Welcome to The Alpha Engineer FAQ. This post answers common questions about implementing the quantitative method, managing a portfolio, and understanding the strategy. I’ll update this regularly as new questions arise. You’ll find a permanent link to the latest version on the home page.
Understanding the Method
Q: Do you follow the trades of the model portfolio?
No. I follow the method, not the specific trades of the model portfolio. The model portfolio tracks the general performance of the quantitative approach using 50 stocks. I personally hold only 20 stocks, which provides good results with better manageability (see my fine-tuning analysis for details). My personal results align completely with what the strategy should produce.
Q: What’s the difference between the method and your deep-dive research?
The quantitative Alpha Engineer method is the core of my stock investment strategy and the foundation of this blog. The deep-dives are complementary research for paid subscribers—the icing on the cake.
Not everyone follows the method exactly. Some subscribers use the ranking as an additional tool in their stock picking process, combining it with other screeners and checklists. I add value through selective research on specific top-ranked stocks, and I have skin in the game by investing in these separately in my deep-dive portfolio.
In the long run, I’ll discover whether this portfolio adds value beyond the quantitative method. Many successful investors with quantitative strategies have found that beating their ranking model through discretionary selection is difficult. I try to do better than my model by posting only high-conviction ideas. As a result, I can’t post ideas regularly—there may be 5 or 10 in a given year, depending on what I discover.
Stock Selection
Q: Which stocks do you buy from the ranking?
The ranking is alphabetical within each percentile rank, and my actual selection process is random. When I need to buy a stock, I choose randomly from available liquid stocks at percentile rank 99. If none are available that I don’t already hold, I move to rank 98.
About 5% of the time, I can’t buy a chosen stock (my broker doesn’t support that exchange). In those cases, I simply move to the next stock.
Q: Why are stocks with the same rank listed alphabetically?
The ranking has a very broad top. This is why selecting just the top 50 or even top 100 stocks delivers similar returns as selecting only 20 stocks. Going for the absolute highest-ranked stocks within the percentile rank 99 group doesn’t make sense. If that were the process, the method’s performance would decline if too many paid subscribers chased the same stocks on Monday morning. The alphabetical ordering within percentiles prevents this crowding issue.
Q: What if too many people follow the strategy?
Successful blogs posting microcap stocks with limited liquidity will influence prices if many followers make the same trades, limiting returns for those buying or selling late. The Alpha Engineer method reduces this risk because:
Not everyone buys or sells the same stocks at the same time
The ranking has a broad top with many equivalent options
The random selection process distributes demand
Stocks that are liquid one week might be illiquid another week, so the buy list is dynamic due to rank and liquidity changes
To ensure future alpha generation in case of huge success, I intend to set a cap on the total number of paid subscribers.
Q: What about concentration risk?
I don’t deviate from the “random selection” discipline to ensure even distribution across sectors or countries. I let the ranking decide which stocks to buy. This means I may hold more stocks from Europe than the US, or more from certain sectors than a typical market index distribution.
This makes sense. Some countries and sectors score higher at certain times, so my holdings should reflect that. On average, a 20-stock portfolio with rank 90 as the sell criterion sells stocks every 6 months—about 40 trades per year. Chosen randomly, this produces a representative sample of the major sectors and countries at the top of the ranking.
Q: What about trading costs for certain markets?
This depends on your broker and position size. For me, Canadian and UK stocks (due to the 0.5% stamp duty) cost more to trade than most US and EU markets, but not enough to exclude them from selection.
Backtests show that an average trade returns 16% in 6 months including 2% transaction costs (1% buying, 1% selling). If transaction costs are 2% instead of 1%, that return drops to 14%—still worthwhile.
Some investors exclude specific markets if costs are significantly higher. For investors with larger restrictions, focusing on specific markets (US only, UK only, EU only, etc.) still works but with somewhat lower return and risk-adjusted return expectations (see my market impact analysis).
Portfolio Construction
Q: How much do you invest in each position?
I start with equal positions: 20 stocks at 5% of total allocation each. When a stock drops below my ranking threshold (or loses its rank), I sell it and buy a new stock with the proceeds. If I need to sell two stocks, I buy two stocks splitting the proceeds equally. I follow the same process for multiple sales.
