Over 244 days on which every reading exists, Bitcoin's daily move travels with the S&P 500 at +0.42, and every one of the 12 crypto assets measured here moves the same way. The chain from policy to crypto holds at every link but one, and the one that breaks is in the middle: the real cost of money does not move what risky borrowers pay, at -0.11. Rate decisions reach the dollar and stop. What reaches crypto is the appetite for risk in equities.
Swipe the diagram sideways to follow it
Every reading above is how closely two things moved across 243 daily changes from 244 level dates on which all of them reported. A reading near zero means they did not travel together. None of it says one thing caused another.
Crypto with equities+0.42Still +0.30 once the price of risk is taken out. This is the channel.
Rates to the price of risk-0.11The one broken link, and it sits in the middle of the chain.
Crypto with the real cost of money-0.11Falls to +0.02 once equities is taken out. Not a channel of its own.
Assets moving with equities12 of 12They travel with each other at +0.76, so nothing like twelve independent readings.
The proposed chain, written down first, then measured.
Each pair below belongs to the usual policy-to-market story. Only after that sequence was written was each reading measured. That order matters: a relationship found first and explained afterwards is a story fitted to a number.
What the market expects the Fed to do sets the real cost of moneythey travel together
If the overnight rate is going to be higher for longer, every longer loan reprices to match. This is the most mechanical link in the chain.
reading +0.74242 matched dayscould plausibly be +0.67 to +0.79
The real cost of money moves the dollarthey travel together
Money goes where it is paid more to sit. When money costs more here than elsewhere the dollar tends to firm. The two moved together on the same days, which is not the same as one moving the other.
reading +0.25242 matched dayscould plausibly be +0.13 to +0.37
The real cost of money moves what risky borrowers payno connection found
A weaker company that has to refinance pays the safe rate plus a premium for the chance it does not repay, so the safe rate should be the floor under that premium.
reading -0.11242 matched dayscould plausibly be -0.23 to +0.02
What risky borrowers pay moves equitiesthey travel together
The same conditions that make lending to a weak company expensive make owning one unattractive. Credit usually turns first, though this reading cannot say which moved first.
reading -0.63242 matched dayscould plausibly be -0.70 to -0.55
Equities move cryptothey travel together
Crypto is held as the high-risk end of the same portfolios, sized out of the same risk budgets, so it should follow. Measured with the two clocks lined up, it does.
reading +0.42242 matched dayscould plausibly be +0.32 to +0.52
Charlie's analysisCrypto has no policy channel of its own. Every outside reading that looks like one turns out to be equities showing through. Bitcoin against what risky borrowers pay is -0.32, and -0.08 once equities are removed. Against the dollar, -0.20 falls to -0.04. Against the real cost of money, -0.11 falls to +0.02. Run the test the other way and equities survives it: +0.42 against Bitcoin, still +0.30 once what risky borrowers pay is removed. One reading carries all of it, and it is the ordinary one.
The mistake this page nearly published.
Read the obvious way, this same data says crypto ignores the stock market entirely. It says that because of a clock, not because of a market. The fix is worth showing rather than quietly applying.
Charlie's analysisThis nearly went out wrong, and the reason is worth publishing rather than quietly fixing. Crypto trades every hour and its daily reading is taken at midnight UTC. The American readings are stamped at the New York close, some hours later. So a crypto reading dated the fifth sits between the American closes of the fourth and the fifth, and matching the two by calendar date compares two different days. Done that way, the link between equities and crypto reads +0.05, which is nothing, and the comfortable conclusion is that crypto ignores the stock market. Lining the clocks up gives +0.42. The gain appears at exactly one day of offset and at no other, which is the signature of two clocks rather than of one thing predicting another. Anyone comparing crypto with American market data by date alone will keep finding a market that answers to nothing.
How many assets carry it.
A reading that shows up in one asset may belong to that asset. Every asset below was measured against the same 243 daily changes, but the assets are not independent votes.
Twelve crypto assets measured against what risky borrowers pay, equities, the real cost of money, the dollar and Bitcoin, over the same window.
