This sign-in link has expired or is invalid. Request a new link.
Follow how we invest our own capital and test ideas about companies and markets. We distinguish live decisions, model observations, and tests on historical data.
What we bought, sold, and decided each trading day.
Free members
The account recorded one buy and one sell on Nov. 9. The article distinguishes those trades from the earlier decisions to keep holding nine stocks.
See how our own capital performed, including the period, benchmark, and drawdowns.
This is the record of an account investing our own capital. See what it earned, how it compared with the S&P 500, and how far it fell along the way.
Performance as of 2026-11-09. Measurement period: 2026-04-14 to 2026-11-09. Primary currency: US dollars. Page updated: 2026-11-10 12:30 UTC.
The as-of date is the last trading day for which the account and benchmark can be compared. The record usually updates the following afternoon in Japan.
Live account · USDAs of 2026-11-09
+34.1%
Cumulative time-weighted return since 2026-04-14
2026/4/142026/7/282026/11/9
AlpAIS&P 500
Account returns include received dividends and trading fees. The benchmark uses SPY with dividends reinvested. Returns are cumulative, not annualized.
Time-weighted return (TWR) adjusts for deposits and withdrawals, so adding cash is not counted as investment profit. These are cumulative returns, not annualized returns.
The difference over this period is +25.2 percentage points. This is a simple difference between returns. It is not a risk-adjusted measure or proof that AI, one stock, or one trading rule caused the result.
Our strategies and trading rules have changed during the period as research progressed. The record includes those changes; it is not the track record of a single unchanged strategy.
The same account returned +33.4% in Japanese yen. Currency movements can make the yen and US-dollar results differ.
Current drawdown is the decline from the account's previous performance peak at the as-of date. Maximum drawdown is the deepest such decline in the measurement period. Neither is a limit on future losses.
| Month and coverage | AlpAI | S&P 500 ETF | Difference (percentage points) |
|---|---|---|---|
| 2026-04 · partial month 2026-04-14 to 2026-04-30 | +4.1% | +3.5% | +0.6 |
| 2026-05 2026-05-01 to 2026-05-31 | +14.5% | +5.3% | +9.3 |
| 2026-06 2026-06-01 to 2026-06-30 | -1.4% | -1.0% | -0.4 |
| 2026-07 2026-07-01 to 2026-07-31 | +2.8% | +0.0% | +2.7 |
| 2026-08 2026-08-01 to 2026-08-31 | +0.2% | +2.7% | -2.5 |
| 2026-09 2026-09-01 to 2026-09-30 | -3.5% | -0.3% | -3.2 |
| 2026-10 2026-10-01 to 2026-10-31 | +7.3% | -0.9% | +8.2 |
| 2026-11 · partial month 2026-11-01 to 2026-11-09 | +7.1% | -0.4% | +7.5 |
The first month starts on the measurement start date; the current month ends on the as-of date. Partial months are not comparable to full months. Differences use unrounded values and can differ by 0.1 percentage point from subtracting the displayed numbers.
| Measure | Method and scope |
|---|---|
| Account return | TWR calculated from the broker's (IBKR) daily account valuations and deposit/withdrawal records. |
| Dividends, fees and taxes | Account valuations reflect dividends received and trading fees. Corporate income tax and similar taxes are not deducted. |
| Benchmark | SPY, an ETF tracking the S&P 500, using dividend-adjusted closing prices. This assumes dividend reinvestment and reflects ETF expenses; it is not the price-only index. |
| Timing | Account and benchmark closing dates are aligned. This is not a live intraday comparison. |
| Scope | The whole account investing our own capital, including changes in rules during the period. It does not isolate a particular stock, rule or AI model. |
| Private information | We keep account balances, cash-flow amounts, lists of holdings, and share counts private. |
Through October 8, 2026, articles compared the account with the S&P 500 price index, excluding dividends. This page now uses the dividend-reinvested ETF comparison described above.
Daily and Weekly articles retain the performance snapshot and comparison basis from their first publication. We do not replace those figures with today's results. If we find an error, we correct it.
This performance page and theme research are public. Daily records document account decisions and confirmed trades; the Weekly review puts the evidence in market context. Free membership provides access to recent member articles.
Past performance does not guarantee future results. Read our research and publication policy.
See the performance page for the latest figures.
We keep account balances, cash-flow amounts, lists of holdings, and share counts private.
Charts, monthly returns and methodologyDoes an investment idea survive realistic costs, data limitations, and account constraints? Read what we compared and what we found.
Latest study
Deposits can lift a brokerage balance without producing investment gains. Learn how time-weighted return separates the two, and why the benchmark, dates and drawdowns still matter.
Read the studyThe week's market moves, the evidence behind them, and what to watch next.
All Weekly reviewsFollow changes in our account from one trading day to the next.
All Daily records