[SHOCK] A Mysterious $194k Fees On A Single Bitcoin Transactions

[pooya87]

Is there any way to tell for sure whether this transaction was mined in secret or was broadcast to the network prior to being mined?

The only reliable way is to run a modified full node that first removes the limit on its mempool size and also has a good connectivity to the rest of the nodes and is running for more than a day. Then that node would mark transactions that aren't in its mempool but are in a block so that they can be analyzed. These transactions could include the payment of the pool to its miners which they don't broadcast (and usually has a zero fee) but also these shady looking transactions.
This is something a blockchain analysis company could do.

[o_e_l_e_o]

This is something a blockchain analysis company could do.

I'm sure some already do. But yeah, so there is no way for us to discover whether or not that transaction was broadcast only using retrospective data from publicly available block explorers.

As above, there are block explorers which record the time that they received a transaction as opposed to only recording the timestamp of the block it is included in. Are there any block explorers which do the same thing with blocks, recording the time that they received a block as opposed to only recording the timestamp of the block? If there were, then the combination of the time the transaction was first seen and the time the block was first seen could be enough to deduce if the transaction was broadcast prior to being mined.

Still, for such a niche case as this, it's not exactly high up on the priority list.

[PrimeNumber7]

Is there any way to tell for sure whether this transaction was mined in secret or was broadcast to the network prior to being mined?

The only reliable way is to run a modified full node that first removes the limit on its mempool size and also has a good connectivity to the rest of the nodes and is running for more than a day. Then that node would mark transactions that aren't in its mempool but are in a block so that they can be analyzed. These transactions could include the payment of the pool to its miners which they don't broadcast (and usually has a zero fee) but also these shady looking transactions.
This is something a blockchain analysis company could do.

This is not guaranteed to determine if a transaction was not broadcast prior to being included in a block.

Someone could mine block n at timestamp x that includes a previously unbroadcast transaction, broadcast said transaction at timestamp x, and broadcast block n at timestamp x + 6 seconds. This would put the chances of another pool including the transaction in another block at around 1%. This chance could be reduced by the delay in broadcasting the block is reduced.

A pool could also broadcast a found block only to a select number of other mining pools, and broadcast the transaction at the same time, but via many very well-connected nodes.

As for the 1st scenario, a poll could charge a 10% fee for including this transaction, and refunding the transaction fee, and could still offer this service profitably if they don't end up being the one confirming the transaction 1% of the time.

[nc50lc]

Fascinating, based from the address' history, it tends to send small amounts from time to time but with change that sends the rest back to itself.
Someone must have messed up coin control or the manually generated txn

Less experienced people or people who dont know how to make use of the Bitcoin transaction estimator site will always pay ridiculous transaction fees when the network is in a congestion state.
Don't be surprised more are still coming or have already happened which we don't know.

It's not about the typical ridiculous fee due to the "congestion" simply because it's impossible to reach that fee with "wrong estimation" for a '1in-1out txn'.
The transaction was obviously created without a change (by accident or with reason) so all of the excess amount were considered as "mining fee".

[pooya87]

Someone could mine block n at timestamp x that includes a previously unbroadcast transaction, broadcast said transaction at timestamp x, and broadcast block n at timestamp x + 6 seconds. This would put the chances of another pool including the transaction in another block at around 1%. This chance could be reduced by the delay in broadcasting the block is reduced.

I don't know how you came up with 1% chance but it doesn't seem right.
In any case the risk of doing this is that someone else could find the same block in that delay time and broadcast it to the network where everyone else starts working based on that block. Result of this is losing nearly $400k, in other words the cost is just too high that doesn't justify the act.

[PrimeNumber7]

Someone could mine block n at timestamp x that includes a previously unbroadcast transaction, broadcast said transaction at timestamp x, and broadcast block n at timestamp x + 6 seconds. This would put the chances of another pool including the transaction in another block at around 1%. This chance could be reduced by the delay in broadcasting the block is reduced.

I don't know how you came up with 1% chance but it doesn't seem right.

The difficulty is adjusted so that blocks will be found, on average, once every 10 minutes, or once every 600 seconds. This means the chances of a block being found in any given 6 second period is 1%.

In any case the risk of doing this is that someone else could find the same block in that delay time and broadcast it to the network where everyone else starts working based on that block. Result of this is losing nearly $400k, in other words the cost is just too high that doesn't justify the act.

If there is a 1% chance of loss, and the amount of the loss would be $400k, as long as the pool stands to gain an amount greater than $4k, the expected value of doing this is positive.