Asset
With what risky borrowers pay
Same, after equity control
With equities
With the real cost of money
With the dollar
With Bitcoin
Move over the year
Bitcoinbelow-zero range with bond premiummoves with equities
-0.32
-0.08
+0.42
-0.11
-0.20
+1.00
-27.8%
Ethereumbelow-zero range with bond premiummoves with equities
-0.27
-0.03
+0.39
-0.13
-0.22
+0.90
-41.7%
Solanabelow-zero range with bond premiummoves with equities
-0.22
+0.01
+0.37
-0.11
-0.13
+0.86
-47.1%
BNBbelow-zero range with bond premiummoves with equities
-0.15
-0.01
+0.24
-0.08
-0.18
+0.76
-19.1%
XRPbelow-zero range with bond premiummoves with equities
-0.25
-0.04
+0.35
-0.04
-0.19
+0.86
-49.5%
Cardanobelow-zero range with bond premiummoves with equities
-0.21
-0.02
+0.30
-0.04
-0.17
+0.78
-74.4%
Avalanchebelow-zero range with bond premiummoves with equities
-0.26
-0.06
+0.34
-0.09
-0.16
+0.76
-69.5%
Chainlinkbelow-zero range with bond premiummoves with equities
-0.26
-0.09
+0.31
-0.07
-0.15
+0.79
-46.9%
Dogecoinbelow-zero range with bond premiummoves with equities
-0.23
-0.05
+0.29
+0.00
-0.15
+0.79
-61.0%
Polkadotbelow-zero range with bond premiummoves with equities
-0.19
-0.04
+0.26
-0.03
-0.12
+0.60
-77.9%
Litecoinbelow-zero range with bond premiummoves with equities
-0.19
-0.01
+0.30
-0.04
-0.15
+0.72
-56.6%
TRONbelow-zero range with bond premiummoves with equities
-0.27
-0.20
+0.18
+0.09
-0.09
+0.45
+2.0%
Charlie's analysisThe reading is not one asset's. All 12 move with equities, from +0.18 to +0.42, over the same 244 days. Remove Bitcoin and Ethereum entirely and an equal-weight basket of the other ten still reads +0.35, and its apparent link to what risky borrowers pay collapses the same way, from -0.26 to -0.05 once equities are taken out. One caution against reading too much into the count: these 12 assets travel with each other at around +0.76, so twelve agreeing readings are nothing like twelve independent ones. They are closer to one observation seen twelve times.
What argues against all of this.
Charlie's analysisTwo things argue against taking this too far. The strongest readings on the page are still crypto against crypto, not crypto against anything outside it: Bitcoin and Ethereum travel together at +0.90 against +0.42 for the best outside reading, so most of what moves these assets is not on this page. And the link is not steady. Over rolling sixty-day stretches it runs from +0.19 in 2026-08-21 to +0.60 in 2026-03-11, and sits at +0.24 now. It never turns negative in this window, but a reader measuring one quarter would get an answer three times larger or smaller depending on the quarter.
Questions this page can answer.
Every answer below is drawn from the readings on this page and nothing else. The last one is refused.
Not directly, and not through the channel people name. Over 244 days, Bitcoin's daily move travels with the real cost of money at -0.11 and with where the market expects the Fed to go at -0.04. What it does travel with is equities, at +0.42. So policy reaches crypto only as far as policy reaches the stock market, and over this window the link between rates and the appetite for risk was itself broken.
With equities+0.42
With the price of risk-0.32
With the dollar-0.20
With the real cost of money-0.11
With expected policy-0.04
How closely Bitcoin's daily move travels with each reading, with the clocks lined up. Faded bars could plausibly be zero. Nothing here is a claim that one causes the other.
Which links of the chain hold?
4 of 5. What the market expects the Fed to do moves the real cost of money at +0.74. That moves the dollar at +0.25. What risky borrowers pay moves equities at -0.63, and equities move crypto at +0.42. Each of those has a mechanism a person can say out loud, and each was written down before it was measured.
What the market expects the Fed to do sets the+0.74
The real cost of money moves the dollar+0.25
The real cost of money moves what risky borrow-0.11
What risky borrowers pay moves equities-0.63
Equities move crypto+0.42
So where does it break?
In the middle, and only there. The real cost of money does not move what risky borrowers pay over this window, at -0.11 across 242 days, and that reading could plausibly be anywhere from -0.23 to +0.02. That is the joint between the two halves. Above it, rates run to the dollar and stop. Below it, the price of risk runs through equities all the way into crypto. Both halves work. They are simply not attached to each other, which is exactly the part everyone assumes.
No chart. A break is the absence of a reading, and drawing it as a bar would give it a size it does not have.
