Kaufman’sMost Popular Books (available on Amazon)
Trading Systems and Methods, 6th Edition. The complete guide to trading systems, with more than 250 programs and spreadsheets. The most important book for a system developer.
Kaufman Constructs Trading Systems. A step-by-step manual on how to develop, test, and trade an algorithmic system.
Learn To Trade. Written for both serious beginners and practiced traders, this book includes chart formations, trends, indicators, trading rules, risk, and portfolio management. You can find it in color on Amazon.
You can also find these books on our website, www.kaufmansignals.com.
Blogs and Recent Publications
Find Mr. Kaufman’s other recent publications and seminars at the end of this report. We post new interviews, seminars, and reference new articles by Mr. Kaufman each month.
MAY Performance in Brief
A mixed month but not the gains in the Trend System that happened in the index markets. We think we are doing well given the erratic moves in the market and comments in the news that all trend followers are suffering. The biggest gains have been in tech stocks, but after devastating losses in that industry, it will be a while before a trend system can add them to a portfolio. Then again, that could be safe.
Short-term trading programs did much better, with the Divergence program for stocks gaining 10%.
We did look at whether last month was “near” the bottom. We’re looking for better performance going forward, even if we are still trying to reduce the risk.
We continue to make small changes to our strategy for both stocks and futures, looking to improve returns and lower risk. It’s an on-going process and there is much to learn when the markets produce a new pattern.
Major Equity ETFs
The index markets recovered much more than expected, but are not out of the woods yet. Tariffs are still the issue as well as the potential for increasing the U.S. debt. Companies are still cautious. Performamce of the S&P and Nasdaq are nearly identical, while the small caps (IWM) show that investors are not as willing to take on more risk.
CLOSE-UP: How to Test a Strategy
Testing a strategy is the only way I know of deciding if an idea might work going forward. But to do that, there are choices that make the result better or worse. The following are the most likely prospects:
- 1. Do you test all the data or only the most recent?
- 2. Do you leave the most recent data as “out-of-sample?”
- 3. Do you leave the oldest data as “out-of-sample?”
- 4. Do you use “walk-forward” testing?
To solve this, I needed to write my own optimization and walk-forward programs. They allowed me to test futures data going back to 1982. For some development platforms, that was too much data.
I used a simple moving average system, where we bought and sold when the trend line turned up and down. There were no commission charges. At the end, I’ll compare the moving average with a breakout strategy – just to make things more complicated!
This study will only use S&P futures and corn futures, both back-adjusted. They both go back to the early 1980’s and they represent very different patterns. The S&P has gone more-or-less steadily higher, while corn has had some ups and downs, but has gained only due to the changes in the U.S. dollar and Consumer Price Index. They represent two extremes.
I’m separating the data into 5-year intervals for these tests. It might have been better to use 1-year intervals, or even 6 months, but that would have been a bigger project and more lines to analyze, and I thought this would tell us enough.
Results of Not Using the Newest Data
Figure 1. Corn futures moving average results 1982-2019
- Best result: 70-day, long $6,931,965, short $15,605,524, total $22,537,510
- Average result: long $5,378,070, short $11,293,397, total $16,671,474
I like using the average result because it tells me whether the strategy is robust. In the case of corn (Figure 1), we see that all calculation periods were profitable over the data through 2019. Because of seasonality, the best results are in the area of one calendar quarter (63 trading days). Prices tend to rally then sell off over a longer period, giving shorts a larger profit.
Doing the same for the S&P (Figure 2), leaving off the last five years, we get a very different picture. Short trades lost money at every calculation period, even though the declines in 2000 and 2008 were large. Most other declines were severe but short-lived, making it difficult to capture profits using a trend system.
Figure 2. S&P futures moving average results, 1982-2019
- Best result: 150-day, long $16,405,873, short ($8,616,694), total $7,789,198
- Average result: long $10,565,453, short ($13,478,597), total ($29,131,145)
Results of Not Using the Oldest Data
We now remove the oldest five years and keep the newest. You might find that a more logical approach, but then if we test the oldest data with the best parameters, we might find that the results are disappointing because the market has changed.
