Nomination pools are just around the corner for Polkadot, and this suggests some changes to be applied to the staking system.
First, let’s recap on the current status quo: Currently, nominators can freely set their intention to nominate in Polkadot, as long as they meet two limits: a minimum of 10 DOTs is staked, and a maximum of 50_000 nominators is not filled. This creates a nomination market, where in every era, the top 22_500 nominators from this open market are chosen to participate in the election process, and possibly receive staking rewards (if other conditions like over-subscription are not met).
This is a complicated and involved process. Nomination pools are exactly here to combat this and provide a simpler experience for DOT-holders to participate in staking.
Based on this, let’s cluster Polkadot nominators into two groups: active (in the top 22500), and waiting. The main concern at this point is to
To achieve the second goal, we can set the nomination limits (10 DOTs, 50_000) such that the system is not so inviting to waiting nominators. In other words, the system should try and prevent nominators from being created, if they are certainly going to end up being in the waiting queue.
For example, we empirically know that the actual amount of DOTs needed to be an active nominator is around 130 DOTs as of this writing. Therefore, we can set the minimum nomination limit to a higher value such as 50.
Note that keeping a buffer of waiting nominators is a positive factor for Polkadot's economic security. It allows for buffer of waiting nominators to be ready, should a (malicious) cluster of nominators decide to chill all at once, preventing nothing at stake issues. Therefore, I am not suggesting lowering 50_000 as well, and find that a reasonable bound. But, with nomination pools, I do think we should immediately forward anyone with less than e.g. 50 DOTs to join pools instead, and the way to do that is to immediately reject their nominate transaction.