Update on - Some interesting observations/analysis for Pool Operators and ADA Stakers

Staking & Delegation Stake Delegation
1

Hi Everyone,

Thank you for the feedback on my original observations located in
Some interesting observations/analysis for Pool Operators and ADA Stakers.

Being a Staker, and not a Pool Operator, my observations about the fate of Pool Operators were based on the stats from the https://pooltool.io/ ](https://pooltool.io/.

I have watched a video by one of my favourite regular podcasters from the Cardano Effect, Phillipe, shown in this link -https://www.reddit.com/r/cardano/comments/ey8eoq/cardano_stake_pool_operator_perspective/.

As a high level summary of this podcast, Phillipe argues strongly, that the current Cardano model of a capped saturation point of 1% of Total Staking beyond which returns for both the Pool Operator and Staker begins to fall will not guarantee a truly decentralised system. As a Pool Operator, Phillipeā€™s experience is that it is unprofitable for the small operators. That in fact, there are some Operators who are creating multiple Pools to achieve a larger percentage of the Total Staking, and thus increasing their staking revenueā€¦as sort of ā€œcorrupting or circumventing the equity model intended by Cardano Stakingā€.

Phillipe is convinced that current the egalitarian model of Cardano staking will likely lead to counter-intuitive centralisation through fewer, well resourced Pool Operators with large numbers of competitive sized pools that can win the right to create blocks regularly and also attract Stakers. Phillipe believes that a free market capitalist system will provide the best model to incentivise Pool operators to maximise their profitability through uncapped percentage of Staking.

My Opinion:
As my observations in the previous posts had shown, that the reality of an open and free market always, without fail, will lead to what economist call ā€œMarket Failureā€. This is why there are Government intervention programmes at all levels of Government to plug the cracks that the disadvantaged fall into, as well as to protect against the brutal and relentless exploitation by those who have financial, political and physical power over the weak in Western capitalist countries.

The answer to the lack of competitiveness of those lacking resources is not in allowing those who already have advantage to gain even more advantage by being able to have uncapped staking leading to immediate centralisation by them. Sure, some of the small operators might earn more, but only for a while so long as profitability of the model is below what the big players see as worth their while. Look at Bitcoin, original small miners have been smashed and are no more, as Bitcoin became lucrative.
I would say, rather it is by fine tuning the game theory models on which staking in Cardano is being developed and perhaps building a Smart Contract into the Pool registration system which will monitor and auto-penalise authenticated registers that seek to directly and/or indirectly aggregate a larger percentage share of the Total Staking than the limit set by the Cardano staking rules.

I believe, the mathematicians and scientists have already figured this out.

Phillipe, should not be discouraged, but during the Testnet may need to temporarily pull-out or consolidate with other Pool Operators to improve competitiveness. After all, that is the very essence of free markets, where viability is part of the survival of the fittest, otherwise its existential.

Look forward to your feedback.

Stevod

1 Like
2

I would agree that the incentives in the protocol will be adjusted accordingly for mainnet. The team Iā€™m sure is preoccupied with a number of different issues that are being fixed in order of priority.

1 Like
3

I agreed that the current Cardano model for decentralization is theoretically sound but human behavior is very unpredictable. Example, a pool reaches saturation. So its rewards are capped. However, the pool operator decides to operate a second pool using his same successful formula (same type server, node location, marketing, etc). Again, he/she successfully attracts more delegates based on the past performance of his first pool. He repeats his magic formula for 10 times and it works each time. Out of sudden, you have one operator controlling 10% of all block creation for Cardano. Itā€™s not a hypothetical. Itā€™s happening right now in the ITN. I would not name the pool. The pool operator is not at fault. In this case, the Saturation Point has no effect of decentralizing Cardano pools. If ā€œmore stakes more blocksā€ approach dominates, I donā€™t see how a smaller pool competes in terms of growth either. Smaller pools, like mine, will always be at a disadvantage until delegates from the bigger pools migrate to a different pool. If the bigger pool is telling his oversaturated delegates to migrate to his second pool, then the smaller pools are going no where. In short, distribution of delegates needs to be decentralized away from one pool operator. Saturation point alone doesnā€™t do it. There need to be a way of restricting one pool operator to one pool. If not , a well-oiled pool operator can start a pool farm and centalized block creation for Cardano.

1 Like
4

Agreed. What your are describing is a Sybil attack. I believe the primary way they will be dealing with this in the future is through stake pool owners pledge. I believe it was not implemented in the test pool as it would have been counter productive to the notion of getting a bunch of pools set up for testing. Interestingly reality did provide a confirmation of the theory that this would be a problem.

More on that all here:

5

Hi All,
When I reflected on my staking habits during this ITN period I found that I was checking daily on my rewards and sometimes adjusting to whom I was staking on a daily basis. Those pools who were producing blocks daily and had a consistent ROI caught my eye and lured me into staking with them. The less saturated pools were less attractive to me as they had not shown that they could produce blocks.