The pattern for corn changes considerably, but the best results are still the same, a 70-day average (see Figure 3).
Figure 3. Corn futures moving average results, 1987-2025.
- Best result: 70-day, long $11,378,196, short $11,496,579, total $22,874,730
- Average result: long $8,687,998, short $7,113,621, total $15,801,611
Perhaps removing 5 years from a total of 43 years is only 12%. That must not be enough to change the results, even though the patterns seem to change and the total profits are different.
Not so with the S&P. (see Figure 4). The pattern is similar, with all short trades netting a loss and the best parameters near the far right.
Figure 4. S&P futures moving average results, 1987-2025
The only way to see the difference is to look at the totals.
- Best result: 150-day, long $13,087,232, short ($9,174,217), total $3,913,019
- Average result: long $10,147,886, short ($12,787,725), total ($2,639,835)
Removing the oldest 5 years instead of the most recent 5 years barely changed the results, and the best moving average periods were still at 150 days.
Out-of-Sample Summary
I’m sure that using a smaller amount of data would have changed the results, but then you would be targeting a narrow set of price moves. And perhaps 30 years of data is too much, but it does cover a lot of patterns. My own inclination is to use more data to see more patterns. The results may not be as good as you want, but are most likely to be more realistic.
The study above shows that removing the first or the last data doesn’t make a difference if you’re testing enough data.
Walk-Forward Testing
I’ll tell you in advance that I don’t use walk-forward testing. There are too many variables and too many restrictions. You need to choose the in-sample and out-of-sample periods to work with your strategy. A macrotrend would need longer periods. If the out-of-sample period is short, you need to retest often.
The biggest problem is when the first results are not to your liking. You can’t change the strategy because you no longer have out-of-sample.
In the following cases I used 5-year intervals. Testing in-sample for 5 years, then out-of sample for 5 years, then moving forward. That may be too long but the results are easier to read! The following results come from my own walk-forward program.
In both corn (Figure 5) and the S&P (Figure 6), the in-sample results are 70% to 80% greater than the out-of-sample results, showing the unseen data is not the same as the in-sample data. Yet the results are good.
Corn Walk-Forward
Figure 5. Corn futures walk-forward tests.
S&P Walk-Forward
Figure 6. S&P futures walk-forward tests.
A Final Accounting, Since 2022
To be fair, Figure 7 shows the results of the most recent out-of-sample performance. The walk-forward test is better than the moving average tests for both corn and the S&P. As odd as it seems, the difference from a 70 to a 75 calculation period for corn made a big difference in the results. I checked that to be sure. Sometimes a small difference in timing can avoid a bad trade. It is also interesting that a big difference in the S&P calculation periods gave similar results. But who am I to question that once I’ve checked the numbers?
Figure 7. Summary of moving average and walk-forward tests from 2022.
I don’t see the advantage of using in-sample and out-of-sample data. Results for these tests show the same calculation periods. Of course you can change that by selecting shorter period, which might prove less stable. The advantage of using more data is that the strategy is robust, although perhaps not as profitable on paper. My answer is to test all the data.
Use Different Intervals for Testing
If you test the same periods for either stocks for futures, you will tend to get the same calculation periods. Most markets react similarly to economic and market events. In particular, the U.S. index markets, interest rates, and the U.S. dollar.
Try to use different test intervals. I have written a spreadsheet that uses random numbers to generate test periods. That not only allows me to see different data, it will tell me how much the performance of these periods differ from each other.
Let’s Not Forget the Breakout Strategy
I always like the breakout strategy. Breakout results tend to have a different pattern than a moving average, shown below for corn in Figures 8 compared with the moving average in Figure 9. The best breakout was at 55 days with a gain of more than $15 million.
The moving average had gains between 70 and 95, with returns near $28 million. Both show positive returns for both long and short positions.
