Great article on the power of combining factor investing with a systematic momentum overlay. It’s rare to see a strategy that effectively mitigates the decade-long underperformance periods often seen in pure factor plays.
Have you tested how sensitive the CAGR is to the specific look-back period—for instance, does using a 6-month vs. a 12-month momentum window significantly change the drawdown protection results?
I have tested every possible look-back period with the monthly data. The CAGR stays significantly above market level, regardless of the lookback period, but the alpha gradually declines the further I move away from the selected monthly intervals, as you would expect.
So, it seems the system would do well with most settings, but I optimized it for the most consistent outperformance based on the 2000-2020 data.
I’ve been following MarketFighter’s site pretty much since the early days — mostly just reading quietly and trying to understand the logic behind the strategy. Over time, the consistency of the updates and the transparency really kept me coming back.
One thing I was curious about was whether the returns in USD would match the EUR version. After checking it myself, the USD performance is basically similar, and in some periods even a bit higher. That gave me enough confidence to stop watching from the sidelines.
So I ended up buying the suggested ETF strategy myself — and yes, the returns are real. Not backtests, not theory. Actual results in my own account.
Big thanks to MarketFighter for putting the work into this. I have decided to subscribe because it’s rare to find something that’s both simple and actually holds up when you test it on your own.
Just sharing this for anyone else who’s been quietly observing like I was.
You can feel the years of trial and patience behind this. The discipline to stick with a system is what most people struggle with. It’s a steady approach that seems to value consistency over excitement.
It does require discipline to stick with a system.
And yes, my main priority was to build a system with high consistency in the returns. There are lots of trading systems available that deliver higher CAGRs but with much higher volatility. Personally I don't have the risk profile to stick with these, so I had to invent my own.
I found MF via ChatGPT. I told ChatGPT to go find me a strategy that beats the SPY but also carries less risk. After a few days of searching ChatGPT came back with a single result...TheMarketFighter.
I signed up and am going to give it a try.
Also, thank you for publishing so much helpful and well written information about it. I appreciate the transparency and the obvious love you put into your work.
I've been reading for a while and find the approach genuinely interesting. Combining sector and factor momentum into a single monthly rotation is a clean idea, and you're more transparent about the mechanics than most people selling signals. Credit for that.
Before I allocate seriously to it, four things I'd like to understand:
1. Benchmark construction — is the MSCI World column price return or net total return? Compounding your annual figures gives roughly 3.8% p.a. over the period, which is about two points below MSCI World's actual net total return. If the strategy column includes dividends and the benchmark doesn't, a meaningful slice of the reported alpha would be a measurement artifact.
2. Live vs. backtested — can you break out performance from the date you started publishing signals, separately from the simulated history? An out-of-sample record is the only part that can't be optimised in hindsight.
3. 2008 — the strategy shows -10.24%. Consumer staples, the strongest global sector that year, fell roughly 15-20%. A long-only rotation between sectors and factors shouldn't be able to beat its own best available holding, so I assume there's a cash or moving-average filter in the model. Could you describe it?
4. Investability pre-2013 — most of the major factor ETFs didn't launch until 2012-13, and several factor indices were backfilled. How much of the 2000-2012 record is simulated on index data rather than tradeable instruments, and does it account for spreads and TERs?
Not trying to be difficult — these are the questions I'd ask any manager, and clear answers would make me more confident rather than less. Thanks.
1. The data I used to build this was based on price returns, so I decided to stick with this for all data to make a fair apples-to-apples comparison. You're right that net totals are higher. This applies to the benchmarks as well as the strategy returns.
3. Yes, there's a drawdown protection layer, also based on relative price momentum. I have tested this part on data back to 1871, and it kicked in last time in 2008-2009. It is described in this article.
4. It's not based on specific ETF implementations as they vary across the globe. It's based on the underlying indices and not accounting for spreads, TERs, or taxes, as all of this varies upon execution.
I like how clearly you separate the factor and sector sleeves from the drawdown filter. The move to cash is the part I’d most like to understand: what market momentum rule triggers the exit and re-entry, and how sensitive were the results to that choice? It would also be useful to see the maximum drawdown alongside the annual returns. That would make the “without taking more risk” claim much easier to assess.
Thanks for your comment, that's a great point. I will try to incorporate a table with annual drawdowns in a future post 👍 For now I can say that the largest drawdown of the strategy over the full backtest (back to 2000) was -15.3% in 2020 during Covid. But this number is based on monthly closing prices, so it would of course be deeper with daily figures. The protection layer only plays a role during major declines, and it's based on simple monthly closing prices just like the rest of the strategy.
Thanks for clarifying that the −15.3% drawdown is based on monthly closes. I’d be interested to see the annual drawdowns when you publish them. Daily figures would also help show what investors experienced between month ends.
