I was exploring this new strategy on finding underperforming stocks that I wanted to share. The analysis is based on the “reversion to the mean” belief. This means that a stock that has deviated too far above or below its mean price will eventually go back to its historical mean in the short term. Obviously this theory has a lot of caveats but the general idea is sound to me. This is something we observe every day, when there is good news the stock price pops and eventually falls back down, when there is bad news the stock price tanks heavily and then creeps up. So the strategy and the technique I will highlight here is to identify the stocks which are undervalued based on this theory.

Here are some key principles to keep in mind –

1. Historical Price Analysis: Begin by analyzing a stock’s historical price data, preferably over a multi-year period. This will help in determining its historical mean or average price. Tools like moving averages, especially the 14 day, 50-day and 200-day moving averages, are commonly used.
2. Standard Deviation Analysis: Standard deviation helps measure the volatility or the dispersion of a set of data from its mean. If a stock’s current price is more than two standard deviations below its mean, it could be considered undervalued.
3. Volume Analysis: Often, the price moves on abnormally high or low volume can be temporary. Look for spikes in trading volume which accompany price dips. If a stock drops significantly on low volume, it might be a sign of an overreaction.
4. Fundamental Analysis: Even if a stock appears undervalued based on the reversion to the mean, it’s crucial to understand the underlying reasons. Check the company’s earnings, cash flow, debt, and other key financial metrics to ensure it remains fundamentally strong.
5. News and Events Analysis: Analyze the reason behind any sharp drop in price. If it’s due to temporary factors or overreactions to news that won’t significantly impact the company’s long-term earnings potential, it could be a buying opportunity.
6. Comparative Analysis: Compare the stock’s valuation metrics, like P/E, P/B, and P/CF ratios, with its industry peers. If it’s trading at a significant discount without any clear fundamental reason, it could be undervalued.
7. Monitor and Re-evaluate: It’s not enough to just identify and invest in these stocks. Continuously monitor the company’s performance, news, and financials. The reversion to the mean might not happen if there’s a fundamental change in the business.
8. Diversify: Like all investment strategies, diversification is key. Not all stocks you identify will revert to their mean, so spreading your bets can minimize risks.

Caveats & Considerations:

• Time Frame: While stocks might revert to the mean, the timeframe in which they do can vary widely. Some might take months while others might take years. Ensure your investment horizon aligns with your expectations.
• Changing Fundamentals: If a company’s fundamentals deteriorate, the stock might not revert to its previous mean. Always stay updated with the company’s performance.
• Market Factors: Broader market or economic factors can also influence stock prices. An overall bearish market can keep even undervalued stocks depressed.
• Risk Management: Always set a stop-loss or a point at which you’ll re-evaluate your investment. If a stock keeps declining, understand the reasons before adding more to your position.
• Behavioral Biases: Investors might be susceptible to confirmation bias, where they seek out information that confirms their belief. Always approach with a neutral perspective and be open to evidence that might counter the reversion hypothesis.

In conclusion, while the “reversion to the mean” strategy can be a valuable tool in identifying potential investment opportunities, it’s essential to combine it with fundamental analysis and a broader perspective on market conditions. Always remember that investing carries risks and it’s important to do your due diligence before making any investment decisions.

Remember this is a short term play strategy – this means this is not for finding tickers to own in the short term – the moment the price cross the upper threshold – what I call upper bound its time to cash out. Also this strategy is more likely to yield fruit during a bear run where there are many more stocks that have been falling.

Methodology

Here’s what we will do for a ticker –

1. Calculate the 14 day and 60 day mean price of the stock – m14 and m60
2. Calculate historical standard deviation of the stock price over a 14 day and 60 day period – sd14 and sd60
3. Calculate the current price of the stock – price and the percentage drop from 14 day and 60 day mean – drop14 and drop60
4. Calculate the average percentage drop from 14day and 60 day mean for other stocks in its sector – sectordrop14 and sectordrop60
5. Pick the stocks where
   • drop14 < sd14 and drop60 < sd60
   • sort them in descending order by (drop14 – sectordrop14)
6. Cash out once the stock price reaches mean – sd14 or the mean if you are feeling optimistic. There is no right answer here.

Why is the sector part important? its because we want to remove any variability caused by a general market drop. For eg a stock that dropped by 5% may seem like a good buy but if the S&P500 dropped by 10% in the same time period, it is likely the stock is still overvalued and can tumble even further.

Why would this strategy work?

The basic tenet is that the stock market gives a higher than expected reward and a higher than expected punishment to stocks on good/bad news respectively.

Why would this strategy not work?

Sometimes a stock genuinely underperforms its sector because of fundamental reasons. In this case its not a great idea to buy the stock even if it is underpriced. Thats why I recommend validating the stocks that this algorithm returns (see my earlier videos on some ways to do validation).

Code

def get_stock_historical_price(ticker):
  data = yf.Ticker(ticker)
  hist = data.history(period="2mo")
  prices = hist['Close'].tolist()
  current_price = data.info['currentPrice']
  industry = data.info['industry']
  return prices, current_price, industry

# calculate mean and standard deviation
def calc_stats(prices, current_price, days):
   last_prices = prices[-days:]
   mean_price = statistics.mean(last_prices)
   stdev_price = statistics.stdev(last_prices)
   stdev = (stdev_price / mean_price) * 100
   price_change = ((current_price - mean_price) / mean_price) * 100
   return price_change, mean_price, stdev

def calculate_score_for_ticker(ticker, ticker_stats, mean_industry14, mean_industry50):
   # if greater than moving average remove this stock
   if ticker_stats[ticker][1] > ticker_stats[ticker][3] or ticker_stats[ticker][2] > ticker_stats[ticker][4]:
      return 0.0
   industry = ticker_stats[ticker][0]
   return (mean_industry14[industry] - ticker_stats[ticker][1]) + (mean_industry50[industry] -   ticker_stats[ticker][2])

Happy investing.

Disclaimer: The information provided here is for general informational purposes only and should not be considered as professional financial or investment advice. Before making any financial decisions, including investments, it is essential to seek advice from a qualified financial advisor or professional.