The patterns are very different with the breakout focusing on a shorter time period, perhaps capturing seasonality, while the moving average is looking at the bigger trend across years.
Figure 8. Corn futures using breakout strategy, 1982-2025
Figure 9. Corn using a moving average, 1982-2025.
The S&P has similar patterns for both the breakout and moving average (Figures 10 and 11). Short sales lose money everywhere, but they are not as severe for the breakout. Breakout profits top at near $25 million, and moving average profits top at about $17 million. The breakout has a peak at 120 while the moving average keeps getting better as the calculation period increases.
Figure 10. S&P futures using breakout strategy, 1982-2025.
Figure 11. S&P futures using a moving average, 1982-2025
Which Would I Use?
Every investor has a different risk tolerance. Mine is probably bigger than yours. I don’t like getting stopped out frequently, so I prefer the breakout. Conceptually, it holds on to a position until something new happens – a new high or new low. It often has more than 50% good trades, compared to 30% for a moving average.
About using in-sample and out-of-sample data, or using walk-forward testing, I prefer just testing more data. The more data, the more patterns, and the lower the result. However, it is probably more realistic.
I have heard that using a shorter data period might be more accurate, much the same as walk-forward testing, but it will decay faster. Using more data implies that it has already decayed! I’m not sure if that’s good or bad; however, I use more data and I accept larger drawdowns.
Results from walk-forward data can vary considerably, from a fast calculation period when the data shows a trend, to a very slow calculation periods when there is no trend, Figure 12 shows the different calculation periods for bond futures, which tend to be very trending. The walk-forward period was very short, perhaps similar to what you might use.
Figure 12. Moving average calculation period for U.S. bonds using walk-forward testing.
A lot of moving average switches does not mean that the walk-forward test didn’t do well. The out-of-sample returns are shown in Figure 13. There are other results that did better, and some not-so-good.
Figure 13. Walk-forward out-of-sample returns for Bond futures.
I’ve tried to give you my insight into testing. There are a lot of decisions and a lot of choices. But testing is still the most important part of developing a strategy. I hope my explanation helps you along the way.
A Standing Note on Short Sales
Note that the “All Signals” reports show short sales in stocks and ETFs, even though short positions are not executed in the equity portfolios. Our work over the years shows that downturns in the stock market are most often short-lived and it is difficult to capture with a longer-term trend. The upwards bias also works against shorter-term systems unless using futures, which allows leverage. Our decision has been to take only long positions in equities and control the risk by exiting many of the portfolios when there is extreme volatility and/or an indication of a severe downturn.
PORTFOLIO METHODOLOGY IN BRIEF
Both equity and futures programs use the same basic portfolio technology. They all exploit the persistence of performance, that is, they seek those markets with good long-term and short-term returns on the specific system, rank them, then choose the best, subject to liquidity, an existing current signal, with limitations on how many can be chosen from each sector. If there are not enough stocks or futures markets that satisfy all the conditions, then the portfolio holds fewer assets. In general, these portfolios are high beta, showing higher returns and higher risk, but have had a history of consistently outperforming the broad market index in all traditional measures.
PERFORMANCE BY GROUP
NOTE that the charts show below represent performance “tracking,” that is, the oldest results since are simulated but the returns from 2013 are the systematic daily performance added day by day. Any changes to the strategies do not affect the past performance, unless noted. The system assumes 100% investment and stocks are executed on the open, futures on the close of the trading day following the signals. From time to time we make logic changes to the strategies and show how the new model performs.
Groups DE1 and WE1: Daily and Weekly Trend Program for Stocks, including Income Focus, DowHedge, Sector Rotation, and the New High-Risk Portfolio
The Trend program seeks long-term directional changes in markets and the portfolios choose stocks that have realized profitable performance over many years combined with good short-term returns. It will hold fewer stocks when they do not meet our condition and exit the entire portfolio when there is extreme risk or a significant downturn.
