Does Insider Trading Predict Stock Market Returns? 2025 Update

June 02, 2025 | InsiderSentiment.com Team


This article updates an earlier article we published on our blog in 2023, titled Does Insider Trading Predict Future Stock Market Returns? In that piece, we conducted a hypothetical exercise where investors would look at realizations of our Insider Sentiment Tracker as they came in, and using the strategy we lay out in the article, determine whether to buy the Russell 2000 ETF (IWM) or to put their money in a non-interest bearing checking account. At that time, we found that utilizing the Insider Sentiment Tracker to determine the timing of going in and out of IWM improved the 10-year total return from 67% (using IWM alone with no going in and out), to 130%, along with a tripling of the risk-adjusted reward-to-risk (Sharpe) ratio.


We are now revisiting this methodology to see how it fares in 2025 from the COVID period onward, and if there are any improvements we can make. We will use the results of this exercise to inform the aggregated graphs that we provide to our paid subscribers. We are inclined to make adjustments to the methodology for aggregating insider transactions as we want to provide as much signal to our users as possible.


Like last time, we are presenting this analysis as an academic exercise and not as any recommended investment strategy or investment advice.


Quick Results


While the Russell 2000 achieved a 29.8% return from January 2020 to April 2025, and the S&P 500 achieved an 86.5% return, using the Insider Sentiment Tracker constructed from profitable insiders to determine going between the Russell 2000 and a zero-interest checking account achieved a return of 145% over the period. By handily beating the S&P 500, these results demonstrate that Insider Sentiment is still highly valuable as a forward-looking macroeconomic indicator in 2025.


The Methodology to Analyze the Predictive Ability of Insider Transactions


We will use the following methodology to measure how effective aggregated insider transactions are as a forward-looking economic indicator. (We publish these aggregated insider transactions for our paid subscribers on the Dashboard at InsiderSentiment.com as a ratio of companies experiencing net insider buying in any particular month.)


We will measure a simple $100 portfolio invested in IWM (the Russell 2000 ETF) as a baseline. In addition, we will define what we call the Tracker-UP strategy: When the Insider Sentiment Tracker returns a value in a given month that is above a certain threshold (which we will define later), we buy IWM on the first day of the following month.


If the Tracker returns a value below the threshold, we sell IWM and put our money into a non-interest bearing checking account. The fact that we earn zero interest when we are out of the stock market understates the true value of our switching strategy.


Similarly, we also define the opposite strategy, which we call Tracker-DOWN. When the Insider Sentiment Tracker is below a certain threshold, we buy IWM on the first of the following month. And when the Tracker is above that threshold, on the first of the following month we will sell IWM and put our money into a non-interest-bearing checking account. 


Advantages of the Strategy


Since we are using the current month’s realization of the Insider Sentiment Tracker to determine the next month’s actions, this is an implementable strategy that investors could undertake. Additionally, the fact that we are investing in a checking account with zero market risk and zero return, and the fact that we are not short the market at any time, means this is also a conservative strategy. More specifically, the risk of our switching portfolio will be lower than IWM while our return is also biased toward zero. Finally, we utilize both the Tracker-UP and Tracker-DOWN strategies in tandem to demonstrate that, not only will using the Insider Sentiment Tracker as a positive indicator create wealth, but additionally that using the Tracker as a negative indicator will destroy wealth.


Full Specifications


The full specifications of the experiment are as follows:


  1. In addition to IWM, we will also test SPY (S&P 500 ETF), IWB (Russell 1000 ETF), and RSP (equally weighted S&P 500 ETF). Since the value-weighted stock market indices these days are driven mainly by the handful of largest companies (think magnificent 7 accounting for about 30% of S&P 500 index), we expect insider sentiment to have valuable predictive ability with small companies, such as those in the Russell 2000 or RSP, but not necessarily for indices that include the largest companies, such as the S&P 500, the Russell 1000, or the Russell 3000.


  1. We will test using aggregates created from all insiders, but also using only insiders who trade profitably. We will also test adding other transaction types in addition to insider purchases and sales.


  1. We will vary the cutoffs that are used to determine the next month’s buy or sell decision, from -3% to 12% above the average value of the Insider Sentiment Tracker to that point in time.


  1. We will run the test starting from COVID onward or last five years only (January 2020 to April 2025).


  1. Finally, we will test the strategy using monthly realizations of the Insider Sentiment Tracker as well as using the 3-month moving average, since we report both on the Dashboard.

Full Results

The graph below shows the best results achieved by the experiment. Our two baselines in the chart, the buy-and-hold $100 portfolios consisting of IWM and SPY, achieved 29.8% and 86.5% increases over the period, respectively.


