I don’t understand fee structure. Mempool still very full , yet fees lowering. Why’s that?
Discussion
Last I checked, the mempool typically has a max MB (in memory) size cap. When full, it drops the lowest sat/vB transactions. Normally transactions get dropped after two weeks of no confirmation, in the case the mempool never overflows. The mempool also rejects dust (too smaller transactions), as basically it can be a spam attack vector.
The fees paid are typically an urgency calculation. If your transaction is less urgent, and still above the current purge sat/vB limits, then you may just have to wait longer.
Hopefully it’s a sign of inscribers running out of their VC funds, shitcoin bubble, or trust fund depleting. More likely it’s just a breather.. there is no urgency in inscribing more crap that you can’t sell and paying a higher fee to get it quicker. Rate limiting your transactions makes it cheaper.
Thanks 🤙
I see what you’re saying and makes sense. So only if people are pumping their fees to guarantee next block then fees will rise so drops as most are now putting lower fee ? 
It could still be many things.
Exchanges and other regular transaction entities (to allow account withdrawals), or OpenTimestamps, don’t have much choice. They grind to a halt.
However people who want to open Lightning channels may way for it to ease.
And people may use lightning instead of Bitcoin to pay while fees are high.
Anyone planning on consolidating will defer as largely no rush.
Likely a typical lag after a few spike, while people and businesses adjust their transaction schedules and usage down.
It’s expensive, however it’s still pretty interesting to see how easy it is to cause mempool related transaction delays for Bitcoin. Overall it’s just growing pains and a catalyst for innovation and adoption in other areas like lightning.
Also, I’m not sure if mempool accounts for it, however you can bump transaction fees using a child transaction.
Let’s say I pay 150sat/vB, realise i need it sooner, I bump it with a second transaction for 50sat/vB, being slightly imprecise, I’m not sure if mempool counts that as a 200sat/vB or 150 and 50 in their statistics. So many of the smaller transaction fees could be trying to bump (stuck) existing higher fee transactions too. You could write a script to investigate it.
Obviously two transactions means more data, so eventually they clear and get mined and the transaction count can drop, as people’s initial transaction fee is higher/better calculated based on their needs.