A Sobering Reality for Cardano Stake Pool Operators & Owners

@Razvan @vantuz-subhuman I meant, why would someone run a smaller pool (besides contributing to te network) if your returns in a year would be greater if you just delegate your ADA to a large pool?

Edit: But I’m not sure this is the case.

Because a larger pool cannot get larger at some point

In Cardano the maximal proportion of the rewards that a stake pool can receive will be limited by 1/k, where k is the number of desired number of pools. This policy is so beautiful simple and effective.

It looks like k will be 1000, so this means:

max reward for each pool = total reward / 1000

Let’s say the total reward is 10000, this means each pools max rewards is 10. No matter how many pools there are. This will force “the system” to have ideally 1000, but probably more pools. → it will not happen that one pool will have 80% of the total supply because everyone would lose money, and people don’t like this.

5 Likes

Please read this as a question and hopefully a correct contribution to this topic:

Certain pools will be able to reach their primary goal “high desirability” with high pledges, heavy investment in service uptime or low costs (or an interesting marketing strategy when it’s a heavy investment and low cost OR low investment, low cost and a claim to guarantee five-nines)

But as soon as these first listed pools will reach their saturation point, it’s not anymore in their interest to be listed on top, because it will end up with additional delegations and a lower revenue to all (also them who delegated to what they saw before as most desirable/profitable).

Other users - who start looking for a pool when the most profitable ones was already saturated - probably will look, understand and delegate to “less desirable” but in this moment “most profitable” pools.

Then we probably will also see delegations removed from currently saturated pools, because users spend their money, move it to other investments or re-delegate to a free pool (because all these nasty bas…s who delegate to a saturated pool without understanding what they are doing).

So bottom line: it looks to me like a pretty dynamic field with plenty of chances and ways.

5 Likes

Yup - to run a pool that is listed favourably to the public - you will need a lot of ADA.

And there is a miss-match between the fan boys who want to run staking pools (English speakers: USA, Europe) and the people who actually hold the ADA (Japan, Asia).

People actually think their 100k ADA is going to make them some kind of major pool operator. Salty tears are coming.

Hi Ruslan

I just want to break this down a bit more to either better understand the point being made by @Razvan or better understand your response.

So let’s says that regardless of whether there is a pool of 50 people’s stake or a single person, it is chosen to create a block in a given slot.

Firstly, I assume the probability of that pool/person being chosen for that slot is proportionate to the percentage of their stake in relation to the total stake. Let’s say 0.0001% for this example. Is this assumption correct?

Next I like @Razvan assumed that the actual reward for creating the block for that slot was a fixed amount. Not necessarily fixed as in 100 ADA, but fixed in as much as whoever had been randomly selected for that slot would have received the same amount regardless of their percentage of stake.

It sounds however like you are now saying that the reward will be “watered down” to be an amount based on their percentage of stake.

So if the reward was say 100 ADA, then our example pool would get 0.0001 ADA, whereas if another pool chosen at random had 0.1% of total stake they would be given 0.1 ADA. Is this what you are saying?

If so then it may be considered unreasonable that you receive a lower reward for performing exactly the same task (i.e. creating a block) and make it difficult for some small pools just to cover costs.

If however everyone received the same reward, then their total reward over time would be proportionate to their percentage stake due to the probability of them being selected multiple times over a given period compared to another pool.

Thanks

He will probably reply you, but since I am bored, this is how I understand it…

The chance of a pool being chosen to make a block is directly proportional to the total amount staked in that pool.

The pool is rewarded for making the block. The reward is then distributed proportionally to the users staking in that pool.

Basically it’s a similar reward regardless of the pool, however there is a slight difference:

When you make a pool, you can commit a certain amount of ADA to that pool. Consider two pools:

  1. owner of pool commits 10 ADA, stakers stake 90 ADA
  2. owner of pool commits 50 ADA, stakers stake 50 ADA

Both pools are equally likely to be chosen to make the block, but the second pool will get a slightly greater reward, because the ratio of commitment ADA vs staked ADA is higher. This is necessary, to discourage people from creating many pools.