The blue line represents the Tracker-UP strategy when going between IWM and a non-interest bearing checking account by using signals from the Insider Sentiment Tracker, derived from profitable insiders only, using a 0.06 cutoff (meaning, buying the IWM when this month’s Insider Sentiment Tracker value is 0.06 above the historical average buying ratio). This strategy achieved a 105% return over the period, beating both IWM and SPY.


The best performing line (green) achieved a 145% return over the same period. This strategy differs from the blue line in that it uses trailing 3 month average values of the Insider Sentiment Tracker to determine the buy/sell decision, and it uses a -0.02 cutoff. That the best results were achieved here with a -0.02 cutoff is not surprising, since taking the average over 3 months significantly reduces the volatility, and so the ideal cutoff should be lower to reflect this.


Finally, the red line represents the opposite of the green line strategy, which is buying IWM when Insider Sentiment is below the historical average, as measured by trailing 3 month average realizations. This strategy destroyed wealth, resulting in a -33% return.



A graph showing the performance benefit of using insider signals compared with portfolios holding IWM and SPY.



Results achieved for each group of insiders is presented in the table below along with baseline ETFs for reference. The best performing delta cutoff is shown in parenthesis. As expected, the strategy only achieved success when used to determine buying and selling of IWM, so all Tracker-UP figures presented here are testing the strategy of buying and selling IWM based on insider signals.




From the table we first notice that the best performing Tracker-UP series when using both monthly realizations and 3-month trailing averages is the series that uses profitable insiders with only purchases and sales, no additional transaction types. Secondly, though both achieve impressive results, using trailing 3-month averages to go in and out of IWM handily beats using monthly realizations. Finally, we notice that the best performing iteration of all Tracker-UP series outperforms plain IWM, and all Tracker-UP series using 3-month averages outperform both plain IWM and plain SPY. That is highly significant. 


The top performing Tracker-UP strategies achieved significant improvements in risk adjusted returns over IWM, but only moderate improvements over SPY. Additionally, Tracker-DOWN strategies consistently destroyed wealth when paired with IWM as predicted.

Discussion

These are strong results. Using profitable insiders allows us to beat the S&P 500 using either monthly realizations of the Insider Sentiment Tracker as well as 3-month moving averages. This is practically significant as only 23.7% of large-cap funds were able to beat the S&P 500 over the past 5 years.1 And again, this is a conservative strategy, where the portfolio was moved into a zero-interest checking account in off months and is not short the market at any time.


In addition, the results are fairly robust with respect to the subset of insiders selected as well as the cutoff value chosen. This is a strong indicator of a real signal as opposed to plain luck, where we might see one or two series perform well while the rest fall flat. The fact that Tracker-DOWN strategies consistently lowered portfolio performance when paired with IWM is also evidence of the positive predictive ability of insider signals.


We used these results to update the series that we display on the dashboard. Current subscribers will notice a new tab called Profitable Insiders where we will represent our new flagship Insider Sentiment Tracker using only purchases and sales, to reflect the most successful series in our experiment. 


We discuss some of the limitations below and why we present this analysis as an academic exercise and not as an investment strategy. Our main goal in presenting this analysis is to convey the value that is contained in insider transactions and validate its importance as a forward-looking macroeconomic indicator, and we believe the results presented here do achieve this convincingly.

Limitations

Limitations of this exercise, and why this should be seen as an academic exercise and not as investment advice, are as follows:


  1. Transactions dated within a month are sometimes filed late, not entering the data until days following the end of the month. Our exercise captures all of these late filings, but for an investor making a decision at the end of each month to determine next month’s actions, these late filings would not be factored in. These late filings can cause the monthly realizations of the Tracker to change after the end of the month.


  1. In theory, an investor doing this strategy in real time would always use the 10 year trailing Insider Sentiment Tracker average to calculate the buy and sell cutoffs, but we did not have access to 10 years of the Insider Sentiment Tracker prior to 2020. The Insider Sentiment Tracker data that we used started in 2015, so we just used all available prior data to calculate the average for the purposes of the buy and sell cutoffs.


  1. There is of course luck involved in choosing the cutoff level that determines the buy and sell decisions, and it may not be the case that the best performing cutoff levels in this analysis would also perform the best going forward. However, there is some comfort in choosing a cutoff which is attributed to the robustness of the results. These results show that the Insider Sentiment Tracker performs the best when monthly realizations are 0.04 to 0.06 higher than the mean, and when 3-month trailing averages are cut off right at the mean or slightly below it.


    References

    1. https://www.spglobal.com/spdji/en/research-insights/spiva/


    Authors’ disclaimer: Neither this article nor anything contained herein constitutes any recommendation or advice as to whether any investment or investing strategy is suitable for any particular investor. For financial advice, consult a licensed financial advisor.