What the first 100 days of a live portfolio actually feel like
Lessons from the first ~100 days of Outpick's live portfolio: early dispersion, why paper returns lie, and how to judge a young book without rewriting the strategy every week.
A backtest is a closed book. A live portfolio is an open one — with marks that move while you are still building the cast of characters. The first hundred days teach different lessons than year five.
Highlights
Outpick's live book started in April 2026. The early months are about seeding positions on a biweekly cadence, surviving noisy P&L, and resisting the urge to rewrite rules after every green or red week. Judge a young portfolio on process adherence and thesis quality first — headline return second.
Live is not a replay of the backtest
Our walk-forward work (see walk-forward backtesting explained) answered a specific question: did the rules work on data the fitting process never saw? That is necessary. It is not the same as sitting with real marks, real liquidity, and a book that is still incomplete.
In the first hundred days you do not yet have a full roster. Returns are path-dependent on which names cleared the bar first. One early winner can dominate the percentage tape; one early laggard can make the whole experiment feel broken. Neither outcome proves or kills a multi-year strategy by itself.
Dispersion shows up immediately
Concentrated books are honest. When you own a handful of positions, day-to-day P&L is not a smooth index line — it is a handful of businesses arguing with the market. That is uncomfortable if you grew up on VOO charts. It is also the point. Alpha, if it exists, comes from intentional difference, not from matching the crowd's calendar.
Early on we have already seen the usual shape: a few names working hard, a few flat, and at least one that makes you re-read the thesis with a colder eye. The discipline is not pretending every position is “fine.” The discipline is separating thesis broken from price noisy.
Process beats narrative in year zero
The temptation in month three is to declare victory or declare failure. Both are usually premature. What you can evaluate early:
- Cadence — are we still adding on the biweekly rhythm we promised, not impulsively?
- Documentation — does every live name have a clear Insight note with risks, not just a bull case?
- Rule fidelity — are we respecting max adds, sizing, and exit logic instead of improvising?
- Honesty — are percentage returns shown without hiding the ugly rows?
Those are leading indicators. Trailing return becomes more informative as the book seasons and as more evaluation cycles complete. Until then, process is the scoreboard that does not lie as easily as a short sample of marks.
How we want you to read the dashboard
Use the dashboard as a living lab notebook, not a mutual-fund fact sheet. Members can open Insights for why each name is there. Compare the live book to the index over time, but give the strategy enough cycles to look like itself. And if you are building your own book alongside ours, size for your sleep — our risk budget is not yours.
OUTPICK MEMBERSHIP
Want to see the picks?
Outpick is a stock research team that publishes high-conviction picks every two weeks — full thesis, live tracking, and transparent performance. Founders: $250/year through December 1, 2026 · then $1,000/year.
START YOUR MEMBERSHIP →Frequently asked questions
Frequently asked questions
When is a live track record 'real'?+
Why not wait until the portfolio is fully built to show it?+
Should I copy positions one-for-one?+
Discussion
—The Market Note
One short read, every Monday. Free.
What the model is seeing across ~3,600 US-listed stocks, which sectors are scoring, and what we make of it. Market commentary — the picks stay behind the membership.
Keep reading
Related articles
Why we publish one stock pick every two weeks
Why Outpick publishes one stock pick every two weeks: conviction over noise, a cadence you can actually follow, and how biweekly research compounds without becoming a second job.
Jul 9, 2026
Japan's Treasury dump and AI's power problem: two clocks on the US economy
America sold euros to buy yen while Japan sold Treasuries to defend its currency — and AI spending is racing a grid that takes eight years to plug in. Here's what the headlines miss.
Aug 23, 2026
Inflation is not a five-month Middle East story
Everyone blames Middle East oil for inflation. The Fed’s scoreboard — PCE — has missed 2% for years. Here’s why debt, money supply, and fiscal dominance matter more than the napkin narrative.
Jul 25, 2026