I compared gold and Bitcoin across every 10%+ S&P 500 drawdown since August 2016. BTC was negative in all five.

Today's divergence is what pushed me to run this test: gold futures gained about 2.5% while Bitcoin fell about 1.5% on August 10 closing data.

But one day is not evidence. So I tested the broader claim.

Disclosure first: I have significant personal exposure to gold and own no Bitcoin. (I have in owned it in the past) That gives me an obvious potential bias. I tried to deal with it by using a fixed rule, checking the numbers against independent data and including the strongest counterexamples to my own thesis.

Method : I used daily closes from August 10, 2016 through August 10, 2026.

1. Identify every S&P 500 price-index drawdown of at least 10%, from the previous closing high to the eventual closing low.

2. Measure continuous COMEX gold futures and BTC/USD over those exact dates.

3. Cross-check the S&P 500 with FRED, gold with the official GLD archive and Bitcoin with Coinbase data published through FRED.

This is an ex-post stress test, not a timing strategy. The trough is only known after it happens.

(This was meant to be a graph and is available as such on my blog but reddit dosen't allow me to post a graph here)

Episode S&P 500 | Gold futures | Bitcoin |

Feb. 2018 correction -10.2% | -2.6% | -26.0% |

Q4 2018 -19.8% | +5.1% | -37.4% |

COVID crash -33.9% | -2.5% | -33.4% |

2022 bear market -25.4% | -7.2% | -58.8% |

2025 drawdown -18.9% | +1.7% | -21.1% |

Bitcoin was negative in all five. Its median return was -33.4%. Gold was positive in only two of five, so this is not a claim that gold always rises in a crisis. Its median return was -2.5%, and it lost less than Bitcoin in every episode.

Independent checks

The FRED S&P 500 series reproduced the same five peak dates, trough dates and drawdown percentages.

The official GLD archive returned -2.4%, +5.0%, -3.6%, -7.3% and +1.6% over the same windows. That is almost identical to the gold-futures result.

Coinbase Bitcoin data through FRED returned -25.9%, -37.9%, -31.4%, -58.8% and -16.0%. Exact BTC returns differ from Yahoo because a 24/7 market needs an arbitrary daily cutoff. The robust result is unchanged: negative in all five, with a median of -31.4% on Coinbase versus -33.4% on Yahoo.

What happens if the threshold is only 5%?

That expands the sample to 13 S&P 500 drawdowns.

- Gold was non-negative in 6 of 13, with a median around -2.3%.

- Bitcoin was non-negative in 2 of 13, with a median between -16.9% and -21.1% depending on the closing source.

The best counterexample for Bitcoin was the April-to-June 2019 correction. The S&P 500 fell 6.8% while BTC gained roughly 52% on Coinbase data. Bitcoin does not fail in every risk-off period.

Gold also failed badly during the January-to-March 2026 correction. The S&P 500 fell 9.1% and GLD fell 12.9%. BTC fell 25.3%, but gold did not protect capital in that episode.

Iran is a warning against cherry-picking

The first week of the 2026 Iran war actually favored Bitcoin: from February 27 to March 6, gold futures fell 1.6% while BTC gained 3.4%.

The July re-escalation also favored BTC over the next five S&P sessions: gold fell 2.0% and BTC gained 2.6%.

The latest seven-day snapshot favored gold: from August 3 to August 10, gold gained 10.3% while BTC gained less than 1%.

These Iran windows were selected retrospectively from the AP chronology. They are context, not a preregistered event study. Depending on the week, either side can cherry-pick a win.

My conclusion

If "safe haven" means an asset that always rises in a crisis, neither gold nor Bitcoin qualifies.

If it means an asset that has reduced the damage during major equity drawdowns, gold has the stronger record in this ten-year sample. Bitcoin may have other valuable characteristics, but it has not yet demonstrated gold-like downside protection.

What definition would you use for a safe haven, and what result would make you change your view on gold or Bitcoin?

