Masternode Rewards in Question
February 26, 2018
February 26, 2018
In this article, I wanted to clarify important and recurring questions about masternodes and rewards while introducing an upcoming service which we hope will satisfy a broad range of magnet investors and users.
Averages Are Noncomprehensive.
When people talk about averages, they usually refer to the arithmetic mean however, the median and the mode are as valid ways to compute an average. Besides, an average is often an incomplete tool if not correlated to other figures such as a standard deviation (especially true in finance).
For example, let’s consider the two series below representing the returns in USD generated by two investments over the past 5 months.
|Month 1||Month 2||Month 3||Month 4||Month 5|
Both investments claim an average monthly ROI of USD 200 over the 5-month period. However, we can clearly see that the average alone is not enough to understand the whole picture and this may lead an investor to premature decisions (such as pulling out after 4 months). If such investor knew that the standard deviation in the first series was 425 versus 0 in second series, his/her decision could have been different.
We must emphasize that averages are given as very rough and partial figures, they should not be used as single source of information. Furthermore, the data used to compute expected averages has to be extrapolated or “guessed”, especially in pseudo-random systems.
This is why, when it comes to such systems, including Magnet masternodes reward distribution, it is more accurate to speak about “odds” than averages. Let’s dig further.
A Game of Odds
Last year, I published an article on my personal blog along with an infographic comparing the odds of winning the lottery game with solo mining bitcoin (you can read the full article here).
At their core, cryptocurrencies rely on mechanics that leverage unpredictability (hashing algorithms). These mechanics are very efficient to strengthen the blockchain because the only full-proof way to break them would be to reform the Information Theory itself – i.e. uncover the puzzle of entropy.
In a very similar fashion to mining (which consists of finding a block header hash value below a specific target difficulty), Magnet uses a deterministic pseudo-random function to compute the masternode ranks and determine the coinbase rewards winner. As long as payments are being properly enforced, this ranking mechanism being the most unpredictable is also the hardest to fraud.
While miners would increase their hash power to increase their odds of “winning a block”, investors would have to increase their stake in Magnet to achieve this with masternode rewards.
The main issue however is that these chaos-based mechanics are not intuitive for human to grasp because we tend (or have) to rationalize, order, classify and generalize to make sense of the world from our partial point of view.
This is the reason that prompts so many questions or even FUD (fear, uncertainty and doubt) as people try to rationalize their earnings (or lack of) from a limited perspective.
Let’s try to demystify all this with some simple use cases in probability.
Heads or Tails
Everyone knows that the chances to guess the outcome of a “fair” coin flip are 1 in 2 or 50%. But, what are the chances to end up with only “heads” when flipping that same coin three times?
To find the answer we simply compute the combined probability of the flipping coin events: 1/2 ^ 3 = 0.125 or 12.5% chance.
We can apply the same method to figure out the chances for a single masternode to earn a reward every day. Since there are currently 1120 masternodes on the network, with each block the chances for a masternode to be rewarded are 1 in 1120 (i.e. 0.089%). To look at it from another angle, the chances for a masternode to not be rewarded are 99.911%.
Provided there is an average of 960 blocks mined every days, the chances of being rewarded in one day are 100 – ((0.99911 ^ 960) x 100) or 57.4%. So statistically, a masternode within a network of 1120 nodes has more than a flipping coin chance to receive a reward every day. Check this spreadsheet for additional stats.
It however does not mean that every masternode will receive a reward every other day. If we get back to our flipping coin example, you would hardly expect a perfect series of alternated heads and tails.
|Flip 1||Flip 2||Flip 3||Flip 4||Flip 5||Flip 6||Flip 7|
The same applies to the masternode rewards distribution. You could see your masternode “very lucky” earning several times in a single day or missing out several days in a row.
Additionally, Magnet rewards come in two flavor due to hybrid mining: Proof-of-Work and Proof-of-Stake. Both have equal chance to occur but, due to the supply distribution curve, Proof-of-Work based rewards are more attractive at the moment. This is likely to change in the future (PoW will be halving while staking amount is likely to grow) but must be taken into account when calculating current ROI.
Steady or Lucky
Theoretically, you could go solo against the bitcoin odds on a 1 GH/s GPU miner and win a 12.5 BTC reward within the next 10 minutes. There is a chance for that to happen – one chance in 21.5 billion in fact (if we take today’s bitcoin block difficulty that would be 3,007,383,866,429 x 2^32 ÷ (600 x 10^9) = 21527692254.71)
This extreme example reflects a concept that Proof-Of-Work miners have understood for quite some time. That is, to counterbalance the irregularities incurred by low odds, they have created pools. These pools allow generation of steady revenues despite a wide disparity of hashrate powers among miners so that everyone can find a fitting shoe. Even a single miner with 1 GH/s can earn regularly from such pool (whether his/her venture would be profitable is another story).
Similarly, the collective revenue from 50 masternodes would be much more predictable than the revenue from a single one (remember how we compute the combined probabilities).
A Masternode Pool Service
As mentioned in the beginning, questions and FUD have been recurring during the past few days, mostly due to misunderstanding the core concepts behind rewards.
This is something that we have been discussing a lot with Roland because we understand that these concepts are not the easiest to grasp and we are also aware that, with a growing network, smaller investors may not find the return irregularities much sustainable.
Based on these feedback, we have been exploring possible solutions and services that we could provide.
Pooling masternode rewards appear to be the most attractive one. Such setups are voluntarily (you do not have to join if you are happy with your returns), relatively easy to setup, do not require a protocol or consensus change (horizontal solution), such models are already used successfully by miners and scale very well with a growing network and partial stakes.
As we are currently designing this solution further we will come back with more details in the coming weeks. We are also happy to hear about your suggestion and feedback in the meantime.
That will be all for now. We hope this article has been instructive and clarifies a lot of the questions you may have had about Magnet masternode rewards.
Happy Masternoding and good “luck”!