Yes, I expect major pools to drift toward OFAC/blacklist templates (or "risk-graded" inclusion policies) unless the miner–pool protocol surface changes (Stratum V2 with job negotiation/Ocean's DATUM + non-custodial pooling) and enough hashrate actually uses it.
Security against reorgs/51% still scales with hashrate. So hashrate increases are not completely irrelevant.
But censorship-resistance does not scale with hashrate if template control remains centralized.
So marginal hashrate under the same pool topology is mostly irrelevant for freedom. Marginal hashrate under miner-templating topology is highly relevant.
What I expect (base case path)
- Soft censorship becomes normal: not outright bans, but risk-weighted delays and fee penalties.
- Pools posture "we follow law", insurers and utilities quietly codify it; most retail never notices — TXs just "feel slower unless you KYC".
- Effective Free Hashrate (EFH) stagnates unless miner-templating spreads; raw hashrate hits ATHs; headlines say "record security", while settlement becomes more steerable.
- Result: BTC functions as supervised SoV with compliant MoE corridors; non-compliant MoE becomes niche/expensive.
So the real metric you care about is Effective Free Hashrate (EFH).
EFH ≈ total hashrate × share using miner-controlled template selection (Stratum V2 job negotiation ON/Ocean's DATUM, or non-custodial pools) × (1 − censorship correlation across pools).
If Effective Free Hashrate (EFH) isn't rising, censorship cost isn't rising, no matter what the difficulty says.
TL;DR
With today's pool topology and perimeter levers, large pools are likely to adopt OFAC/blacklist-style templates (explicitly or implicitly).
Without miner-level template control becoming mainstream, more hashrate mostly strengthens the appearance of security, not the censorship-resistance Bitcoiners actually care about.
View quoted note →
Login to reply
Replies (8)
MARA already did OFAC transaction censorship. They stopped because their blocks were getting orphaned by other miners, losing them block rewards. Nobody wants that.
Censoring blocks don't get orphaned. MARA was mining blocks that did not have the highest fee transactions.
If a miner/pool builds a valid block that merely omits certain transactions (OFAC list, "risk-graded" UTXOs, etc.), other miners do not orphan it.
They extend the heaviest valid chain tip. Orphaning a valid-but-omissive block would require:
- A coordination rule (social soft fork) to reject such blocks, or
- A protocol change that makes omission invalid (not happening without civil war).
So a censoring miner's blocks are paid normally. The cost they bear is opportunity cost (they skip some fees) and potentially slightly higher stale risk if their template construction is slower. That's it.
I have not extensively looked into why MARA stopped, but my guess is it was a short-term test for the future. Predictive programming if you will.
Fred Thiel would sacrifice his shareholder base in a nanosecond if it means he gets to sniff Trump's jockstrap.
MARA was doing OFAC and their blocks were getting orphaned. Which is why they eventually were forced to capitulate. It's not about fees but about timing of other blocks being mined. It's wasn't 100% of the time, but enough to hurt their profits. Yes they were concensus valid, but nobody else was doing OFAC, nor does anyone want to. Why do you even care about OFAC?
Maybe try to read my comment again. Censoring blocks don't get orphaned.
MARA's blocks were valid (just included fewer txs/fees).
They allegedly abandoned the experiment within weeks amid industry backlash and to align with Bitcoin Core 0.21.1/Taproot signaling, saying they would validate like everyone else (no censorship).
Orphan/stale events happen for timing races, not because a miner filters tx by policy if the block is otherwise valid.
So again, none of MARA's OFAC-compliant blocks were orphaned.
BitMEX Research noted one of the OFAC-compliant blocks contained ~0.0033 BTC less in fees than a normal maximized template would have, implying foregone revenue — not invalidity.
I can't find any reference to their blocks being orphaned, so perhaps I'm wrong. In any case, it wasn't effective since they were the only miner doing it. I highly doubt the Chinese miners would do it. Still not sure why it's important since it wasn't effective and isn't happening anymore.
It being effective or not depends on their true intent, which we don't know for sure.
I'm pretty sure F2Pool (China) was already caught censoring transactions multiple times.
-
-
It is important to me because it's an attack vector and I like to know about these things. If it's not important to you, it's all good 🤝

Bitcoin News
Bitcoin Mining Pool F2pool Acknowledges OFAC Transaction Censorship; Backpedals After Community Backlash
F2Pool, a Bitcoin mining pool, has admitted to filtering Bitcoin transactions coming from addresses flagged by OFAC.

Atlas21
F2Pool accused of transaction censorship and vulnerability in Bitcoin Core
Last month, developer 0xB10C discovered that the US mining pool was filtering transactions linked to OFAC-sanctioned addresses. A bug was
And by the way, you maybe got confused because in a separate event, much later, Marathon did mine an invalid block in Sept 2023 due to a transaction-ordering error — unrelated to OFAC filtering and years after they'd stopped the "clean block" trial.
If different pools censor different transactions it can delay them but won't prevent them. I don't think it would have any effect on L2 either.
You're right, it is important to think about. I guess I'm more focused in node implementation centralized and arbitrary data storage attacks right now.
🤝