PoW provides significant economies of scale for those with lots of scale and makes mining with a single GPU a waste of time. PoS doesn't offer much economies of scale at all as the main cost is the ETH, so it's much fairer to smaller validators.

When you put your money in a bank, it doesn’t just sit there. Deposits, while possibly not the most efficient way to do it, support useful economic activity.

Proof of work is also terrible, obviously. Proof of stake seems like an awful thing to _aspire_ to, tho.

The minimum stake of 32 eth is probably going to require something like a laptop to run. Those with larger stakes are going to need to devote more computer resources to the network.

32 ETH is ~$4,800 at today's price of $151.07[1]. What's the expected "payout" for this stake?

[1] CoinMarketCap used as source.

Expected payout depends on the total amount staked. With a large amount staked it's only a couple percent annually; with a small amount it's up to 18% annually. The idea is an equilibrium will be reached in the middle somewhere.

How are participants prevented from simply splitting up their stakes?

The return isn't based on how much you stake. It's based on how much everybody stakes. It's targeting an approximate amount of total ETH staked and doesn't really care how much is staked by each individual.

That's an interesting formula.

More than the interest paid on savings accounts.