Jim Cramer’s Recommendations: A Six-Year Analysis

Luís Fernando Torres
8 min readMar 30, 2023

--

Introduction

In recent years, we’ve seen a growing interest in the stock market and financial industry, with an increasing number of people investing in stocks, crypto, etc. This has also increased the number of social media influencers on platforms like YouTube, TikTok, and Instagram, offering their perspectives on the market and providing various stock buying and selling recommendations.

In the United States, one of the most well-known financial commentators is Jim Cramer, host of CNBC’s “Mad Money” and co-founder of TheStreet.com, one of the world’s most significant financial news and education websites. He is known for giving numerous stock buying and selling recommendations in the American market to his followers, recommending, on average, about 20 different stock trades per episode of his CNBC show, which is significantly higher than the average of 10 to 15 recommendations made by American research firms per quarter.

Jim Cramer gained significant notoriety in the financial market for his unique and straightforward approach to market analysis. This has both gained him fans and critics. You can even find on the internet some strategies and portfolios that do the opposite of everything he recommends on his show.

In this context, I decided to conduct a Data Science project to observe the recommendations made by Jim Cramer between 2016 and 2022, analyze his best and worst recommendations, and evaluate the performance of a portfolio based on his recommendations. With this project, I aim to answer the question: is it worth listening to these “influencers”?

Methodology

The database used for this project is the “Jim Cramer’s Picks 2016–2022”, available on Kaggle (https://www.kaggle.com/datasets/diamondprox/jimcramer), which contains data obtained from TheStreet and Quiverquant sites. It is composed of attributes such as the company name, the ticker (code used in the stock exchange to identify each financial asset), the recommendation date, the type of recommendation (buy or sell), and the variations in the stock prices one day, one week, one month, and one year after the recommendation, as well as the variations of the same periods for the American market benchmark, the S&P 500.

Jim Cramer’s Picks From 2016 to 2020

The first step for project execution was data treatment, seeking to make the necessary changes for subsequent analysis. First, I changed some data types, changing columns that were in text format to numerical values. Then, I checked for any missing data and found that several data in the returns were empty.

I then noticed that most companies in the dataset were listed as “Not Available”. To solve this problem, I used the BeautifulSoup library to extract the name of the companies in Yahoo Finance through the tickers that were already in the database.

Creating a function with BeautifulSoupt to obtain companies’ names

I could realize that Jim Cramer not only made buy and sell recommendations but also recommended his followers to hold certain stocks and also gave positive and negative mentions to others. Additionally, I noted the occurrence of a “3” recommendation. I found that it was a probable error, in only one of the instances, and I chose to remove this record from the database, as I could not try to guess what Cramer’s recommendation for these particular stocks would be.

Finally, considering that the goal of this study was to analyze the performance of stocks after the recommendation, I chose to remove all records with null data from the dataset, leaving only those that could provide relevant information for this study.

Dataframe after data pre-processing

Results

To analyze Cramer’s results, I used data visualization techniques with Plotly to generate graphical representations of his recommendations and stock performances after they’ve been mentioned on his show.

In the first analysis, I performed a count of all the recommendations made between the years 2016 and 2022. I found that 61.27% of his calls were ‘Buy’. One of the main arguments from Cramer’s critics is that he tends to be way too bullish in his calls, and now we can conclude that these critics are grounded in reality, at least for the last six years, as Cramer has made far more buy calls than sell calls. Only 10.32% of his calls were for selling, while negative and positive mentions represent 14.39% and 13.87% of data, respectively.

Cramer’s calls during 2016 and 2022

I’ve organized some of Cramer’s best ‘buy’ calls, plotting charts that show stock performances one month after his recommendations. A buy call for Palantir Technologies stocks, for example, resulted in a monthly return of 154.9% for those who bought it. ‘Buy’ calls for shares of Plug Power and Moderna also generated good results, with both having returns above 100% in the following month after being recommended on Cramer’s show.

Best returns for ‘Buy’ calls

Of course, Cramer couldn’t be always right! Those who bought EPR Properties stocks following Cramer’s recommendations saw the company’s shares drop 74.30% in the following month. In addition, those who bought shares in Toll Brothers, Amarin, Novavax, and Ventas also experienced negative returns of way over 50%.

