About our Quant ratings :
Quick and powerful insight
Our Quant ratings provide a quick and powerful insight on stocks by addressing 5 factors that have historically been highly correlated to stock returns:
1) Growth, 2) Valuation, 3) Quality, 4) Revisions and 5) Momentum.
For each factor, we compare a variety of quantitative metrics for each company with the rest of its sector to determine a score from A+ (best in sector) to D (worst in sector). Our overall Quant Rating considers the aggregation of the 5 factors.
Machine learning
The metrics we consider in our ratings and their relative importance are different from one sector to another, based on their historical relation to stock returns in that sector. As these relations change over time, our scoring algorithm adapts to focus on the most meaningful indicators. Every month, our scoring algorithm is refined by expanding its dataset. Machine learning enables us to adapt to a changing world where investment styles and criteria change over time.
In backtesting the performance of the strategy, our algorithm considers the financial dataset and knowledge available every month since January 2008 to calculate quant ratings for every stock in the same fashion it does in the real world everyday. By constantly optimizing our scoring algorithm based on data available at the date of the decision, our machine learning process reduces the risk of bias and improves the probabilities of continued alpha generation over time.
Our Quant Picks
Our Quant Picks highlight about 30 stocks with the best quantittive rankings in their sector. Our recommended stock weights always represent an equalweight and the sector weightings closely match the index, making sure we generate alpha purely from stock selection, not from sector allocation.
Quant Picks backtest performance
The backtest performance below represent the results of a simulation of our quant-based strategy starting in 2008. It assumes a monthly rebalancing of the portfolio with equal weights in every Quant Pick on the 1st of each month. It represents the return an investor would have achieved by copying and equally weighting our Quant Picks every month. Past performance is not indicative of future returns. We present the backtest performance below for informational purposes.
U.S. equities
Cumulative annual returns
Risk metrics (10 years)
Civil year returns
Backtest returns as of December 31, 2025
DISCLAIMER: The returns shown above are the result of a backtest simulation implementing our quant-based strategy. Although we take great care to minimize potential biases, backtest returns may present some biases and may not be indicative of future performance.
Canadian equities
Cumulative annual returns
Risk metrics (10 years)
Civil year returns
Backtest returns as of December 31, 2025
DISCLAIMER: The returns shown above are the result of a backtest simulation implementing our quant-based strategy. Although we take great care to minimize potential biases, backtest returns may present some biases and may not be indicative of future performance.
Backtest & methodology :
Backtesting intrinsically comes with limitations and biases. Machine learning eliminates most of them.
At QDD Research, we've been using machine learning before even knowing it was machine learning. We simply leverage a very wide range of financial data, some of which we create from qualitative data through content analytic, and we constantly optimize how we use it based on its historical effectiveness across different sectors and geographies.
Before hearing the term "Machine Learning", QDD Research's founders found out that the world was evolving and that traditional backtesting was useless and deceptive. With the amount of data available today, it's easy to create a strategy that would have worked in the past 10 years. The machine learning process we've been implementing since the begining is fundamentally different. Every day, our algorithm recalculates the historical correlations between stock fundamentals and their excess returns. Every day, the process evolves, improves itself, always using only data available at that day. Our backtest plays back the strategy starting in 2008, always using data available on that day to determine factor weightings and improve itself every month.
Content Analytic? Using Python, we read thousands of press releases every day and convert them into data. For instance, a press release stating "announces stocks split" or "announces share repurchase" becomes a quantitative data in our dataset. This expands our dataset with data unavailable to most. Bloomberg tells you the date a stock splited, but by analyzing the content of press releases, our data shows the date the stock split was announced, which is the most important input for a trading strategy. With over 500 qualitative signals translated to quantitative variables, we use a much wider range of data than most investors.
A+ Quant Picks: Our Quant Picks are the stocks rated A+. They are the very best in every sector based on quantitative data about Growth, Valuation, Quality, Revisions, and Momentum. Historically, A+ rated stocks have consistently been outperforming the market.