Really enjoyed reading this because it focused on something many investors overlook: the importance of having a repeatable process. Outperformance rarely comes from chasing every market trend or constantly reacting to noise. More often, it comes from having a clear framework, staying disciplined, and sticking to it through different market cycles.
What stood out to me is that good investing often looks boring from the outside. It is patience, consistency, and sound decision making under uncertainty repeated over time. That is where the real edge usually comes from. Great breakdown.
Well said, and thanks, I appreciate it! 🙏 You’re absolutely right. Patience, discipline, and consistency can take you far, and that combination seems to be underrated.
Thanks for the article. As a fellow computer scientist I loved reading this and your approach. I've downloaded the stock data for the last few years, calculating all the technicals for each stock and running various queries with a view to something similar in an individual sock basis, but your approach is so elegantly simple (appreciate the work to get there wasn't!). Will definitely give this a try.
Thanks, I really appreciate your comment! I've been down that road with individual stocks as well, attempting to build a similar system, but I didn't manage to find the same consistency as I did with the ETF setup. It was too volatile for me :-)
You should execute the trade after you receive the monthly trade signal in your inbox. Ideally on the first day of the month, but the sooner the better.
The whole back-test and my five years of trading the system in real life have been based on trades executed at the turn of the month or on the first trading day of the month.
I haven’t, because we don’t have access to these as retail investors in the European Union. But historically, they would most likely have performed better in the crisis years but also introduced more volatility.
I understand how the dividends are distributed for ETFs. As I understand it, all you return calculations assume the investor fully reinvests all dividends back into the portfolio. What I was asking for is the annual and period returns with the assumption that the investor does NOT reinvest those dividends. This can make a significant difference over the long run. As mentioned above I am looking for income as well as appreciation of principal. Thanks for your quick responses by the way. Looking forward to joining soon.
Thanks for clarifying. Sorry, I don’t have any numbers for the exact US dividends, as I track the strategy with the European ETFs.
But if we assume an average 1.5% annual dividend in total, and the strategy has delivered around 18% CAGR over the past 20 years, the math would be like this:
18% CAGR with dividends reinvested: 2,739%
16.5% CAGR with no reinvested dividends: 2,037%
Be aware that these numbers are based on assumptions. But yes, withdrawing the dividends does reduce long-term compounding, but it gives you the ongoing income stream you're looking for.
Hope this helps, and feel free to send me a DM if you have any further questions :)
Are you able to share the annual returns if the investor did not reinvest the returns back into the portfolio, I.e. non CAGR returns. I am at a point in life where I will need the income vs growing the principal. I am US based so in USD would be appreciated.
Hi Patrick. While the European ETFs are accumulating, the US-listed ones pay out quarterly dividends. Typically they will take up 1-2% of the annual returns, but it depends on the allocations throughout the year. Did this answer your question?
Thank you so much for the article. I am always afraid of overfitting. Is your method to avoid drawdown the same as for the momentum signal? Or do you use another? For me less moving parts are always more stable. Best regards!
Totally agree, the simpler the better! And yes, the drawdown mechanism is based on momentum in monthly closing prices just like the rest of the strategy. It’s so easy to overfit a backtest, and I didn’t want to add other parameters for the same reason :)
The exact rules behind the strategy are proprietary, but the paid subscription provides access to the monthly trading signals covering every trade I execute myself, based on the output of the system.
Quick clarification on the drawdown protection piece — is the move-to-cash decision only evaluated once a month, at the same time as the momentum signal, or do you also monitor and act on the drawdown filter mid-month if conditions shift before the next signal is due? Just want to make sure I'm executing the mechanics correctly.
It’s only evaluated as part of the monthly signal. This is to reduce the risk of false alarms and maintain the simplicity of only having to deal with the strategy once a month.
Great article and inspiring, Question ? I’m an Australian investor , prefer to use ETFs domiciled in AU , tax reasons etc what support do you provide if we join ?
Thank you! Tax is one thing I can’t help with, as rules differ so much from one country to another. I always recommend reaching out to a local tax advisor.
Sure, ETFs domiciled in AU can be just as good as those in US or Europe, as long as they track the same or similar indices.
Was there anything else you needed support for? :)
Thks for responded so quickly , just considering the subscription,concerned if we use the domiciled ETFs can’t really compare with yours , filters ,or indices you mentioned not sure
The majority of my subscribers actually live outside of EU, meaning their returns are different and affected by currency fluctuations. They will never see the exact same numbers as me. What matters is finding ETFs that track the right (or similar) indices as those listed here: https://www.marketfighter.com/p/the-watchlist-see-the-15-tickers
Great article on the power of combining factor investing with a systematic momentum overlay. It’s rare to see a strategy that effectively mitigates the decade-long underperformance periods often seen in pure factor plays.
