Weekly winners portfolio
I remember reading an opinion piece in one of the major US newspapers about the current bull maket and how it might last more than we imagine. The mood in wall street seems to very good.
I thought that this was then a good time to test a little simple theory.
Lets buy the best performing stocks - each week. Specifically:
buy the ten stocks that have risen the most over roughly the last trading month, then refresh the portfolio once a week.
How it works:
-
After the market has completed a session, the system looks at every stock available in Wall Street Game.
-
It measures each stock’s split-adjusted return over the previous 20 trading sessions—roughly four weeks.
-
It removes stocks that:
- Have fewer than 21 completed daily prices.
- Closed below $5.
- Averaged less than $20 million of daily trading value.
- Do not have a usable live quote when the trade is prepared.
-
It ranks the remaining stocks from strongest to weakest and selects ten different companies.
-
Approximately 98% of the portfolio is divided equally between those ten companies—about 9.8% each. The remaining 2% stays in cash.
-
At the next weekly rebalance:
- Stocks no longer in the selected ten are sold.
- New winners are bought.
- Stocks that remain are increased or reduced to restore equal weighting.
The investment idea is momentum: stocks that have recently performed strongly may continue performing strongly for a while. But it is intentionally a very pure experiment:
- No sector diversification rule.
- No volatility or risk adjustment.
- No preference for existing holdings.
- No judgement about whether a company is good or expensive.
- No protection against several selected stocks belonging to the same market theme.
I have tested this - for a while and you can check the results by yourself in the wall street game - IT Works !! Yes there are all the caveats I discussed above - this is a disaster waiting to happen - but so far I have made lots of (fake) money with it
Originally published on Substack on 29 August 2026.