Let's forget "ETH PoW" as it is completely irrelevant. All of the community has moved on to PoS, all the relevant projects were kept on the fork, etc. To talk about "ETH PoW" on the same footing as ETH or BTC shows that you are basing your arguments on ideology and not objective truth.
Besides that, your breakdown has other problems:
- None of these three criteria you presented are binary attributes. How decentralized and secure are BTC vs ETH? How performant is ETH vs BTC?
- One could easily argue that PoS is more decentralized than BTC ever will. The capital and resources to start a viable node on Ethereum is a lot lower than it is to get a sustaninable/profitable BTC miner.
That is not true. This shows a terrible misunderstanding of how PoS works.
Stakers are only rewarded in proportion to their stake. "Power" in PoS comes from being a validator node, which requires the minimum stake of 32 ETH. Having more ETH on a node does not give them any power. Yeah, whales could try to create many nodes, but they would need 66% of the nodes to even consider the possibility of blocking a transaction or making a double spend.
So you are saying that by some crypto-magic it is technically impossible for groups which collectively own more than 2/3 the coins to create enough nodes to dominate the protocol?
Because that would be news to me and if so I would be misunderstanding things.
But my bet is that you are dismissing the threat of whales splitting their holdings to maximize their influence. Don’t know why though.
> So you are saying that by some crypto-magic it is technically impossible for groups which collectively own more than 2/3 the coins to create enough nodes to dominate the protocol?
Yes, that's exactly what I am saying. Not because of "crypto-magic", but because that is not how PoS works. Stakers have no control over the protocol. Their stake only gives them the benefit of validating and producing blocks and collecting rewards for it.
> But my bet is that you are dismissing the threat of whales splitting their holdings to maximize their influence. Don’t know why though.
You mean "joining their holdings", in order to achieve 2/3 of the staked capital?
If so, the threat is possible but incredibly stupid. There is more than 10M Ether locked in the staking contract. This means that they would have to control two-thirds of the tokens (that they do not have, which means that they will have to go to the market and shoot the price up). And even if they did get to collectively have that much, they would be willing to risk losing it all on an attack that (a) could be easily detected (b) would have to entail a single profitable operation that is more profitable than the current stake they are holding and (c) would lead to people losing faith in the ecosystem and consequently plunging the value of the token they have at stake and, worst of all, (d) if needed be, could still be remedied with an irregular state change like the DAO hack was.
IOW, even though such an attack is theoretically possible, in practice "malicious whales" could do nothing but bring a major inconvenience to the thousands of the developers, and they would have to spend ~$7B for it.
Those people don't have any governance power, they can't change the rules of the game. They only have the power to double spend or change the order of transactions in blocks.
It costs currently over $10B to pull off this attack, and the rest of the network can do a fork and burn all your tokens as soon as they notice. So game theory wise it makes no sense.
> They can’t block certain protocol updates by continuing to use old/custom clients?
They could try, but if that was clearly against the interests of the rest of the community, the minority could counteract by adopting a client that forked the chain to a state where the majority holders are slashed.
Besides that, your breakdown has other problems:
- None of these three criteria you presented are binary attributes. How decentralized and secure are BTC vs ETH? How performant is ETH vs BTC?
- One could easily argue that PoS is more decentralized than BTC ever will. The capital and resources to start a viable node on Ethereum is a lot lower than it is to get a sustaninable/profitable BTC miner.