Have you tested how sensitive the CAGR is to the specific look-back period—for instance, does using a 6-month vs. a 12-month momentum window significantly change the drawdown protection results?
Thank you! Appreciate your comment 🙏
I have tested every possible look-back period with the monthly data. The CAGR stays significantly above market level, regardless of the lookback period, but the alpha gradually declines the further I move away from the selected monthly intervals, as you would expect.
So, it seems the system would do well with most settings, but I optimized it for the most consistent outperformance based on the 2000-2020 data.
Thank you for the knowledge 👏
I’ve been following MarketFighter’s site pretty much since the early days — mostly just reading quietly and trying to understand the logic behind the strategy. Over time, the consistency of the updates and the transparency really kept me coming back.
One thing I was curious about was whether the returns in USD would match the EUR version. After checking it myself, the USD performance is basically similar, and in some periods even a bit higher. That gave me enough confidence to stop watching from the sidelines.
So I ended up buying the suggested ETF strategy myself — and yes, the returns are real. Not backtests, not theory. Actual results in my own account.
Big thanks to MarketFighter for putting the work into this. I have decided to subscribe because it’s rare to find something that’s both simple and actually holds up when you test it on your own.
Just sharing this for anyone else who’s been quietly observing like I was.
That’s great to hear, thank you so much for sharing this! 🙏
You can feel the years of trial and patience behind this. The discipline to stick with a system is what most people struggle with. It’s a steady approach that seems to value consistency over excitement.
Thanks for the words! 🙏 You are absolutely right.
It did take many years to reach this point.
It does require discipline to stick with a system.
And yes, my main priority was to build a system with high consistency in the returns. There are lots of trading systems available that deliver higher CAGRs but with much higher volatility. Personally I don't have the risk profile to stick with these, so I had to invent my own.
I found MF via ChatGPT. I told ChatGPT to go find me a strategy that beats the SPY but also carries less risk. After a few days of searching ChatGPT came back with a single result...TheMarketFighter.
I signed up and am going to give it a try.
Also, thank you for publishing so much helpful and well written information about it. I appreciate the transparency and the obvious love you put into your work.
Keep up the good work!
Thank you.
Thank you so much for sharing this information!
Appreciate it 🙏
I've been reading for a while and find the approach genuinely interesting. Combining sector and factor momentum into a single monthly rotation is a clean idea, and you're more transparent about the mechanics than most people selling signals. Credit for that.
Before I allocate seriously to it, four things I'd like to understand:
1. Benchmark construction — is the MSCI World column price return or net total return? Compounding your annual figures gives roughly 3.8% p.a. over the period, which is about two points below MSCI World's actual net total return. If the strategy column includes dividends and the benchmark doesn't, a meaningful slice of the reported alpha would be a measurement artifact.
2. Live vs. backtested — can you break out performance from the date you started publishing signals, separately from the simulated history? An out-of-sample record is the only part that can't be optimised in hindsight.
3. 2008 — the strategy shows -10.24%. Consumer staples, the strongest global sector that year, fell roughly 15-20%. A long-only rotation between sectors and factors shouldn't be able to beat its own best available holding, so I assume there's a cash or moving-average filter in the model. Could you describe it?
4. Investability pre-2013 — most of the major factor ETFs didn't launch until 2012-13, and several factor indices were backfilled. How much of the 2000-2012 record is simulated on index data rather than tradeable instruments, and does it account for spreads and TERs?
Not trying to be difficult — these are the questions I'd ask any manager, and clear answers would make me more confident rather than less. Thanks.
Thanks for the kind words, I appreciate it!
Let me try to answer your questions:
1. The data I used to build this was based on price returns, so I decided to stick with this for all data to make a fair apples-to-apples comparison. You're right that net totals are higher. This applies to the benchmarks as well as the strategy returns.
2. Fully agree, the out of sample data is the most important part. I displayed a chart for this in my latest article here: https://www.marketfighter.com/p/i-switched-my-strategy-to-usd-and
Performance since I started publishing on Substack in March can be found here:
https://www.marketfighter.com/p/performance-update-august-2026
3. Yes, there's a drawdown protection layer, also based on relative price momentum. I have tested this part on data back to 1871, and it kicked in last time in 2008-2009. It is described in this article.
4. It's not based on specific ETF implementations as they vary across the globe. It's based on the underlying indices and not accounting for spreads, TERs, or taxes, as all of this varies upon execution.
I hope this answered your questions :)
I like how clearly you separate the factor and sector sleeves from the drawdown filter. The move to cash is the part I’d most like to understand: what market momentum rule triggers the exit and re-entry, and how sensitive were the results to that choice? It would also be useful to see the maximum drawdown alongside the annual returns. That would make the “without taking more risk” claim much easier to assess.