This is educational analysis, not investment advice.

Sources:

- S&P 500 methodology and independent data: https://fred.stlouisfed.org/series/SP500/downloaddataand

- Official GLD description and historical archive: https://www.spdrgoldshares.com/usa/gld/

- Coinbase Bitcoin through FRED: https://fred.stlouisfed.org/series/CBBTCUSD

- Iran chronology: https://apnews.com/article/iran-us-timeline-trump-hormuz-war-ceasefire-04da58cbae991183f8b52ef5bf615963

- Academic safe-haven definition: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=952289

Kraken holds the instant liquidation switch on a crypto firm’s 479 Bitcoin if price drops to $45,094

This post contains content not supported on old Reddit. Click here to view the full post

Coldcard: The Gold Standard in Bitcoin Security

The banner of this subreddit still reads "The Gold Standard in Bitcoin Security". Considering recent events I think this wording should be removed. Curious to know peoples thoughts on this.

The Quiet Attack on Bitcoin

How Economic Attrition Is Doing What a Direct Assault Never Could

There is a particular kind of threat that is difficult to defend against because it never announces itself. It does not arrive as an attack. It arrives as a series of reasonable decisions, each defensible on its own terms, each moving imperceptibly in the same direction. By the time the cumulative effect becomes visible, the damage is structural and the reversal is costly.

This is the nature of the most sophisticated threat Bitcoin has ever faced. And it does not require a single overt rule change.

The Mechanism: Five Stages of Economic Attrition

The process operates in stages, and each stage is designed to look like something other than what it is.

Stage one begins with a policy decision inside Bitcoin Core. The 80-byte OP_RETURN limit, which governed how much arbitrary non-financial data could be embedded in a Bitcoin transaction, is removed. The new ceiling is approximately 100 kilobytes, a 1,250-fold increase. Miners, incentivized by fees generated from Ordinals inscriptions, BRC-20 tokens, and Runes activity, continue mining these transactions. They are profitable. A community response, BIP-110, proposes to cap this data and restore the original constraint. Major mining pools ignore it. The proposal collapses.

Stage two follows as a consequence of stage one. Every arbitrary data inscription mined into the blockchain is permanent. Every full node on the network must download it, store it, and validate it forever. The blockchain grows not just from financial transactions but from data payloads that have no monetary purpose. The cost of running a full node, measured in storage, bandwidth, and processing power, rises continuously and without a natural ceiling.

Stage three is where the real damage occurs, quietly and without drama. Average individuals find node operation increasingly uneconomical. Not because of a single threshold event, but because of gradual cost creep that compounds year over year. They abandon their nodes not because anyone compelled them to but because the economics no longer justify it for a private individual operating without institutional resources. The exit is voluntary. The effect is identical to a forced removal.

Stage four is the consolidation that follows. The node population contracts into entities that can absorb the rising cost: large mining operations, institutional holders, custodial exchanges, data centres. These are precisely the entities subject to government jurisdiction, court orders, regulatory oversight, and the full institutional apparatus of state power. They are incorporated. They have legal addresses. They have compliance departments. They can be compelled.

Stage five is where the theoretical protections of Bitcoin governance become academic. The nodes that matter economically are no longer run by private individuals in homes and apartments across the world. They are run by legal entities that any sufficiently motivated government can reach. The private node runner, who was always the backbone of genuine decentralization, has been priced out of existence. Not by force. By economics.

Why This Is More Dangerous Than a Direct Attack

A direct attempt to change the 21 million cap or cancel the halving schedule would be immediately visible. It would be controversial from the moment it was proposed. It would mobilize the Bitcoin community in organized, passionate opposition. The attack would be loud, and loud attacks generate resistance.