Equity Trend
A quiet month after a series of nasty losses. The Weekly program did a bit better than the Daily program after taking more of a loss. We’ve made a few small changes to the strategy, looking to increase returns and reduce risk, but nothing that would change the basic approach. Our new risk hedge worked, although we applied it a bit late in the sell-off. The market is always teaching us new things.
Income Focus and Sector Rotation
Fractional gains and losses in the Income Focus program for May. Continued confusion about higher and lower rates prevents any serious gains. However, rates have not gone up and it is possible that they will (eventually) go down. That would be good for this program.
Weekly Sector Rotation
A reasonable return of more than 4% while still holding the same ETFs – Utilities, Staples, and Financials. While investors are rushing back into high-tech stocks, we expect these ETFs to outperform.
DowHedge Programs
Gains of better than 2% in May are a bit below the Dow Index gains, showing that the “best” of the Dow are not living up to their potential! Because this program tracks the market using the Dow, it has lost less than most strategies and is likely to recover faster.
High-Risk Portfolios
Small losses in May, even though we thought that the tech stocks were driving the market. The problem seems to be that a macrotrend program has not gotten on board with the rally. Perhaps that’s good.
Group DE2: Divergence Program for Stocks
The Divergence program looks for patterns where price and momentum diverge, then takes a position in anticipation of the pattern resolving itself in a predictable direction, often the way prices had moved before the period of uncertainty.
As remarkable as it seems, this program gained 10% for the smaller portfolio and nearly 6% in the larger portfolio, putting it ahead of all the equity index market for the year. This program takes advantage of a pause in the trend, looking for a further advancement in the trend.
Group DE3: Timing Program for Stocks
The Timing program is a relative-value arbitrage, taking advantage of undervalued stocks relative to its index. It first finds the index that correlates best with a stock, then waits for an oversold indicator within an upwards trend. It exits when the stock price normalizes relative to the index, or the trend turns down. These portfolios are long-only because the upwards bias in stocks and that they are most often used in retirement accounts.
Small gains of 2% to 3% in both portfolios moderates the losses for the year. Buying a pullback in a declining market has not been a strategy that has worked this year, but losses have been smaller than other strategies and gains this month are a good sign.
Futures Programs
Groups DF1: Daily Trend Programs for Futures
Futures allow both high leverage and true diversification. The larger portfolios, such as $1million, are diversified into both commodities and world index and interest rate markets, in addition to foreign exchange. Its performance is not expected to track the U.S. stock market and is a hedge in every sense because it is uncorrelated. As the portfolio becomes more diversified its returns are more stable.
The leverage available in futures markets allows us to manage the risk in the portfolio, something not possible to the same degree with stocks. This portfolio targets 14% volatility. Investors interested in lower leverage can simply scale down all positions equally in proportion to their volatility preference. Note that these portfolios do not trade Asian futures, which we believe are more difficult for U.S. investors to execute. The “US 250K” portfolio trades only U.S. futures.
Still posting losses in May after adding more markets and switching to the micro gold contract. We like the new portfolio, although the market keeps flopping around, first a stronger dollar, then a weaker one. It will stabilize eventually, we just don’t know when!
Group DF2: Divergence Portfolio for Futures
A good month in May, adding 4% to 5% and showing small losses for the year. While still in a prolonged drawdown, I am ever hopeful that this marks a turnaround.
Insert Futures Divergence charts
Blogs and Recent Publications
Perry’s books are all available on Amazon or through our website, www.kaufmansignals.com.
June 2025
Yes, another article! “There is Money To Be Made On The Weekends – But You Need to Know The Market,” appeared in the June issue of Technical Analysis of Stocks & Commodities.
Perry will be giving a Webinar to “Trading Heads” in Mumbai, India on June 5 at 9:30 AM New York time. He’ll be talking about “Not the Usual Diversification.” Join him if you can.
May 2025
You’ll find Perry’s article “Trading the Channel” more interesting than usual. Published in the May issue of Technical Analysis of Stocks & Commodities, it looks at various ways of construction a channel, and one very profitable one.
