Author: Chen Xiaomeng
If we define the problem more precisely:
Buying at any point and holding for a full 4 years results in a positive nominal total return.
The conclusion is very interesting:
Among high-risk, freely tradable mainstream assets, none can achieve "100% historical rolling 4-year profitability"—except this one.
But if low-risk fixed-income products are included, instruments such as 4-year government bonds, rolling over short-term bonds, and term deposits can also achieve this—and the “100%” in the latter case is entirely different.
I. The S&P 500 cannot do it
Many people automatically assume:
If you hold U.S. stocks for four years, won't you eventually make money?
Actually, it's completely not.
The long-term data maintained by NYU professor Damodaran includes dividends in the S&P 500 total return, covering the period from 1928 to 2025.
Here are a few very typical 4-year windows.
1929–1932
Annual returns are approximately:
1929: -8.3%
1930: -25.1%
1931: -43.8%
1932: -8.6%
Cumulative over 4 years:
Approximately -64.8%.
In other words:
$1 million in, only about $350,000 left after four years.
1999 to 2002
Even including the final surge of the 1999 internet bubble:
1999: +20.9%
2000: -9.0%
2001: -11.9%
2002: -22.0%
Total return over 4 years is still approximately:
-24.4%.
2007–2010
Experiencing a financial crisis:
2007: +5.5%
2008: -36.6%
2009: +25.9%
2010: +14.8%
Even after the sharp rebounds in 2009 and 2010, over four years later:
Approximately -3.2%.
So:
The long-term win rate of the S&P 500 is very high, but it has never reached 100% over any four-year period.
Two: The Nasdaq 100 is even less likely
According to official Nasdaq data, during the bursting of the dot-com bubble, the Nasdaq 100:
2000: -36.4%
2001: -30.8%
2002: -38.9%
2003: +48.5%
Even with a nearly 50% surge in year four, the total accumulation over this entire four-year cycle remains:
Approximately -60%.
Nasdaq also noted that after the 2000 bubble peak, the index fell by approximately 83% in total.
So the historical performance difference between BTC and tech stocks on this point is very clear:
Tech stocks may encounter a downturn so severe that it cannot be recovered over a full four-year cycle.
Bitcoin has not done so yet.
Three: Even gold can't do it.
Gold is often regarded as the classic long-term store of value, but it can still lose significant value over a four-year period.
For example, from 1981 to 1984, according to Damodaran's data:
1981: -32.6%
1982: +15.6%
1983: -16.8%
1984: -19.4%
Cumulative over 4 years:
Approximately -47.7%.
Even more strikingly, the World Gold Association's own research shows:
From November 1987 to August 1999, gold experienced a nearly 12-year bear market, with a cumulative price decline of approximately 48%.
So although gold has very strong long-term monetary properties,:
Four years certainly does not equal a safety period.
Four, real estate is not either
When looking at the overall U.S. real estate market rather than a single lucky property purchase, a similar four-year negative cycle exists.
In Damodaran's data:
Approximate annual real estate returns from 2007 to 2010:
2007: -5.4%
2008: -12.0%
2009: -3.85%
2010: -4.12%
Total approximately:
-23.3%.
Moreover, real estate has another optical illusion:
Houses don't trade daily like BTC or stocks, so their price curves appear smoother.
In reality, if there were a public market offering prices for your home every day, real estate volatility would be much greater than we typically perceive.
Five: Even long-term government bonds can't do it
This is actually the most likely to cause misunderstanding.
Many people say:
Aren't U.S. Treasuries risk-free assets?
The key is to distinguish:
Hold a government bond to maturity and trade a long-duration bond.
For example, the total return on 10-year U.S. Treasury bonds:
2021: -4.42%
2022: -17.83%
2023: +3.88%
2024: -1.64%
Approximately over 4 years:
-19.8%.
The reason is the aggressive interest rate hikes that began in 2022, causing older bonds with low coupon rates to plummet in market price.
Therefore:
TLT, the 10-year Treasury bond index, and long-term bond funds are not "four-year guaranteed win assets."
Six, even corporate bonds cannot do it.
Baa corporate bonds during the same 2021–2024 period:
+1.02%
-15.23%
+8.74%
+1.74%
Still approximately after 4 years:
-5.3%.
Therefore, credit bonds cannot do it either.
Seven: What else can truly achieve "100% nominal profit over four years"?
Here, we enter another category of assets.