If rewards were distributed less frequently, which would effectively allow smaller pools to show the greater rewards resultant from the production of fewer blocks shared between less stakers, I think I would have been more willing to commit my stake to those smaller pools.

It is my understanding that the slot leader lottery is meant to distribute the rewards evenly per ADA staked i.e.ā€œsmallerā€ pools are rewarded with greater amounts less frequently.

If rewards were awarded weekly, or maybe monthly, or even quarterly, it would hide the fact that the less saturated pools spent many epochs producing no blocks. Wouldnā€™t this even out the perception of how stakers are rewarded, making smaller pools seem more competitive and therefore garner a greater number of stakers?

Less frequent awarding of income derived from staking might also make management of taxes an easier prospect.

Cheers,
D

1 Like
6

I M doing nearly the same: looking every day and sometimes switching. The difference between us is that i decided to delegate my ada, and im not a whale, to smaller pools, preferably located in the netherlands or somewhere out of asia/usa. I switched once a week. It is not a strategy that gives me the highest rate i guessšŸ˜‰, but it feels good.
Your suggestion of not giving the wards on daily basis is very good en honest. I thought that ive read the mainnet should have a epoch/reward time of 5 days,

7

@HarryJacques Coincidentally, right after I tweeted the below link, I read your comment. Youā€™re giving smaller pools a fighting chance to truly and fully decentralized the Cardano network.

2 Likes
8

Yep, the present pool has 500k, the former 900k, and the one before has at the moment 48k. All are situated in the Netherlands and have their own hardware, not cloud based My plan is to switch between them after a week or so.

9

Yes, I have mentioned this in the past (back when everything was still being researched). The notion that the majority of people will put the idea of decentralisation over their own financial gains is naive.

Of course all those from the early days who believe in the social implications of blockchain and want decentralisation to be a core tenet may be prepared to take slightly less profit to support the idea. But the majority donā€™t care. It will only work if increasing decentralisation causes rewards to increase.

That is why I have always taken Charlesā€™ comment about Cardano being 100x more decentralised that Bitcoin with a grain of salt. Yes, there might be a thousand stake pools. But half of them may as well not be there.

I think the pledge idea is a good one for preventing a Sybil attack, but I am sure we will see syndicates forming to collect the funds required for a competitive pledge (hopefully just not enough to control 51% of pools :slight_smile:)

10

Isnā€™t what Binance is doing right now a confirmation of the described above? Those who have the highest stake will always be able to increase their stake in the most efficient way possible compared to minor ADA owners, which will naturally lead to increased centralisation no matter the protocol parameters.

11

Hi, this is a super discussion that needs to be had at a much higher level. I have stumbled on posts like these multiple times and the message is always the same - how to avoid having majority of stake in a handful of participants.

I had a think about how could this be fixed

As any economic systems agents will behave selfishly and that is not necessarily a bad thing. Think of Adam Smith and the invisible hand ā€¦ where the baker produces bread not because he is worried about the community, but because he needs to make a living. The invisible hand of the market guides decisions.

Then there are market failures in a free market economy, such as health and educations where the optimal provision in a truly free market economy might be suboptimal. And the society as a whole benefits if there is an authority that ensures there is a minimum provision through mandatory actions (you must go to school) or though incentives (please come to school and we will give you free lunch). Without digressing too much, the point is that a market failure can be fixed with correctly designed incentives. And the changes often donā€™t need to be dramatic, but rather a nudge in the right direction.

Everyone reacts to incentives.

So a few key principles which I think many would agree on concerning staking:

a. Delegators will move their capital to a pool that has the potential to generate highest returns

b. It is preferred to have the stake distributed across as many pools as possible, and as evenly as possible

c. avoid the possibility of rogue pools being created and attracting stake

How the right incentive should look like:

1 - give smaller pools a higher % reward to compensate for more volatility in the return as smaller pools have more variability in the number of blocks they produce. Standard risk / reward trade off

2 - give pools that have been around for longer a slightly higher return, not significant, but higher. Besides eliminating the possibility of new pools spring up and attracting stake, it also incentivizes the operators to be in the game for the long term

I have found that a lot of stake holders are first after a higher return, and second after a meaningful pool with a purpose. Rarely are the priorities flipped. And this is fairā€¦ if someone is holding ada and their main mean of subsistence is a staking return then they will look to optimize it first.

Solution 1 would in my view give the most leverage to attract delegators to smaller pools as it drastically changes the conversations that can be had. From ā€œā€¦ stake with a smaller pool as it benefits decentralisation ā€¦ā€ where decentralisation is still ambiguous to many, to ā€œā€¦ stake with a smaller pool because it potentially gives you a higher return, although with more vol ā€¦ā€

Happy to elaborate on any of the points

All the best

1 Like