2 Likes

Yes, correct

Nope, incorrect

Yes

The number of blocks you are selected to produce is proportional to your stake as well, so on average over time it would be the same, it’s just that it’s easier to distribute it according to the stake share.

It’s the same thing, but other way around. If you own 0.00001% of total stake - you are guaranteed to be eligible for 0.00001% of total rewards (as long as you produce all the blocks you were selected to produce). I don’t see how it’s not fair.

Nope. Pools receive reward proportional to their stake, regardless of the exact number of blocks they were selected to produce. It’s just that this number of blocks will be in the same proportion over time.

I did not know that. So even if a pool makes zero blocks in an epoch, they still get a reward?

And transaction fees also get distributed like this, across all pools?

Nope. Pools only receive rewards if they were selected at least for one slot, and if they have shown non-zero performance.

Yes.

Hmmm, ok - I’m trying to think this through… does that very slightly advantage smaller pools?

thanks for this easy to read explanation.
I wonder and try to bring it up here, because I always read from “higher reward” for high pledged pools.

As resources (fees and delta to max supply) are given, a “more” reward to high pledges in my understanding must mean, there is also “less” reward for the lower half (pledged) pools.

This brings up the question:
How much more reward in % are we talking about? because this gives some “tolerance” above the 1/n saturation limit. And - assuming the high<>low theory is right - the lower pledged pools with not have at disposition the full 1/n size.

I expect the a0 parameter will have some impact but will not separate high and low pledged pools in 150 - 50% (from 1/n)

In my opinion the right setting for a0 is to let some people only and permanently looking at the slightly highest possible reward (not caring about anything else) versus another group of participants who simply want to stake but don’t spend any time every 5 days (epoch) to permanently re-optimize their delegations. They may prefer a “never top listed”, friendly, educative, supportive, inconspicuous, … pool, with a stable average revenue and not that much artificial drama.

Potentially, yeah. Like if you have barely enough stake to win on average 2.2 slots per epoch, but keep winning 2 - you still gonna receive rewards for 2.2, etc :man_shrugging:

Sorry mate, it sounds unreasonable to me.

You are basically saying that a 0.0001% stake pool will be selected 0.0001% of the time and when they are selected they will get 0.0001% of the reward for that slot.

So if the reward for every slot was 100 ADA and there were say 1 million slots (total of 100 million ADA), the 0.0001% pool will receive 0.01 ADA over this period (i.e. be selected for 1 slot and then receive 0.0001% of the slot reward of 100 ADA); whereas a 0.1% pool will receive 10,000 ADA (i.e. be selected 1000 times and receive 0.1% of the reward of 100 ADA).

The smaller pool is 1000 times smaller than the larger, yet receives 1 million times less reward. Reasonable???

1 Like

I basing my last post on:

I see what you are saying… I’m getting the feeling (as often happens when trying to communicate complex systems) that we all have a slightly different idea in our heads as to how it will work… I am going to wait for the staking testnet, at which point it will all be transparent.

1 Like

There’s no “reward for a slot”. I am saying they will get 0.0001% of the TOTAL reward. There’s no reward per slot in Cardano, conceptually. Reward-per-block is Bitcoin and PoW model. In Cardano we have reward per epoch only. And all pools will receive portion of the total reward always proportional to their stake.

Think of it as all rewards for all blocks going into one big pot, and then at the end of the epoch - they are fairly distributed among all pools, always according to their stake.

6 Likes

OK, now we are getting somewhere. @Razvan does that clarify it for you too?

Sorry, one last clarification. When you say

I presume by “all pools” you mean all pools that were selected for a slot in that epoch, or are you saying ALL pools, even those that weren’t included?

1 Like

Ah yes. Sorry. It’s my constant problem that I don’t make this point clear. Not the first time I confuse someone by this :smile:

All pools that participated in an epoch as slot leaders will receive portion of the total epoch reward pool, proportional to their stake share (with a correction for their performance in this epoch)

1 Like