The stocks generate dividends—currently about 1.5% yield for the model portfolio. Other portfolios created from the ranking will have similar yields. I keep dividends in the portfolio and add them to the cash balance for the next purchases.
When buying new stocks, I apply minimum and maximum position sizes:
Minimum position: 2.5% of portfolio value (0.5 × average position)
Maximum position: 7.5% of portfolio value (1.5 × average position)
Average position: Total portfolio value ÷ target number of stocks
Sometimes a stock sale doesn’t provide enough cash for a full new position, or provides too much. In both cases, I leave cash in the portfolio for the next purchases. The model portfolio uses the same method. An alternative approach is adding the cash to top-ranked positions if you prefer not holding cash for a week or two.
Q: Should I buy all portfolio stocks at once when starting?
I did that initially, though some investors prefer dollar-cost averaging to build the portfolio over time if they believe market risks may provide better prices in the future. For example, you could buy a few stocks each month and slowly build a 20-stock portfolio over several quarters. In the meantime, simply trade the stocks you own following the method.
Trading Mechanics
Q: How do you buy stocks?
I set a limit order prior to market open at the latest transaction price. I might set it slightly above that price if I have a large order relative to the previous day’s volume, but typically I only adjust the next day if the order didn’t succeed or only partially filled.
I try to buy at that slightly higher price during the week. I never raise my price more than 2% to catch a stock that’s running away (which does happen). I check the ranking the following week to determine whether to place the higher-priced order or abandon that stock and choose another.
This means I sometimes have partial stock positions. I generally consider them full positions unless they’re too small (less than a half position). In those cases, I may sell them immediately—typically they’re already significantly higher in price, so they’re a win.
Note that the model portfolio also includes 1% transaction costs and slippage and buys/sells at the next day’s average transaction price to correspond to real trading.
Q: How do you sell stocks?
I follow a similar process. I set a limit order prior to market open at the latest transaction price. The next day, I check if the order succeeded. If not, I sell at a limit price up to 2% lower. If I still can’t sell, I check the following week to see if it makes sense to sell at the new lower price.
In general, stocks that drop below my ranking threshold don’t return immediately (as momentum fades), so I normally sell those stocks the next week—sometimes at a significantly lower price. This happens, but it also happens in the model portfolio, which sells at the average price of the next day.
Special Considerations
Q: What about the 3-month holding period?
The model portfolio has a 3-month holding period. Stocks are held for 90 days before being considered for selling based on their ranking. This reduces turnover slightly without impacting return (check the fine-tuning post for details). Since transaction costs are already included in the return calculations, the actual benefit is zero. My main purpose was to demonstrate that the method relies on longer-term price appreciation, not short-term trading.
For my personal portfolio, I don’t use a 3-month holding period. I simply sell based on ranking. The portfolio manager tool also doesn’t account for a holding period. If you use the portfolio manager with a holding period, you would need to disregard sell signals until the stock is available for selling.
Q: Will my subscription price change?
Subscriptions at Substack renew at your original rate as long as you don’t cancel and resubscribe. As long as you maintain your monthly or yearly subscription, your price remains the same.
I do plan to update pricing yearly for new subscribers to account for inflation and the growing value of the service. The earliest subscribers will benefit the most, as it should be. Pricing for 2025 was $39/month or $390/year. In 2026, subscriber pricing increased to $45/month or $450/year.
Learn more about my investing framework:
Read about my overall portfolio strategy.
See the full details of my systematic method for stock picking.
The Alpha Engineer --- Investing with a quantitative edge
Disclaimer: The Alpha Engineer shares insights from sources I believe are reliable, but I can’t guarantee their accuracy—data’s only as good as its inputs! This content (whether on Substack, via email newsletters, X, or elsewhere) is for informational and educational purposes only—it’s not personalized investment advice. I’m not a registered investment advisor, just an engineer crunching numbers for alpha. My opinions are my own and may shift without notice. Investing involves risks, including the chance of losing money. Past performance, whether from back-testing or historical data, does not guarantee future results—outcomes can vary. So, please consult your financial advisor to see if any strategy fits your situation. Full disclosure: I may own positions in the securities I mention, as I actively manage my own portfolio based on these strategies.
So I have been tracking your work for a few months now and am ready to move forward. But basically, I plan on taking a sampling of 20 companies with 99 ranking and buying. Then if/when they move down, replacing them with another one of the 99. Is this basically the strategy?