Does crypto respond to anything equities has not already brought it?
No. Every outside reading that looks like its own channel disappears when equities is removed. What risky borrowers pay goes from -0.32 to -0.08. The dollar goes from -0.20 to -0.04. The real cost of money goes from -0.11 to +0.02. Run the test the other way and equities survives it: +0.42 before, +0.30 after the price of risk is removed. One reading carries all of it.
Price of risk, alone-0.32
Price of risk, without equities-0.08
Dollar, alone-0.20
Dollar, without equities-0.04
Equities, alone+0.42
Equities, without the price of risk+0.30
Each reading measured on its own, then again with the other one taken out.
Why does everyone else find that crypto ignores the stock market?
Because of a clock, and this page nearly published the same mistake. Crypto trades every hour and its daily reading is taken at midnight UTC. American market readings are stamped at the New York close, some hours later. Matching them by calendar date therefore compares two different days: read that way, equities and Bitcoin travel together at +0.05. Line the clocks up and it is +0.42. The whole gain appears at one day of offset and at no other, which is what a clock difference looks like. Something predicting something else would fade away gradually instead.
crypto moved 2 days earlier-0.04
crypto moved 1 day earlier-0.01
same calendar date+0.05
crypto moved 1 day later+0.42
crypto moved 2 days later-0.10
crypto moved 3 days later-0.06
The same comparison at every offset. One clean peak, not a slope.
Is that one coin's finding, or the whole market's?
All 12 assets measured here move with equities, from +0.18 to +0.42, over the same 244 days. That said, these assets travel with each other at around +0.76, so twelve agreeing readings are nowhere near twelve independent ones. Counting them as twelve separate confirmations would overstate the case considerably.
Bitcoin+0.42
Ethereum+0.39
Solana+0.37
BNB+0.24
XRP+0.35
Cardano+0.30
Avalanche+0.34
Chainlink+0.31
Dogecoin+0.29
Polkadot+0.26
Litecoin+0.30
TRON+0.18
Each asset against equities, clocks lined up.
What happens if you take Bitcoin and Ethereum out?
The reading survives. An equal-weight basket of the other ten reads +0.35 against equities. Its apparent link to what risky borrowers pay collapses the same way Bitcoin's does, from -0.26 to -0.05 once equities are removed. So the two largest assets are not carrying any of this, and the smaller ones behave the same way.
Equities+0.35
Price of risk-0.26
Price of risk, without equities-0.05
The dollar-0.18
Real cost of money-0.06
Ten assets, equal weighted, with Bitcoin and Ethereum removed entirely.
Rates went up and Bitcoin fell. Is that not the transmission?
That is one observation, not a relationship. Over this window the real cost of money rose +0.72 points and Bitcoin fell -27.8%, so the story fits. Measured day by day, 244 separate times, the two travel together at -0.11, and nothing survives once equities is removed. Over the same window the S&P rose +18.7% while Bitcoin fell, which fits the opposite story equally well. A year that moves in one direction gives a single data point dressed up as a trend.
No chart. Two lines that both slope are the illusion being described, and drawing them here would sell it rather than test it.
What actually happened after the rate cuts?
There were 3 moves of a quarter point or more in the window, found in the rate itself rather than taken from a calendar. Bitcoin fell in the five sessions after every one of them, and so did the S&P after most. That is worth stating and worth not over-reading: 3 events is far too few to conclude anything, and five sessions after a scheduled decision contains everything else that happened in those five sessions.
2025-09-184.33% to 4.08% -0.25 points
Bitcoin, next five matched market days -2.7%S&P 500 -0.4%Real cost of money +0.07 pointsPrice of risk +0.05 points
2025-10-304.12% to 3.87% -0.25 points
Bitcoin, next five matched market days -5.6%S&P 500 -1.5%Real cost of money +0.01 pointsPrice of risk +0.28 points
2025-12-113.89% to 3.64% -0.25 points
Bitcoin, next five matched market days -6.3%S&P 500 -1.8%Real cost of money -0.01 pointsPrice of risk +0.07 points
Does the money-printing story hold up?
Not over this window. Money available to the financial system, after the Treasury's cash pile and money parked back at the Fed are taken out, went from $5.90 trillion to $5.77 trillion, a change of -2.3%. Compared week against week with Bitcoin the reading is +0.08 over 51 weeks, which could plausibly be zero. It is measured weekly because the underlying figures are published weekly, and kept apart from the daily readings for that reason.