Sample: For US and Canadian equities, our stock sample from the beginning of the backtest to this day is the S&P500 and the S&P/TSX index as of the date of the sampling. By always ranking and choosing stocks that were in the index back at the time of the selection, we avoid surviroship bias. As a result of sampling in these large cap indexes, our Quant Picks are always focused on large cap stocks. For international stocks, we provide rankings using the same system as US and Canadian equities but do not provide performance data to this day.
Sector neutral : Many competing quantitative services show off great historical returns but they've been overweight in tech for years. Is it skill or is it luck? At QDD, we believe there are great opportunities in every sector. We think sector picking is hazardous and believe the best long-term approach is to stay sector neutral and generate alpha across all sectors. That's why our Quant Picks always cover every sector, with weightings closely matching the index.
Frequently Asked Questions :
What's your stock dataset?
To avoid survivorship bias, our backtest considers all the stocks in the S&P500 and S&P/TSX as of the date of every decision. This way, our backtest performance applies our quantitative screenings to the stocks that actually comprised the index back in the days, therefore avoiding survivorship biases.
Equal weight: As you may have noticed, every stock has the same weighting in our Quant Picks portfolio. The weights are simply 100% divided by the number of stocks recommended. The number of stocks in the sector is based on the sector weight in the index.
Trade frequency and turnover rate: If you apply our Quant Picks monthly following our email alerts, you can expect an average of 3 trades per geopgraphy per month. Many subscribers use these recommendations as systematic stock calls, while others use them as stock ideas.
What are the factor weights? It depends on the sector and the geography. Every day, our algorithm backtests what have historically been the best factor weightings, specifically on that sector and geography. So some days, stocks in the energy sector get more weight on momentum. Sometimes, they get more attention based on their earnings revisions. Portfolio factors will drift over time based on the historical correlation to stock returns. But at every time, in every sector and geography, the overall score considers at least 3 factors.
For instance, in the Tech sector, in the US, for the Quality factor alone, we consider 84 different quantitative inputs from every Tech company in the S&P500. For each quantitative input, we rank the companies in the sector from the best to the worst. If there are 100 companies in the sector, the one with the best ROE gets 100 points for that input. The company with the worst ROE in the sector gets 0 for that input. We then sum up the percentile for each data input in the Quality factor to determine a Quality score for each Tech stock in the US. We proceed in the same way for each of the 5 factors, and our own in-house qualitative data generated through content analytic. This way, each stock gets a score for each factor that is based on its percentile rank in its sector for various fundamental data.
Then, what is the relative importance or the weighting of every factor? It depends on the sector and the geography. Based on historical data, some factor are more correlated to returns for US Tech stocks than they are for Canadian Energy stocks. Our algorithm constantly determines the factor weightings based on their historical correlation to stock returns in that specific sector and geography. This is also how the backtest has been implemented, by recalculating the factor weightings every month of the backtest based on the past historical correlations as of the date of every decision. For instance, the table below shows what multiplier were used for each factor score to determine the overall quant score, as of September 30th, 2025. This matrix evolves every day as the correlations between our factor scores and stock returns evolve, hence adapting to changing market conditions, preferences and styles.
Factor weighting matrix for US Equities as of September 30th 2025
How to use our email alerts? Our Quant Picks identify the top-ranked stocks in every sector. For instance, in the financials sector in the U.S., our Quant Picks identify the stocks ranked #1 to #4. If a stock drops from rank #4 to rank #5, it will be removed from our Quant Picks and replaced by the new #4. It does not mean that the stock ranked #5 is a bad stock. It’s still one of the best, and there’s a high probability that it might come back to the Quant Picks soon. Since data can change rapidly, rankings can change rapidly too, and generate a lot of additions and deletions in our Quant Picks.
Our general advice is to consider every new addition to the Quant Picks as a buy candidate, but not necessarily to interpret deletions as sell signals.
To reduce the noise caused by daily data changes and keep a reasonable portfolio turnover, our Quant Picks are updated monthly. We send the new additions and deletions to our subscribers monthly. Our Quant Picks backtest performance also assumes a monthly rebalancing of the portfolio on the 1st of each month. You could follow a similar approach and rebalance your portfolio monthly based on our Quant Picks and expect similar outcomes.