Blog
Graham Stephan's housing market predictions, checked
By Callsight · Published Oct 7, 2026 · Updated Oct 12, 2026
Figures as of Nov 4, 2026. They change as new windows settle.
As of Nov 4, 2026, Callsight has judged 104 of Graham Stephan’s forecasts Correct or Incorrect at one year: 63 Correct and 41 Incorrect. On the 84 with a comparable base rate (how often the called move happened anyway), 57.1% were Correct, against a base rate of 47.2%. Housing-crash calls: 16 said none would come, 1 said one would.
What is Graham Stephan’s track record on predictions?
Callsight checks every call Graham Stephan made in his own words at the same 6 windows, from 30 days to 5 years, and the table shows all of them, because a call can be Incorrect at three months and Correct at a year. Beside each hit rate is its base rate: how often the move he called happened anyway over a window that long, measured on the same calls.
| Window | Correct | Incorrect | Pending | Hit rate | Base rate | Above the base rate |
|---|---|---|---|---|---|---|
| 30 days | 49 | 48 | 1 | 50.5% (97) | 33.0% on 60 of 97 | +10.3 pts |
| 90 days | 55 | 51 | 4 | 51.9% (106) | 40.1% on 71 of 106 | +4.9 pts |
| 180 days | 60 | 42 | 13 | 58.8% (102) | 44.3% on 80 of 102 | +12.0 pts |
| 1 year | 63 | 41 | 20 | 60.6% (104) | 47.2% on 84 of 104 | +10.0 pts |
| 3 years | 64 | 25 | 42 | 71.9% (89) | 51.9% on 73 of 89 | +17.9 pts |
| 5 years | 40 | 15 | 91 | 72.7% (55) | 54.3% on 39 of 55 | +17.5 pts |
The table counts a call he repeated on the same subject, in the same direction, within 90 days once, so its counts are smaller than the call-by-call counts below. His record on every subject is on Graham Stephan’s page, and his list of calls has each call’s words and the verdict at every window.
Did Graham Stephan predict a housing market crash?
Mostly he predicted the opposite: of his 17 counted calls on a housing crash, 16 said none would come. At one year, the calls that said none would come stand at 11 Correct, 0 Incorrect, 1 Partly correct and 4 still open. The one call that said a crash would come is Incorrect at one year. Another 5 calls on the subject are listed but not counted: 2 forecasts he read out from someone else, shown with their sources, and 3 remarks of his own that the record leaves out, each with the reason on its page.
A housing crash call is judged on the Case-Shiller national home price index (CSUSHPINSA). A fall of at least 10% from its highest point since the call, inside the window, is a crash, and half that fall is Partly correct. A call that no crash will come is judged the other way round. Two of the calls that said none would come, in his words:
“we're not going to be seeing a 10 to 20% real estate collapse anytime soon outside of a few localized markets.”Reading when said
331.51
Jun 9, 2025
Reading at close
336.66
Jun 9, 2026
Change
+5.15
The house price index (CSUSHPINSA, first releases) stood at 331.5 in June 2025, the figure at the start of the window. Between then and June 2026, its deepest fall from a running peak was −1.5%, in January 2026. A fall of at least 10.0% counts as a housing crash, and at least 5.0% as partial. The call was that no housing crash would come.
30 days
Correct
90 days
Correct
180 days
Correct
1 year
Correct
“even though that sounds severe, it's probably not going to lead to an all-out crash.”Reading when said
300.41
Sep 14, 2022
Reading at close
312.31
Sep 14, 2023
Change
+11.9
The house price index (CSUSHPINSA, first releases) stood at 300.4 in September 2022, the figure at the start of the window. Between then and September 2023, its deepest fall from a running peak was −2.6%, in January 2023. A fall of at least 10.0% counts as a housing crash, and at least 5.0% as partial. The call was that no housing crash would come.
30 days
Correct
90 days
Correct
180 days
Correct
1 year
Correct
Neither call stated a deadline; the headline for each is the one-year window.
Every call on the subject, from every channel, is on the housing crash page.
What did he say about house prices, and how is that judged?
Apart from the crash calls, 54 of his calls on house prices are counted. Another 99 on the subject are forecasts he read out from someone else or remarks of his own that the record leaves out.
A house price call is judged on the index’s year-on-year growth rate: the rate for the month the call was made against the rate for the last month inside the window, both as first published. A change of less than 2.0 points either way is Within band. A call that prices will keep rising can therefore be Incorrect while prices are still rising, if the growth rate falls by 2.0 points or more. At one year his counted calls come to 7 Correct, 13 Incorrect, 4 Within band, 5 still open and 25 that could not be checked. The rule in full is on the methodology page, and every call is on the house prices page.
