Azz on Nostr: Back of the envelope: ~3,300 tx per block 52,560 blocks per year ~175M on chain ...

Back of the envelope:
~3,300 tx per block
52,560 blocks per year ~175M on chain tx/year
6 channels per user, ~2 year lifespan ~6 on chain tx per user/year

That supports ~29 million self custodial Lightning users worldwide if every block is used for channels and nothing else.
That’s the ceiling under generous assumptions. Reality is lower.

So scaling to billions doesn’t happen via sovereign users. It happens via: • fewer channels per user
• extremely long lived channels
• shared channels
• custody

And that’s where the centralising pressure comes from.
Lightning shifts economic power from proof of work → proof of liquidity.
Large hubs dominate because they have: • deep capital to lock in channels
• cheaper rebalancing and liquidity management
• better uptime and routing reliability
• pathfinding preference → traffic concentrates

Routing revenue compounds to the largest nodes. Smaller nodes become edge users, not infrastructure.
Custodial Lightning removes the need for users to open channels at all. At scale, that looks indistinguishable from banks running payment hubs with internal ledgers.

Meanwhile miners lose relative influence: • fewer on chain transactions per user
• suppressed fee growth as subsidy declines
• security budget pressure over time

Lightning works. But its incentive structure rewards liquidity concentration and custodial aggregation.
That’s not a bug. It’s the economic gravity of the system.

The real question isn’t whether Lightning works.
It’s who ends up controlling the rails when it scales.