Category 1: Rolling Over U.S. Short-Term Treasury Bills
A 3-month T-Bill is a very typical example.
It doesn't make money through asset price increases, but rather:
Buy at a discount → Redeem at face value upon maturity.
For example, if you buy a $100 face value T-Bill for $99.50, the U.S. Treasury will pay you $100 at maturity.
TreasuryDirect describes the mechanism for T-Bills in the same way: the purchase price is typically below face value, and upon maturity, you receive the full face value—the difference is the interest.
Damodaran's annual U.S. T-bill data since 1928 has consistently shown positive nominal returns, so rolling over 3-month T-bills over any complete 4-year period in history has naturally resulted in positive returns.
But note:
This is called:
The dollar amount does not lose value.
Does not represent:
Good value for money.
For example, with 8% inflation and a 2% T-Bill yield, your account balance increases, but your real purchasing power declines.
Eight, Four-Year Term Deposit / CD
Suppose you now find:
4-year 4% fixed-rate CD
And you:
No early redemption;
The bank is not covered beyond the insurance limit;
The interest rate contract is valid;
Then, 4 years later:
The nominal USD profit is essentially determined on the day of purchase.
The U.S. FDIC insures eligible bank deposits and CDs, with the current standard coverage amount being:
$250,000 per depositor, per bank, per ownership category.
So the "four-year win rate of 100%" for this type of thing essentially comes from:
Contract + Credit Protection.
Rather than an increase in asset prices.
Nine: U.S. Treasury bonds with maturities of four years or less
This one too.
For example, if you buy today:
A U.S. Treasury bond maturing in four years with a yield to maturity of 4%
Then hold until maturity.
As long as the U.S. government makes its payments on time, you don’t need to worry about how bond prices fluctuate in the meantime.
Expiration:
Principal and interest will be paid as agreed.
Therefore, the nominal yield is essentially locked in at the time of purchase.
But if you're buying:
10-year government bond, but must be sold in year 4
That's not the case at all.
So there is a particularly important distinction here:
A 4-year bond does not equal holding a long-term bond for 4 years.
The former can lock in profits.
The latter carries duration risk.
Ten: The One True God—Bitcoin
The strategy report uses daily BTC/USD data from July 2010 to August 2026 to statistically analyze rolling holding periods starting each day.
The result is:
Holding Period Final Profit Ratio Worst Total Return 1 year: 73.1% to -83.6% 2 years: 84.0% to -68.3% 3 years: 99.3% to -34.7% 4 years: 100.0% to +32.6%
In other words, none of the 4,419 rolling four-year periods in its data ended in a loss.
And the four worst years in history were:
April 16, 2021 → April 16, 2025
Total earnings are still:
+32.6%
Approximate annualized rate:
7.3% CAGR.
Two words—unbeatable~
Eleven: So what truly deserves comparison is this table.
So an interesting dumbbell will appear:
Those achieving a 4-year nominal win rate of 100% are concentrated at two extremes.
One end is:
Very low-risk, contract-based payment assets.
The other end is:
Bitcoin is an asset with high growth potential, inherently shorting fiat currencies, and still in the early stages of monetization.
And those traditional risk assets in between:
Stocks, gold, real estate, and long-term bonds all fail to do so.
Twelve, the "100%" of BTC and the "100%" of government bonds have entirely different values.
This is the most critical part.
100% T-Bill
The logic is:
Today, you’ll know exactly how much someone owes you in the future.
You primarily bear the sovereign credit risk of the United States.
100% of BTC
The logic is completely different.
Bitcoin:
No issuer;
No one promises you a specific amount of money four years from now;
No principal repayment;
No coupon;
No cash flow.
So the 4-year 100% for BTC comes entirely from:
Historical market prices eventually rose.
Precisely because of this, it is particularly unusual.
Thirteen: Final Investment Perspective
BTC may be one of the most prominent modern major risk assets with a record of positive nominal returns over any four-year holding period in history.
But what’s truly impressive isn’t the three digits “100%.”
It has experienced repeated declines of 70% to 90%, yet still has not produced a single complete four-year cycle with negative returns.
This is much stronger than simply saying "Bitcoin will rise in the long term."
This also illustrates the opposite issue:
Holding on is far more important than constantly trading.
In particular, selling calls, leverage, swing trading, and yield enhancement all essentially involve altering this very rare long-term return distribution in some way.