Helium Mining Reward Distribution
In no way am I professional at this, I am just a student studying math. I got bored from waiting for my hotspot to arrive so I thought it would be interesting to understand how Helium rewards the network. These are conclusions I got from my analysis, by no means is this advice of any kind.
While Helium has been very transparent about how hotspots are rewarded, I decided to do the analysis myself to confirm some of the claims using real data collected.
I’ve compiled my data from Sitebot.com and a web scraper I built, scraping information found on the coverage map like H3, recent witness, reward scale, and nearby hotspot.
The data has a sample size of just under 1000, and I’ve only compile data on active hotspots in Canada on April 20th, 2021. On April 20th, there was a total of 27,656 hotspots on the network.
Understanding Reward Distribution
Lets take a simple look at the histogram of 30 day rewards. The data shows that there is 138/924 miners that are earning less than 10 HNT per month. This could explained if these miners are lone wolf miners, meaning they have no nearby hotspots to do witnessing and POC challenges. (see graph below)
If I were to filter out hotpots earning less than 10 HNT a month, the histogram looks more like a Normal Distribution heavily skewed to the right. (see graph below)
There is a monthly distribution of 5,000,000 HNT tokens per month. By excluding 34% for the Security tokens, HNT awarded to Helium, Inc and other Network investors who hold Security Tokens, this leaves us with 3,300,000 distributed among the hotspots. The mean for monthly HNT per hotspot should be 119.32 HNT tokens (3,300,000 HNT/27,656 hotspots).
I have created a box plot to understand the median and different percentage quartiles. (see graph below) Within my data set:
25% Quartile: 45.1
Median : 130.32
75% Quartile: 252
It seems that the median is higher than the mean. With the rapid pace of increasing number of hotspots, the delta could be explained due to higher rewards in the start of the month since there are less hotspots.
There were 23,515 hotspots on March 21, 2021. The average for March is 140.34 (3,300,000 HNT/ 23,515hotspots). Now the average between 119.32 and 140.34 is 129.83 HNT/Hotspot, which is very close to our 130.32 HNT/Hotspot per month.
From this analysis, There is a good reason to believe in Helium’s claims regarding the reward structure for mining. Props to Helium being so transparent about it!
Reward Potential Optimization
To better understand reward potential, we must analyze the witness and POC rewards. I am using recent witness data scraped from the coverage map as a proxy for this.
As seen in the graph below, there is a decreasing marginal return for each witness your hotspot witnesses. The sweet spot seems to be around 10–25 witnesses. After that, the reward potential starts to flatten. People below this range could try to move to better or higher placements or, if needed, improve antenna. There are many tools articles online that can help you help you optimize your RF signals. For people above this range, congrats, your hotspot is already optimized to perform.
There is also a reward scale variable. Ideally, you want to be placed in a hex that does not have too many hotspots giving you a scale of 1.00. Using number of nearby hotspots from the coverage map as a proxy to understand the scale, it seems as though people should not worry too much about having too many hotspots around you. (see graph below)
The more hotspot is around you, the lower your reward scale, however, HNT reward actually flattens out. This could be due to higher witnessing and POC activities. The sweet spot for nearby hotspot is around 10, but do not be concerned if your area has more.
In this article I have verified the HNT token distribution among the network and discovered some insights regarding reward optimization.
While profitability in mining HNT is important, I believe the Helium network is putting in the correct methods to reward those building and expanding the network; with exception of lone wolf hotspots. People should aim to have at least 1 nearby hotspot to do witnessing and challenges.