I was just browsing the following mempool visualization and couldn't come up with a good explanation why the drops in unconfirmed transaction count are not of pretty much constant height but vary wildly.

My assumption is that every drop represents a newly mined block. Why are there apparently some blocks which contain very few or almost no transactions while others in contrast contain a relatively large number of transactions?

My only theory that miners are just too stupid to include all the available maximum fee transactions up to the maximum block size they can find seems .. too stupid to be true.

Please explain! Thanks :)

enter image description here

2 Answers 2


A transaction is not a fixed size. Removing one large one will have no impact on the count, but removing ten thousand small ones will.

  • Oh! I get it! And large ones are primarily defined by MANY inputs and/or MANY outputs, right? This starts to make sense, thanks! Still kinda strange, that this results in such a huge variation across blocks.
    – cryptoNoob
    Jan 5, 2021 at 12:59
  • Additionally, this graph only has a data point every minute, and sometimes more than one block is found in a minute.
    – Murch
    Jan 5, 2021 at 13:31
  • Oh, fair point Murch, this better explains HUGE drops
    – cryptoNoob
    Jan 5, 2021 at 14:01

Not every drop is because of a block. Some are dropped because they have been in the mempool for 14 days and it's time for them to be dropped (this is set using mempoolexpiry in Bitcoin Core)

But yes, this graph shows a large drop because of a block.

  • 1
    oh this is also a good point, thanks, interesting
    – cryptoNoob
    Jan 5, 2021 at 14:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.