Worst returns for ‘Buy’ calls

When we analyze ‘sell’ calls, it is noteworthy that Cramer was assertive on two occasions in recommending people to sell Norwegian Cruise Line shares, which went down 79.4% and 78.8% twice in the following month after Cramer’s recommendations. In the analyzed period, the highest monthly return for a ‘sell’ call was for Callon Petroleum stocks, which went down 80.80% in the following month.

Best returns for ‘Sell’ calls

The biggest losses, however, are found when we look at stocks that did the opposite of what Cramer predicted they would do when he suggested people sell them. If you had sold Sunrun stocks, based on what you saw on Cramer’s show, you would have seen the stock rise 115.10% just one month after selling. A similar thing would happen if you sold United Natural Foods shares and would’ve seen the shares rise to 114.30% in the opposite direction to your trade.

Worst returns for ‘Sell’ calls

Analyzing only Jim Cramer’s best and worst results, however, provides only a partial view of how a portfolio could perform based on his recommendations alone. To have a more complete view, it is important to analyze the average returns of different periods available to know how assertive he was in his recommendations.

Average returns by call

In the table above, it can be seen that those who followed Cramer’s buy and sell recommendations, on average, had a positive return on the day following the recommendation since their ‘buy’ stocks went up, on average, 0.04%, while their ‘sell’ stocks went down, on average, 0.01%. However, when looking at the one-year performance after the recommendations, it’s possible to see that the stocks that Cramer recommended to sell went up by 14.25% on average, which resulted in losses for those who went short in the hopes that Cramer was correct. In addition, it is possible to notice that the shares of companies that received negative mentions in Cramer’s program overperformed those that he recommended buying.

When we look at the boxplots to evaluate the performance of Cramer’s recommendations, we can draw other interesting conclusions. For example, the biggest outlier in the boxplot of returns one-day after the recommendations shows an asset that traded extremely higher the day after Cramer recommended people to sell it. Looking through the dataset, we can see that those who sold PG&E shares on January 23rd, 2019, following Cramer’s recommendation, had to deal with the fact that the company’s shares went up 76.6% in the following trading session!

Boxplots for one-day performance after the recommendation

PG&E stocks return one day after Cramer’s selling recommendation

Another interesting outlier appears when we analyze the boxplot for the one-year returns after recommendations. Those who sold Nio shares in July 2020, following Cramer, saw the company’s shares rise 8% the next day, in addition to a surprising 1,501% appreciation for the year!

Returns for Nio stocks

Boxplots for one year returns after the recommendation

The boxplot above, displaying returns one year after the calls, provides very important information. Most of the outliers, regardless of whether the recommendation or mention indicated an increase or decrease in share prices, point to a general rise in share prices in the US stock market, which is a portrait of the analyzed period. Between 2016 and January 2022, the US stock market was in a period of general appreciation, which made it easy for anyone to make money just by buying shares of good companies, which justifies why, in general, the stocks mentioned by Cramer went up on average, including those he recommended selling or made negative mentions.

Another important point to mention is that, when analyzing the returns boxplots, it’s clear that the median of Cramer’s recommendations is close to zero in all of the different periods, which indicates that these recommendations, in general, do not bring significant returns for those following them.

Conclusion

This study concludes that several stocks that Jim Cramer recommended buying went up, and several others that he recommended selling also went up. The period from 2016 to 2022 was a period of a general rise in the prices of shares listed on the American stock market, which generated positive returns for those who bought shares of good companies.

Analyzing the boxplots provided information that Jim Cramer doesn’t seem to be any more assertive than other people on average, and some of his selling recommendations ended in poor results for people who have followed them in the long term, as well as some buying recommendations that also generated losses in the short term.

In conclusion, the results suggest that Jim Cramer’s opinions regarding any given stock should not be considered the most important decision-making factor for buying or selling any stock on the stock market.

Thanks for reading!

Luís Fernando Torres

LinkedIn

Kaggle

Like my content? Feel free to Buy Me a Coffee ☕ !

--

--