No chart. A reading this close to zero drawn as a bar would look like a finding.
Is the link steady, or does it come and go?
It is always there and it is never the same size. Read over rolling sixty-day stretches, Bitcoin against equities runs from +0.19 in 2026-08-21 to +0.60 in 2026-03-11, and sits at +0.24 now. It does not cross zero in this window, so the direction is stable even though the strength is not. Anyone measuring a single quarter would get an answer roughly three times larger or smaller depending on which quarter they picked.
2025-12-05zero means no connection2026-08-28
Sixty-day readings of Bitcoin against the S&P 500, stepped forward one day at a time.
Which reading argues against the rest of this page?
TRON. It is the one asset whose link to what risky borrowers pay does not vanish when equities is removed, holding at -0.20 where every other asset here falls to nearly nothing, and it has the weakest equity link on the board at +0.18. It also rose +2.0% over the window. One asset out of 12 is not enough to overturn the reading, and it is enough to stop anyone calling it a law.
TRON-0.20
Chainlink-0.09
Bitcoin-0.08
Avalanche-0.06
Dogecoin-0.05
XRP-0.04
Each asset's link to the price of risk after equities has been removed.
What would change this answer?
The middle link is the one to watch. If the real cost of money starts moving what risky borrowers pay again, the two halves join up and rate decisions would reach crypto the long way round, through credit and equities. A period of genuine stress would push every reading here up at once and would say more about forced selling than about policy. And a change in who holds crypto would show up first as the equity link strengthening and staying strong across several rolling windows rather than one.
No chart. These are conditions to watch, not measurements.
Which asset should I buy, and where does this go next?
Neither question is answered here, and nothing on this page could answer them. Every reading above describes how things moved together over a window that has already finished. None of it forecasts, none of it establishes that one thing causes another, and a relationship that ranges from weak to moderate depending on the quarter would be a poor foundation for a decision even if it were stable. Anyone turning these readings into a trade is adding an assumption that is theirs, not this page's.
Refused. The data cannot support the question.
Ask this page anything.
Type a question in your own words. Answers come from the same readings printed above, measured over the same days. Where the readings cannot answer a question, it will say so rather than guess.
Nothing here is advice, a forecast, or a claim that one thing causes another.
Every reading, with the day it was observed.
Levels now, a year ago, and the move between. These are the raw readings the rest of the page is built from.
Every policy and market reading used on this page, with its current level, its level a year ago, the change, and the date it was observed.
Reading
Now
A year ago
Change
Observed
The rate the Fed setsThe overnight rate the Federal Reserve actually sets
3.63%
4.33%
-0.70%
2026-09-03
Where the market expects the Fed to goWhere the bond market thinks that rate goes over two years
4.34%
3.49%
+0.85%
2026-09-03
The real cost of moneyThe ten-year yield after the inflation the market expects
2.42%
1.70%
+0.72%
2026-09-03
The ten-year Treasury yieldThe plain ten-year Treasury yield, before inflation is taken out
4.77%
4.05%
+0.72%
2026-09-03
Inflation priced over five yearsThe inflation rate the bond market is pricing over five years
2.37%
2.41%
-0.04%
2026-09-04
Ten-year minus two-yearThe ten-year yield minus the two-year
0.41%
0.56%
-0.15%
2026-09-04
The dollar against a broad basketThe dollar against a broad basket of currencies
118.75
120.10
-1.35
2026-08-28
What risky borrowers payWhat riskier companies pay to borrow above Treasuries
2.65%
2.84%
-0.19%
2026-09-03
The S&P 500The S&P 500
7,719
6,495
+1,223
2026-09-04
Expected swings in the S&PExpected swings in the S&P over the next month
14.3
15.1
-0.8
2026-09-03
What would make this reading wrong.
Charlie's analysisThis reading would be wrong if a year is too short, which it may well be: one year holds one regime, and the middle link that breaks here has held firmly in other periods. It would be wrong if weekends are doing the work, though reading only the 138 runs that are genuinely one day apart gives +0.40 rather than +0.42, so they are not. It would be wrong if daily changes are the wrong clock altogether and the transmission runs over weeks. And none of it establishes cause in either direction. Two things moving together is a description of what happened, not an explanation of why.