Akshually, I think I have a better definition.

The metric shows how close a mined block came to including the “ideal set of transactions” in terms of “miner incentive,” i.e. the theoretical most lucrative lineup of transactions based on fees. And we all know that most spammy txs get a break from the witness discount, which makes their data cheaper, thus allowing them to bid effectively against normal monetary txs.

So if a miner deliberately leaves out spammy high fee-rate transactions, the block ends up with a lower health score, because from the metric’s perspective they left money on the table.

That’s an idiotic way to describe a block’s “health” imo. It’s really more like a miner’s paycheck “health”. A more intuitive measure might be the proportion of monetary transactions in a block versus all other types of datacarrier transactions: if all are monetary, it’s 100%; if it’s 50-50, it’s 50%; if zero, it’s 0%, and so on. Maybe the dudes at mempool.space might consider renaming it. Just my two sats.

yeah we know that. what someone's mempool has to do with the generated block? by same logic then you also have to lower the health when they add additional txs to blocks that would other wise wouldn't be there.

And they were expected by a specific mempool and block construction policy.

There is no way to define block “health” without ignoring the tens of custom block construction algorithms and variations between versions

> Unless you are trying to say about 30% of the block was spam and that's why they were expected but not included but that assumes the miners decision.

That’s exactly what I was trying to say.

i wonder what is healthy for bitcoin the money.