What were Graham Stephan’s stock market calls?
On stocks and funds at one year, 14 of his calls were Correct and 7 Incorrect, a hit rate of 66.7% on 21. Measured on the same 21 calls, the base rate is 68.0%, a difference of −1.3 pts. With a base rate that high, a call that the market will rise comes true most of the time whoever makes it, which is why the hit rate is set against it.
“we'll continually hit new all-time highs as the market and economy grows hopefully”Recorded as the market, a word for a group: this call is judged on SPY, the stand-in the method assigns to it.
Entry price (first close after the video)
$586.10
Dec 19, 2024
Price at close
$676.47
Dec 18, 2025
Total return, dividends included
+16.8%
SPY closed at $586.10 on Dec 19, 2024. On Dec 18, 2025 it closed at $676.47, a total return of +16.8% (dividends included). The call was that it would rise.
30 days
Correct
90 days
Incorrect
180 days
Correct
1 year
Correct
No deadline was stated; the headline is the one-year window.
What has Graham Stephan said about Bitcoin?
Callsight counts 11 of his calls on Bitcoin and 4 on other coins. In the record, which reads the one-year window and counts a repeated call once, his crypto calls stand at 3 Correct and 7 Incorrect, too few to say much on their own. Every call on Bitcoin, from every channel, is on the Bitcoin page. One of his Bitcoin calls:
“longterm I am bullish for Bitcoin I do think the ETFs make it incredibly easy to buy or sell without worrying about self- custody and I have a feeling over time more people are going to begin diversifying into it causing the price to go higher”Entry price (first close after the video)
$101,329
Feb 3, 2025
Price at close
$75,770
Feb 3, 2026
Return
−25.2%
S&P 500 +16.7%
BTC closed at $101,329 on Feb 3, 2025. On Feb 3, 2026 it closed at $75,770, a return of −25.2%. SPY returned +16.7% over the same window. The call was that it would rise.
30 days
Incorrect
90 days
Incorrect
180 days
Correct
1 year
Incorrect
He spoke of the long term and named no date, so the headline is the one-year window, where the call is Incorrect. That verdict covers the first year only. The call’s windows at 3 years and 5 years are still open, and its page will show them as they settle.
What did he say about interest rates and the Fed?
21 of his counted calls are about the Fed’s rate and 26 about interest rates in general. A Fed call is Correct when the target rate moves at least 25 basis points the called way inside the window. At one year the Fed calls are at 14 Correct, 1 Incorrect, 2 still open and 4 that could not be checked, and the calls on rates in general at 7 Correct, 6 Incorrect and 13 that could not be checked. Every call on the Fed’s rate, from every channel, is on the Fed rate page.
“That's why I wouldn't be surprised if we see another small rate cut when they meet in October and maybe another one when they meet in December.”Reading when said
4.5%
Sep 17, 2025
Reading at close
4%
Sep 17, 2026
Change
−0.50 pp
The policy rate was 4.50% at the start of the window and 4.00% on Sep 17, 2026, a net change of −0.50 pp. The largest cut inside the window was 75 basis points. The stated threshold was 25 basis points.
30 days
Correct
90 days
Correct
180 days
Correct
1 year
Correct
Does this show what Graham Stephan owns?
No. Callsight records what he predicts on air. A position he says he has already taken is listed and never scored, and the 169 forecasts he read out from banks or other people are shown with their source and never counted for or against him. The same goes for Bitcoin: the record does not show whether he holds any, or how much.
How does his record compare with other channels?
The record of every tracked channel sets each one-year hit rate against its own base rate, and the comparison page puts Andrei Jikh’s and Graham Stephan’s calls side by side. Two other articles check Andrei Jikh’s Bitcoin calls and Joseph Carlson’s stock calls the same way. His housing-crash and recession calls also appear, next to the other channels’, in who predicted the recession, and a comparison of all four channels asks whether any of them beat the market.
How was each forecast checked?
Each call is quoted word for word from the video and linked to the second it was said. A price call is measured from the first closing price after the video went public, and an economic call on the series it is about, as first published. The rules are the same for every channel, and the methodology page gives them in full. If you think a call was misread, the corrections page says how to report it.
How these examples were chosen
The calls shown here follow a fixed rule: settled at one year, from widely watched videos, on well-known instruments, recent where possible, and with at least one Correct and one Incorrect. A call whose recorded reading is being checked again is left out. Every call is listed on the channel's page.
Figures are computed by code from the same record as the rest of the site. This article was drafted with the help of AI tools and reviewed by a person before publication. It is not investment advice.