Guys,
would you consider the increasing mempool as an indication that the transactions on bitcoin are increasing or would be the price for a tx a better indication?
Thx for your thoughts
use mempool.
the reason to not use tx fee as the indication is that the tx fee is no longer 'reactive' but instead its now 'averaged'
meaning if a time period had 2500 transactions then the next time period had 0 transactions. the 'estimated fee' would go down due to low demand on the old 'reactive' metric. but due to 'averaging' the tx fee remains high even when theres no transactions waiting
here ill show you.
imagine the demand was average.. a block can take 2500tx and is happily seeing 2000tx. and lets say for 24 blocks people pay 0.0001 just for the sake of it.
now imagine the next block(25th) averaged 0.00250 fee due to a fee war maker. even when tx demand is low.. .. here is the results of what the "average" metric does to the estimated fee, simply due to one block with a high fee, triggering it
![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FVCsbAiH.png&t=665&c=17q0lhjV8pB-Zg)
EG
block26 is not 0.0001 again. instead it goes up to 0.000196 due to 1-24 at 0.0001 and 25th at 0.0025
block27 goes up goes up further because theres only 23 blocks at 0.0001, a block at 0.0025, and now a block at 0.000196 swaying the average up
block27 goes up goes up further because theres only 22 blocks at 0.0001, a block at 0.0025, and now 2 blocks over 0.000196 swaying the average up
so even without actual tx's waiting.. the fee estimator (fee war game) is causing issues unrelated to actual transactional demand.
note. the example above is not the code concept of bitcoin. its a demonstration that averages rather than reaction to demand, can affect the price and has little to no correlation to demand