The best way to find profitable Lightning nodes? Release of routing ranking based on real routing data.
I’m happy to announce that Diamond Hands has released a beta version of Diamond Hands Routing Ranking(β). I have had this idea for a while but we finally managed to find the time and resources to put everything together to create a prototype.
Diamond Hands Routing Node Ranking
https://diamondhands.shinyapps.io/ranking/
This ranking is based on real routing data on the Diamond Hands node and is updated twice a day. It tracks routing volume, average transaction fee rate, and APY for each node connected to the DH Node. We hope this will provide other node operators with valuable insights and help improve their profitability or capital efficiency.
The concept of this ranking is inspired by CoinMarketCap/CoinGecko. Just like those crypto ranking services, we aim to provide both current and historical data combined with other key metrics to provide an overview and trends in the ecosystem; but instead of different cryptocurrencies, we apply the same approach to Lightning nodes.
If you want to be on the ranking and check how you’d compare with other nodes, consider opening channels to our node and you’ll get added to the list automatically.
What is this ranking good for?
There are already other node rankings such as Terminal by Lightning Labs and capacity/channel ranking on Amboss.
In particular, the Terminal considers not only for capacity and the number of channels, but also network connectivity, uptime, and channel stability to evaluate the overall quality of nodes. This is a useful tool to discover new peers and we freqently use it ourselves as well.
Having said that, however, we often find that our actual routing data is quite different from what those rankings suggest and we are aware that other node operators also share this pain.
For example, Wallet of Satoshi and Kraken are ranked No. 1 and No. 2 respectively on the Terminal ranking, but in our ranking they are only ranked 29th and 53rd in terms of trading volume over the last 7 days.
On the other hand, Moon (paywithmoon.com), NNN, lnmarkets are our top 3 nodes for the last 7days, but their rankings on Terminal are #297,#1094, #348, respectively. This huge gap was a bit surprising even for me.
So, you get the point. While routing node rankings based on public metrics are a useful source of information for certain things, they’re not good enough on their own to accurately predict real-world routing activity or to identify active/profitable routing nodes for you.
This is where our little ranking comes in to fill the gap and possibly help solve the problem.
The ultimate goal of our ranking is to complement other rankings and data sources such as Terminal or Amboss by providing routing data based on real routing activity rather than theoretical scoring based on publicly available information. They provide more of a bird’s eye view of the network, while ours provides more of an on-the-ground view and both provide useful perspectives for node operators in my opinion.
Methodology
This ranking is still a prototype. Consider the current methodology as just a starting point and we intend to continue to improve it and provide more useful statistics as we go.
Criteria for ranking
Currently, we only list nodes that have active channels open with the Diamond Hands node. We also exclude nodes that are inactive or have no routing record for the last 7 days.
Volume Share(7D)
Volume includes both inbound and outbound volume for each node for the last 7 days. If there are multiple channels open with the same node, we aggregate the volume across different channels.
Historical Volume(24H, 7D, 30D)
24H: Compare volume between 24 hours from the last update time and 24–48 hours ago from the last update time.
7D: Compare volume between 7 days from the last update time vs 7–14 days ago from the last update time.
30D: Compare volume between 30days from the last update time vs 30–60 days ago from the last update time.
Average Fee Rate(7D)
Average Fee rate for outbound transaction for each node for the last 7 days.
Channel APY
We add both inbound and outbound revenue over a given time period (30D and 90D) and divide it by the aggregated channel size (channel liquidity) at the time of the last update. Then annualize it.
We are aware that the APY calculation, most notably, needs some more improvements.
For example, the current method doesn’t take the onchain cost(open and close) into account and also the APY rate might become quite volatile and inaccurate when there is frequent channel open/close. Also, APY tend to be overstated and when the channel size is small.
So, don’t consider our channel APY metrics as very accurate indicators of real profitability but rather, use it as a relative indicator to compare different routning nodes.
Amboss recently announced LINER in an attempt to track the network-wide annualized rate of return and we will consider emulating their formula to improve the accuracy.
In addition, we are planning to add following metrics in the future;
- Relationship between different nodes
- Categorization of nodes(sink, wall, rebalancer, etc)
- Optimal channel size estimation
- Historical routing trend charts
Limitations and potential concerns
Some of the limitations and concerns for this ranking include:
- Applicability to other nodes
The routing data may only apply to our particular node and may not be very useful for others. In general, routing activity and tendency vary quite a bit between different nodes, even with similar channel composition. Having said that, Diamond Hands node is a relatively large and well-connected node, so routing activity on our node should be relevant for other nodes at least to some degree.
- Privacy concerns
Although our ranking shows only aggregated information and we refrain from sharing specific transaction details, publishing real routing data may lead to a decrease in payment privacy. Also, publishing our routing data reveals information about our neighboring nodes without their consent (how active they are, etc.) and may not be well received by some of them.
Regarding privacy concerns, there is a bit of a trade-off between privacy and data transparency, which can be used to improve capital efficiency and payment stability. Personally, I believe that publishing a bit more aggregated routing data is more beneficial than detrimental to the ecosystem, but we are open to any feedback or criticism on this as well. Feel free to leave comments or contact us and let us know what you think.
