Investing advice, checked against the data.
Plenty of investing wisdom gets repeated and almost none of it gets tested. I run the numbers on the claims people actually act on — and publish the code, the sample period, and the cases where the answer flips.
-
Does Fibonacci Retracement Actually Work?
14.7 million times price reached a level in 499 stocks. It bounced off Fibonacci levels 54.86% of the time, and off the levels right next to them 54.79%.
-
Does Asset Allocation Really Explain 90% of Your Returns?
The 90% figure is real but answers a question nobody is asking — and it isn't a constant: on the same 17 funds it was 85% in one era and 93% in the next.
-
Does CAPE Actually Predict Stock Returns?
CAPE explains 30% of 10-year returns in-sample, but goes sharply negative out-of-sample when tested honestly. Today's CAPE sits off the chart it was fit on.
-
Does Momentum Investing Actually Work?
A real, replicable academic premium with a real way to lose 49% in three months. The buyable ETF version barely, not significantly, beats the market.
-
Does Low-Volatility Investing Actually Work?
In a no-lookahead backtest, high-volatility stocks won on raw return and low-volatility stocks won on drawdown. Sharpe ratios were nearly identical.
-
Does the Small-Cap Premium Still Exist?
The full-sample small-cap premium is not statistically significant. Split at 1990 and nearly all of it happened before that date, not after.
-
Is Value Investing Dead?
The value factor had a real, significantly negative alpha from 2007-2020, then rebounded hard. The real ETFs show a flatter comeback than the factor does.
-
How Much Do Fees Actually Cost Over 30 Years?
Fee calculators show one number from one assumed return. This uses real historical outcomes and separates the fee you paid from what it actually cost you.
-
What a Bad First Decade Does to Retirement
Every 4% rule failure came from the worst fifth of first-decade returns. The 2000 cohort's first decade was even worse than 1966's, and it hasn't failed.
-
Is the 4% Rule Still Safe?
Real history and textbook Monte Carlo agree the 4% rule works below 4.5% and disagree above it. No historical retirement started this expensive.
-
Sequence of Returns Risk, Quantified
Reshuffle the exact same 30 years of returns 10,000 times. A lump sum never notices. A retiree fails or thrives by it. A saver benefits from the reverse.
-
What Dividend Investing Costs You in Tax
Every calculator multiplies yield by tax rate and stops. Here is what the tax actually cost, measured against the performance effects it is competing with.
-
Dividend Growth vs High Yield: Which Actually Wins?
Dividend growth beat high yield by 0.78 points a year. Both lost to just buying the index. And the difference between them turns out to be one factor.
-
Do Dividend Stocks Actually Outperform?
High-yield stocks did beat low-yield stocks. But the entire advantage vanishes once you control for value, and here is the regression that shows it.
-
Does Gold Actually Help a Portfolio?
Yes, but not for the reason it is sold, and the optimal allocation everyone quotes was the right answer in zero of the last six decades.
-
How Long Does the Market Actually Take to Recover?
The average is about five years. The median is two. And if you did not buy at the exact peak, the median is two days.
-
How Many Stocks Do You Actually Need?
The famous answer is 20 to 30, and it comes from a volatility curve. That curve is real. It also answers a different question from the one you are asking.
-
Is the 60/40 Portfolio Dead?
It has been declared dead eleven times since 1926. Here is what happened in the three years after each one, and what actually drives the stock-bond correlation.
-
Your Portfolio Is Less Diversified Than You Think
Correlations did not spike in a crisis. Something worse was true: eleven asset classes were only ever worth about three and a half independent bets.
-
The 10 Best Days Chart Is Missing Half the Story
Missing the 10 best days cost 55 percent. Missing the 10 worst days gained 140 percent. The famous chart only ever shows you one of those numbers.
-
What Perfect Market Timing Is Actually Worth
Buying at every yearly low instead of every yearly high was worth about 27 percentage points. Here is every window since 1928, not the one usually quoted.
-
Does Buying the Dip Actually Work? 98 Years of Data
Waiting for a pullback sounds obviously smart. Across 225 rolling 20-year windows of S&P 500 data, waiting for a 5% or 10% dip won exactly zero times.