’t concern ourselves, as from here on out we will just be looking at the top 25.

We will be removing the top address from this list. We are interested in people with accounts and not contracts participating in the network infrastructure.

0x8624b5… The mother of all Flow address Outliers

• Note: Let us remain aware that this study is only dealing with the Crypto Currency $FLOW. It is probable that there are addresses holding a much larger fiat equivalent than those on the list of top $FLOW holders. Many large investors and traders may actively be utilizing the Flow blockchain, but only hold the necessary Flow funds for fees.
• All currency values are in $FLOW unless otherwise specified
• The addresses have been shortened to the first six characters without the leading ‘0x’. You can find the full addresses if you click on the arrow of a relevant chart or table . You will be taken to the query window. Just type “select * from whales” after cloning the query.
• Percentages, in the donut chart reflect the percentage of the total balances of the listed accounts. For example, in the ‘Original rankings’ donut below, those proportions are based on the top ten addresses.

As can be seen in the ‘Original Rankings’ Donut chart above, this address has the lions share of the Flow token wealth. This was the one address, whose totals did not jell. The balance, according to Flowscan was about 18 million Flow as opposed to the 630 million Flow which shows up with my query! I soon discovered that this address is singled out in the Flipside documentation.

“ … Note, on the Flow blockchain the account and the contract are separate as the account may have created more than one contract, in rare instances, such as the following FLOW address 0x8624b52f9ddcd04a.”

In other words, because there is more than one contract activated, involving the same transfer of tokens, it will get duplicated in the tables. One more note to the Flow table curators: on flowscan we see an additional two event types in addition to the listed in the Flipside dim_contract_label table; FlowEpoch and FlowClusterQC.

Outlier number two: 8f77d4 a mystery address with the lion’s share of the wealth

There was no contract label in the tables and no special mention on Flowscan. I treat it as a normal address, as no contracts are listed on Flowscan. You can see from the ‘Top 25 Whales by balance’ chart on the right, that this address makes up nearly 40% of the aggregate total of balances, from the top 25 list. Determining the source of this address remains a mystery to me at the moment. I expect it is a Dapper Labs controlled address, but that is only a guess.

Below left, we see

Surprising Result – Not a single one of these addresses has a single swap interaction, according to Flipside’s ‘ez_dex_swaps table ?!?!

This was quite a shock to me. I expect the problem in the analysis, was limiting our view to the top 25 ‘whales’. We need a better job of screening out certain addresses, which may be reserves for a DAO, hot-wallet addresses from Centralized exchanges, or other blockchain related functionality. This is also due to the fact that the defi market on Flow is still in early stages, as NFT sales has been the focus.

Summary and insights

• Sadly, I was unable to discover a method to give names associated with addresses. It appears that at least some of the addresses in our list are from central exchanges. Still others may be somehow connected to the Flow team itself.

• Looking at transaction events, we don’t see a big diversity of activity. There is some delegator action and some of the addresses are actively transacting small amounts, but for the most part, these addresses are receivers.

• The bar chart confirms that most of the wallets are not involved in a lot of transacting, other than in receiving funds. The outliers are easily identifiable.

• The scatter plot shows a negative correlation between the average amount transferred and total transactions. A more interesting insight is the grouping of a small handfull of addresses with relatively low amounts transferred, but large numbers of transactions. Again one might isolate these for further study.

• Address ‘0x8f77d4’ is an extreme outlier and I sadly was unable to track down exactly what this wallet is doing. Not only does its balance comprise a huge percentage of the total Flow in circulation, but unlike most of these addresses, it is transacting frequently and for big amounts.

  \

If you are looking at this using the new Flipside app. Don't. Some things created , using the old app, do not render properly in the new app. This bounty won a grand prize. You wouldn't know it looking at it with the new app. Please go to the old app dash