Mechanical Investing FAQ

1. What is Mechanical Investing (MI)?

Mechanical investing is a type of investment strategy in which the investor relies on predetermined rules and algorithms to make investment decisions. This approach aims to remove emotion and subjectivity from the investment process, and to increase the consistency and efficiency of the portfolio management. The MI approach can be useful in theory for investors who lack the time or expertise to manage their investments actively, however in practice it tends to involve time and attention for most investors engaged in it. The goal? Usually to have a higher long-term average return than the S&P500, but some aim for a similar return but with reduced losses during market downturns. Mechanical investors may use a variety of tools, such as stock selection by filtering over a large population of stocks each month, or each year, backtesting, and tracking progress of post-discovery data.

2. Once upon a time

Before Shrewd'm, the Mechanical Investing had first became popular at T. M. F. in the late 1990s with their 'Foolish Four', and the community became frenzied whilst the NASDAQ was skyrocketing and momentum screens were the rage. Over over time, MI has given mixed results, but the concept is sound. After all, when you purchase the S&P500 index, you are already inadvertently investing mechanically - in this case by selecting large cap weighted stocks systematically over many years. You could switch strategy to an equal weighted S&P500 index, which through the whole 20th C outperformed the standard S&P500 by 1% or 2% a year after costs.

3. What are the core principles of Mechanical Investing?

• Rule-Based Decisions: Investments are chosen strictly by following mechanical rules or signals.
• Backtesting: Strategies are tested against historical data to evaluate performance and robustness over various periods, as seen in discussions of screen performance metrics like CAGR (Compound Annual Growth Rate) and Sharpe Ratio.
• Systematic Rebalancing: Portfolios are regularly rebalanced, as backtested, so your stocks match the most up to date screening.
• Emphasis on post-discovery results: Real-life results occurring with the MI strategy, after its initial publication, are given much more weight than the earlier backtested results susceptible to data-mining.
• Transparency: Clear rules and metrics for entry, exit, and position sizing.
• Avoiding Emotional Bias: By sticking to rules, investors avoid panic selling or greed-driven buying.

4. What types of mechanical strategies are popular at Shrewd'm?

• Trend Following: Making use of upwards momentum, such as buying stocks, has proven effective over time for market outperformance. (Note: Momentum screens are immune to arbitrage, unlike value screens, because purchasing stocks from momentum-based screens adds to their momentum).
• Value and Quality Screens: Combining valuation metrics (P/E, P/B) with quality factors (ROE, high cash, low debt levels).
• Capital Allocation: Cash/stock ratio can be determined mechanically (e.g., Arezi Ratio).
• Signal-Based Trading: Buy/sell signals generated from technical or fundamental indicators (e.g., Mungofitchs market bottom indicator).

5. How do Shrewd folks pick stocks mechanically?

• Apply a stock screen (ranking systems) using a subscription to either Value Line or SI Pro rankings - we like these two as they match where the MI folk here have conducted much of their backtesting. For example, filter to stocks having no debt, then filter to the 5% with the highest ROE, finally sort by RS26 and select the top 5 in the list.
• Re-balancing every month (or quarterly, etc, consistent with the backtest) applying the stock screen again and executing the necessary trades to hold only the updated stocks in the screen
• Don't have a Value Line or SI Pro subscription? We've got you covered! The rankings for all the popular MI screens, from the classical screens to their more modern marvels, are posted regularly at the Mechancial Investing Board at Shrwed'm by the committed and wonderful generous Shrewd'm community.

6. How often should a mechanical portfolio be rebalanced?

• Common intervals are monthly or quarterly.
• Some members prefer weekly or biweekly adjustments if the system is highly responsive.
• Rebalancing frequency depends on the strategys nature—trend-following may require more frequent updates, whereas value-based systems might rebalance less often.
• The key is consistency and adherence to the models predefined schedule.

7. What are the sneaky traps in MI?

• Data-Mining (Overfitting): Designing rules that work well historically but fail in live markets.
• Signal Noise: False signals leading to whipsaws and unnecessary trades.
• Transaction Costs: Frequent trading can erode returns if costs are not managed.
• Emotional Interference: Temptation to override mechanical rules during market stress.
• Data Quality: Reliance on accurate, timely data, devoid of survivorship bias, is critical for model performance.

