Some Observations of a Cardano Stake Pool Owner
I am super excited that my own Cardano stake pool is finally live!!! Please find it at our website, ADApools and PoolTool.
During the past few weeks, I did quite a lot of research and studies on production stake pool data. I will present my findings in attempts to answer the following questions that may of your interest.
- How much I can earn as a stake pool owner?
- How much stake is needed to make my stake pool competitive?
- Should I build my own stake pool or delegate to others?
How much my pool can generate?
My primary objective of building own stake pool is to learn more about Cardano and Proof of Stake (POS). With that said, I still need to get funded to support the daily maintenance and hopefully get more rewards than delegating to others. That leads to my 1st question — how much my stake pool can generate?
To answer this question, we have to understand how stake rewards are calculated. The formula can be found on the official doc. A recent excellent video is produced by IOHK for explanation. Other references can also be found below section as well.
While it needs time to digest, the key points are:
- Stake pool will be given rewards only if it’s selected to produce blocks successfully. If stake pool is not functioning somehow and fails to produce blocks when it’s selected, no reward will be given.
- Transaction fee contributes to part of the rewards. Since the no. of transactions inside a block varies, the rewards of successfully minting a block will be a variable as well.
- Pool owner will get the “fixed cost” from the rewards first. Then, he/she can get the “margin” from the rest. The remaining will be shared between all delegators in proportion to their stake. Here, we need to note that the owner’s pledge is also considered as stake and entitled to get rewards.
Cardano staking stated at around Aug, 20. Data is already enough for our analysis. I picked some big as well as small stake pools and checked their rewards. From my observations, the rewards of minting 1 block successfully is around 800₳ to 1,000₳.
To illustrate, let’s assume a simple case:
Active stake: 1M₳, pledge: 35K₳, fixed cost: 340₳, margin: 1%
Assuming the rewards for the whole block are1K₳, then rewards that I can get as pool owner are:
a.) Fixed cost: 340₳
b.) Margin: (1000–340)*1/100 = 6.6₳
c.) The rest: (1000–340–6.6)*35/1000=22.869₳
d.) Total: a)+b)+c) = 369.469₳
The current price of Ada is ~US$1.2. So, I can get US$443.36
How much stake needed?
US$443.36 seems quite a decent amount of money. In fact, it is nearly enough to cover my annual expense of running the hardware. But don’t forget we can get rewards only if we are selected to produce the block. So, my 2nd question is what is the chance of being chosen as a block producer. To answer this question, we need to come back to the basic principle of Proof of Stake (POS) which simply states that the higher the stake, the higher the chance of being chosen.
We need to consider how much Ada in all stake pools in total and how much Ada in our own stake pool. At the time of writing, the total stake pool size is 22.47B.
For example if I have 1M (pledge+stake) in my own stake pool, I have the chance of 1/22470 being chosen. Assume there are 4,320 blocks minted in 1day (24hr) , I have to wait 5.2 days (124.83hr, close to 1 Epoch) before I can be chosen. In other words, I can mint 70.19 blocks every year.
However if my stake pool is much smaller, say 35K, I have to wait 148.58 days (3,566hr). In other words, I can mint 2.45 blocks every year.
Please note that the above calculations are over-simplified. I just want to give you a basic idea on how roughly the figures look like. The actual situation is more complicated. It depends on the parameters stated in the equation that are subject to change. Besides, the actual value will fluctuate due to the stochastic nature, or random process of the Ouroboros Praos consensus algorithm. Furthermore, since the block producers are chosen randomly, it depends on how lucky you are (especially for those pools with small stake).
Own Pool vs Delegation?
With the above 2 questions answered, I then challenge myself whether I should delegate my Ada to existing big pools or build my own.
For the 1st option, the reward is quite predictable. If I choose a big stake pool that is stable enough not to miss any block, the APY is expected to be 5–5.5%.
The 2nd option is more complicated. It depends on the following factors:
- Pledge
- Fix cost
- Margin
- Stake
- Total Stake of all pools
- The values of adjustable parameters in the formula
- How lucky you are of being chosen
My observations are:
- If I run my own stake pool, there is no guarantee when I will chosen as a block producer to get my rewards. This is in contrast to delegating to big pools that I can get predictable rewards for every epoch.
- Need more than 1M₳ (a big amount of money) in my stake pool before I can expect to mint 1 block per epoch.
- For small pools (e.g. 50K), the return is only slighter higher than delegation. It is even smaller if I take into account the operating expenses of maintaining my stake pool.
- The best case is to recruit big Ada holders to delegate to my stake pool. In this case, the return will be much greater than delegate my Ada to others.
In short, I should be confident enough to recruit delegators (through various marketing channels) to my stake pool to make it competitive with others. Otherwise, I should just delegate my Ada to the big pools.
Conclusion
The process of setting up my own stake pool was fun. It gave me an enjoyable experience to learn deeper about Cardano and blockchain knowledge in general.
However, from the analysis I made so far, the current Cardano staking mechanism is not without drawbacks. The biggest issue being the big stake pools’ dominance over the small ones, with > 40% stake pools never produced any block. This in certain sense defeats the purpose of de-centralization that the blockchain platform should never be controlled by small portion of big resource owners.
The adjustment of K and a0 parameters of the reward calculation formula is a solution to the problem. It’s disappointing that these values are not adjusted this month as intended. I really hope that this can be revised in the near future so that small Ada holders like me has the incentive to build our own stake pool.