Perry also addressed a Spanish class where they are building algorithmic strategies. Called ROBOTRADER, it in ETSIT-UPM (Escuela Técnica Superior Ingenieros Telecomunicación- Universidad Politécnica Madrid). The presentation is about Diversfication (in English) and available on youtube.
April 2025
Perry did a studio interview with Jeff Baccaccio (“Rfactory”) in London in March. It is a fine production and a good interview. He has put it on youtube. I hope you enjoy it.
YouTube: https://youtu.be/jmR359jHYBQ?si=IHQ5bVLijGFM19qF
Another article in Technical Analysis of Stocks & Commodities for April, “Do Stops Really Work?” The conclusion even fooled Perry.
March 2025
Perry looks at an old standard in “Revisiting the 3-Day Trade,” which appeared in Technical Analysis of Stocks & Commodities in the March issue.
February 2025
Another article, “Chasing the Market” appeared in the February issue of Technical Analysis of Stocks & Commodities. It answers the question, “Can you make money entering the market after a big move?”
Perry enjoyed the “Fireside Chat” at the Society of Technical Analysts (STA) in London on Tuesday, February 11. It should be available for viewing on their website. He also taped another interview and we’ll let you know how to see it when it’s released.
He also posted “If you think the market will tank, here’s a plan” on SeekingAlpha. It has received lots of view and good comments, although it is advising deleveraging.
December 2024
“Overlooked Strategy Rules” appeared in the December issue of Technical Analysis of Stocks & Commodities. We tend to overlook certain rules that can make a big difference to results. This article looks at scaling in and scaling out of a position, delayed entries, correlations, and other simple but important rules.
October 2024
“Trading a Breakout System” was published in Technical Analysis of Stocks & Commodities. It looks at whether it’s better to enter on the bullish breakout, wait for confirmation, or buy ahead of the breakout. It’s a practical look at improving breakout results.
September 2024
Two articles posted by Perry, “The N-Day or the Swing Breakout,” (Technical Analysis of Stocks & Commodities) looking to see which is better. You would be surprised.
A look at deleveraging Artificial Intelligence stocks, a shorter version of the article posted in our “Close-Up” section. It appeared in Seeking Alpha earlier in September.
August 2024
“Theory Versus Reality” was published in the August issue of Technical Analysis of Stocks & Commodities. It discusses price shocks, diversification, predicting performance, and more.
July 2024
Perry posted a new article on Seeking Alpha, “Capturing Fund Flows.” It a good strategy for someone that wants to add some diversification. It only trades 3 days each month!
Older Items of Interest
Perry did a studio interview with Jeff Baccaccio (“Rfactory”) in London in March 2025. It is a fine production and a good interview. He has put it on youtube. I hope you enjoy it.
YouTube: https://youtu.be/jmR359jHYBQ?si=IHQ5bVLijGFM19qF
Perry was interviewed on June 27, 2024 by Simon Mansell and Richard Brennan at QuantiveAlpha (Queensland, Australia), a website heavy into technical trading. It appears on their website.
On April 18th, 2023, Perry gave a webinar to the Society of Technical Analysts (London) on how to develop and test a successful trading system. Check their website for more details, https://www.technicalanalysts.com..
Perry’s webinar on risk, given to the U.K. Society of Technical Analysts, can be seen using the following link: https://vimeo.com/708691362/04c8fb70ea
For older articles please scan the websites for Technical Analysis of Stocks & Commodities, Modern Trader, Seeking Alpha, ProActive Advisor Magazine, and Forbes. You will also find recorded presentations given by Mr. Kaufman at BetterSystemTrader.com, TalkingTrading.com, FXCM.com, systemtrade.pl, the website for Alex Gerchik, Michael Covel’s website, TrendFollowing.com, and Talking Trading.com.
You will also find up to six months of back copies of our “Close-Up” reports on our website, www.kaufmansignals.com. You can address any questions to perry@kaufmansignalsdaily.com.
© May 2025, Etna Publishing, LLC. All Rights Reserved.