8. How do MI investors handle risk management?

• If the volatility disappoints you, MI may not be for you. But if you are applying strategies to reduce volatility, make sure they are backtested to confirm they helped, rather than harmed, long-term returns in the past.
• Keeping the MI screen simpler to avoid data-mining.
• Stop-loss rules or trailing stops embedded mechanically - but make sure you backtest the procedure because it might cause more harm than good.
• Mechanical market timing strategies to enter/exit the market, or change cash/stock percentage allocation based on mechanical valuation - but again, these should be backtested. It is eerily difficult to improve your CAGR with any strategy that reduces holdings to less than a fixed 100% stock allocation through the best and worst market moods.
• Avoiding unnecessary concentration by means of highly correlated screens.
• Keeping costs low (e.g., less frequent trading, being aware of buy/sell spread loss).

9. Can mechanical investing be combined with discretionary investing?

• Some members use mechanical systems as a core portfolio and overlay discretionary trades.
• Others use mechanical signals as alerts but apply human judgment before executing.
• The board generally emphasizes sticking to mechanical rules to avoid emotional mistakes but acknowledges hybrid approaches.

10. What tools and resources do MI board members recommend?

• Jamie Grittons MI Backtester with classic Shrewd'm screens: https://backtest.org
• Custom backtester at https://backtest.org/SB which uses the field guide https://backtest.org/fields.html
• Hugh Todds WER Trade Simulator uses the Weekend Review data: http://tradesim.info
• Spreadsheet models for ranking and signal calculation.
• Programming languages like Python, R, or Matlab for backtesting and automation.
• Data sources such as Yahoo Finance, Quandl, or paid services for reliable data.
• Shared spreadsheets on Shrewd'm for collaborative improvement.

11. How can a beginner Shrewd start with Mechanical Investing?

• Skim through the Mechanical Investing Glossary below, and then to study a term in more detail, post questions about it at the Shrewd'm Mechanical Investing board.
• Experiment with Jamie Grittons MI Backtester above.
• Allocate an isolated small portion of your portfolio (e.g., 5% or 10%) to MI to test strategies without risking capital.
• Keep it simple and focus on the big picture—master core ideas like avoiding data-mining, ensuring backtest data is survivorship-bias-free, and why stable returns matter when tweaking variables.
• Start a trading journal to track your wins and stick to the rules.
• Post to the Mechanical Investing board at Shrewd'm to soak up the wisdom.

12. What mindset is recommended for successful Mechanical Investing?

• Patience and discipline to follow rules strictly.
• Willingness to accept drawdowns as part of the process.
• Openness to learning and adapting models based on evidence.
• Avoiding emotional reactions to short-term market noise.
• Long-term commitment to the mechanical approach given that great MI strategies that beat the market over 10+ years can, and generally will, have periods of multiple years with underperformance.