What is described above is a quiet attack. No one announced that the goal was to price individuals out of running nodes. Bitcoin Core developers framed the OP_RETURN limit removal as a philosophical commitment to permissionlessness and censorship resistance. Mining pools framed their opposition to BIP-110 as protecting fee revenue and user choice. Each individual decision is defensible in isolation. The cumulative effect is the slow strangulation of the decentralization that makes Bitcoin resistant to the institutional capture these decisions are actively enabling.

This pattern is not new. It is the same mechanism that operates throughout the broader financial system. The fiat financial system does not announce that its structural goal is the progressive dispossession of the majority. It operates through incentive structures that produce that outcome without requiring anyone to state the intent openly. What is happening inside Bitcoin is structurally identical. You do not need to announce an attack on decentralization. You simply need to set the incentive structure so that decentralization becomes economically unviable, and the system does the rest automatically.

Incentives in Alignment, Without Coordination

None of the actors in this process need to be coordinating explicitly. That is what makes the threat so resilient.

Miners are incentivized by inscription fees to mine large OP_RETURN transactions. The revenue is real and immediate. Bitcoin Core developers are incentivized by a genuine philosophical commitment to permissionlessness, a commitment that in this context produces the same outcome as deliberate sabotage without requiring any malicious intent. Institutional holders are incentivized by their legal and regulatory relationships to prefer a Bitcoin that operates within a framework governments can interface with. A compliant, monitorable Bitcoin is a Bitcoin they can hold without regulatory risk.

No smoke-filled room is required. The incentives align these actors naturally toward an outcome that progressively concentrates node operation into regulated, compellable entities. The system selects for that outcome the same way any incentive structure selects for the behavior it rewards.

What BIP-110 Actually Revealed

BIP-110 was contentious in its specific technical implementation, and reasonable people disagreed about its merits. But regardless of where one stands on its particulars, its failure carries a governance lesson that cannot be dismissed.

The entities that ensured its failure were large mining pools controlling the overwhelming majority of network hashrate. These are precisely the entities whose economic interests align with a Bitcoin that is expensive to independently verify. A blockchain bloated with arbitrary data requires more resources to validate. More resources means higher costs. Higher costs means fewer independent validators. Fewer independent validators means more power concentrated in the hands of those who can afford to validate. Those hands belong to the same institutional actors whose cooperation with government authority is not a hypothetical but a legal requirement.

A Bitcoin that only wealthy institutions can validate is a Bitcoin that wealthy institutions effectively control. Not through any formal rule change. Through the simple arithmetic of who remains in the room when everyone who cannot afford to be there has left.

Decentralization Is Not a Fixed Property

This is the critical point that Bitcoin's cultural narrative has consistently obscured. Decentralization is not a property that was baked into Bitcoin at genesis and cannot be removed. It is a dynamic condition that requires active maintenance through the continuous participation of individual node runners distributed across the world.

The moment that participation becomes economically inaccessible to individuals, decentralization becomes a historical description rather than a present reality. The technical architecture remains intact. The rules remain formally unchanged. But the population enforcing those rules has consolidated into a small number of institutional entities, and the decentralization that gave those rules their meaning has quietly ceased to exist.

Satoshi's design required not just a hard cap and a proof-of-work mechanism. It required those properties to be maintained by a sufficiently distributed network of independent participants that no coalition of powerful actors could capture enough of the governance infrastructure to override them. The cap without the decentralization is a rule written by people who are no longer in the room when the decision to change it gets made.

The path to that outcome does not require a dramatic assault on Bitcoin's consensus rules. It requires only that node operation become slightly more expensive, year after year, through decisions that each appear reasonable in isolation, until the only entities left running nodes are precisely the ones that Bitcoin was built to make irrelevant.

That process is already underway.

EMG PVE PC US Chernarusplus

https://i.redd.it/guuge06c9fih1.png

EMG PVE PC US Chernarusplus

https://i.redd.it/02gc2zenc8ih1.png

Bitcoin : « Les clés privées sont le péché originel de la crypto »

This post contains content not supported on old Reddit. Click here to view the full post