13. Mechanical Investing Glossary

• Anchoring: Fixating on irrelevant past data when making decisions. For example anchoring to a stock's 52-week high, expecting it to return to that level despite new market conditions. Or anchoring to a PE of 15 as an historical average, despite the stock consistently trading above a 20 PE (or below a 12 PE), and waiting indefinitely for it to return to the 15 PE.
• Arezi Ratio: The attractiveness of stocks as measured ratio of the yield on 3-month Treasury bills to the trailing 12-month earnings yield of the S&P 500 index
• Backtesting: Simulating a strategy on past data to estimate performance.
• Benchmark: Standard (e.g., S&P 500) to compare strategy performance.
• Beta: Measure of a stocks volatility relative to the market.
• CAGR: The annual growth rate of an investment. 90% return over 3 years? That's 23.9% a year, because 1.239^3 = 1.9. To the same CAGR from the 3 year return, use (1.90 ^ (1/3)) = 23.9%.
• Capital Allocation (using MI): Using fixed rules to decide how to allocate between stocks and cash (e.g., Arezi ratio).
• Confirmation Bias: Seeking data that supports existing beliefs about a strategy.
• Correlation: Measure of how two assets move together, used in diversification.
• Data-Mining: (Same as Overfitting) Finding misleading patterns in historical data that fail in real trading.
• Debt-to-Equity: Ratio of total debt to shareholders equity, indicating leverage.
• Disposition Effect: Selling winners too soon and holding losers too long.
• Dividend Yield: Annual dividend payment divided by stock price.
• Drawdown: Decline from a portfolios peak value to its trough.
• EBITDA: Earnings before interest, taxes, depreciation, and amortization, a profitability measure.
• Efficient Market Hypothesis (EMH): Stock prices reflect all public information, so (other than having insider information) it is not possible to get market-beating returns.
• EPS Growth: Percentage increase in earnings per share over time.
• FOMO: Fear of missing out, driving impulsive investment choices.
• Free Cash Flow: Cash generated after expenses and capital expenditures.
• Gross Margin: Revenue minus cost of goods sold, divided by revenue, showing profitability.
• GSD: Volatility of returns, lower values indicate more consistent performance.
• GTR1: Backtesting tool for simulating investment screens.
• Herding: Following the crowd in investment decisions, often irrationally.
• Holdings: Number of stocks held in a mechanical investing screen at one time.
• Kurtosis: Measure of extreme returns (fat tails) in a distribution.
• Liquidity: Ease of buying/selling a stock without impacting price.
• Look-Ahead Bias: Using future data in backtests, inflating performance.
• Loss Aversion: Tendency to fear losses more than value equivalent gains.
• Market Cap: Total value of a companys outstanding shares.
• Mean Reversion: (or Reversion to the Mean) Expecting certain variables, such as net margins or the P/S ratio, to return to average levels over time.
• Momentum: Strategy betting on stocks with recent strong performance continuing.
• MxDD (Max Draw-down): Largest peak-to-trough decline in portfolio value.
• Overconfidence: Overestimating ones ability to predict market outcomes.
• Overfitting: (Same as Data-Mining) Creating a strategy too tailored to past data, reducing future effectiveness.
• Out-of-Sample Testing: (Same as Post-Discovery Data) Evaluating a strategy on data not used in its development.
• Outperformance: Excess return of a strategy over a benchmark.
• P/B Ratio: Stock price divided by book value per share.
• P/E Ratio: Stock price divided by earnings per share.
• Parameter Drift: Changes in screen criteria over time, affecting consistency.
• Peg26: Screen using PEG ratios and momentum to select stocks.
• PEG Ratio: Price-to-earnings ratio divided by earnings growth rate.
• Portfolio Turnover: Frequency of trading, impacting costs and taxes.
• Post-Discovery Data: (Same as Out-of-Sample Testing) Evaluating a strategy on data not used in its development.
• Price-to-Cash-Flow: Stock price divided by cash flow per share, assessing cash efficiency.
• Price-to-Sales: Stock price divided by revenue per share, a valuation metric.
• Rebalancing: Adjusting portfolio to maintain original allocation or screen criteria.
• Regret Aversion: Avoiding decisions to prevent potential regret.
• Relative Strength: Stocks price performance rank. Noted RS13, RS26, etc. For example, stocks with the highest (or lowest) RS13 rose in price (or fell in price) the most over the last 13 weeks.
• Reversion to the Mean: (or Mean Reversion) Expecting certain variables, such as net margins or the P/S ratio, to return to average levels over time.
• Round-Trip: The cost of both entering and leaving a position (2 x bid-ask spread and 2 x commission).
• ROA: Return on assets, net income divided by total assets.
• ROE: Profitability ratio of net income to shareholders equity.
• RS13: Relative Strength over 13 weeks, measuring short-term outperformance.
• RS52: Relative Strength over 52 weeks, measuring long-term outperformance.
• Screen: Predefined criteria to filter stocks for investment, usually with a sort order as the final step and the top 5 or top 10 stocks selected.
• Screen of Screens: Combinng the listings of multiple screens, adding their ranks and picking the stocks that score highest overall.
• Sharpe Ratio: Risk-adjusted return, dividing excess return by volatility.
• Skewness: Asymmetry in return distribution, impacting risk assessment. Positive skewness means occasional big gains with limited losses. Negative skewness means risk of rare but large losses.
• Trading Costs: (See Round-Trip) Including bid-ask spreads and commissions.
• Sortino Ratio: Risk-adjusted return, focusing on downside volatility only.
• Survivorship-Bias: Overestimating performance by only including surviving stocks in data.
• Total Return: The screen's overall performance, including price changes and dividends, expressed as a percentage of the initial investment.
• Value Investing: Buying stocks undervalued based on fundamentals like P/E or P/B. True value investing, however, also takes into account growth. An effective mental model, as developed by Manlobbi, is to estimate the company's per-share private purchase value 10 years away, adding in the per-share dividends along the way, which he notes as IV10, and comparing that to the quote today. Purchase firms with the highest IV10/price ratio.
• Volatility: Degree of price fluctuations in a stock or portfolio.