Thanks for your comment, that's a great point. I will try to incorporate a table with annual drawdowns in a future post 👍 For now I can say that the largest drawdown of the strategy over the full backtest (back to 2000) was -15.3% in 2020 during Covid. But this number is based on monthly closing prices, so it would of course be deeper with daily figures. The protection layer only plays a role during major declines, and it's based on simple monthly closing prices just like the rest of the strategy.
Thanks for clarifying that the −15.3% drawdown is based on monthly closes. I’d be interested to see the annual drawdowns when you publish them. Daily figures would also help show what investors experienced between month ends.
Really enjoyed reading this because it focused on something many investors overlook: the importance of having a repeatable process. Outperformance rarely comes from chasing every market trend or constantly reacting to noise. More often, it comes from having a clear framework, staying disciplined, and sticking to it through different market cycles.
What stood out to me is that good investing often looks boring from the outside. It is patience, consistency, and sound decision making under uncertainty repeated over time. That is where the real edge usually comes from. Great breakdown.
Well said, and thanks, I appreciate it! 🙏 You’re absolutely right. Patience, discipline, and consistency can take you far, and that combination seems to be underrated.
Thanks for the article. As a fellow computer scientist I loved reading this and your approach. I've downloaded the stock data for the last few years, calculating all the technicals for each stock and running various queries with a view to something similar in an individual sock basis, but your approach is so elegantly simple (appreciate the work to get there wasn't!). Will definitely give this a try.
Thanks, I really appreciate your comment! I've been down that road with individual stocks as well, attempting to build a similar system, but I didn't manage to find the same consistency as I did with the ETF setup. It was too volatile for me :-)
I love this! This is exactly the kind of edge I’ve been looking for 🙏 Subscribed and looking forward to follow!
When exactly are the trades supposed to be executed?
Thank you so much, I really appreciate this!
You should execute the trade after you receive the monthly trade signal in your inbox. Ideally on the first day of the month, but the sooner the better.
The whole back-test and my five years of trading the system in real life have been based on trades executed at the turn of the month or on the first trading day of the month.
Have you ever backtested a short ETF when you go to cash ?
I haven’t, because we don’t have access to these as retail investors in the European Union. But historically, they would most likely have performed better in the crisis years but also introduced more volatility.
Having trouble with subscribing on substack .. please send me a link.
I understand how the dividends are distributed for ETFs. As I understand it, all you return calculations assume the investor fully reinvests all dividends back into the portfolio. What I was asking for is the annual and period returns with the assumption that the investor does NOT reinvest those dividends. This can make a significant difference over the long run. As mentioned above I am looking for income as well as appreciation of principal. Thanks for your quick responses by the way. Looking forward to joining soon.
Thanks for clarifying. Sorry, I don’t have any numbers for the exact US dividends, as I track the strategy with the European ETFs.
But if we assume an average 1.5% annual dividend in total, and the strategy has delivered around 18% CAGR over the past 20 years, the math would be like this:
18% CAGR with dividends reinvested: 2,739%
16.5% CAGR with no reinvested dividends: 2,037%
Be aware that these numbers are based on assumptions. But yes, withdrawing the dividends does reduce long-term compounding, but it gives you the ongoing income stream you're looking for.
Hope this helps, and feel free to send me a DM if you have any further questions :)
Are you able to share the annual returns if the investor did not reinvest the returns back into the portfolio, I.e. non CAGR returns. I am at a point in life where I will need the income vs growing the principal. I am US based so in USD would be appreciated.
Hi Patrick. While the European ETFs are accumulating, the US-listed ones pay out quarterly dividends. Typically they will take up 1-2% of the annual returns, but it depends on the allocations throughout the year. Did this answer your question?
Hi Sir, may I ask, for paid members, do they get the full strategy settings/parameters?
Or we will only get the trading signals?
Thanks in advance
Hi! I don’t give the full recipe with the exact rules, but I provide the output every month with the exact signals I’m trading myself.
Thank you so much for the article. I am always afraid of overfitting. Is your method to avoid drawdown the same as for the momentum signal? Or do you use another? For me less moving parts are always more stable. Best regards!
Totally agree, the simpler the better! And yes, the drawdown mechanism is based on momentum in monthly closing prices just like the rest of the strategy. It’s so easy to overfit a backtest, and I didn’t want to add other parameters for the same reason :)
with the paid subscription do i get the actual strategy? i.e. the rules i can backtest on my own, or do i only get the monthly signals?
The exact rules behind the strategy are proprietary, but the paid subscription provides access to the monthly trading signals covering every trade I execute myself, based on the output of the system.
Quick clarification on the drawdown protection piece — is the move-to-cash decision only evaluated once a month, at the same time as the momentum signal, or do you also monitor and act on the drawdown filter mid-month if conditions shift before the next signal is due? Just want to make sure I'm executing the mechanics correctly.
It’s only evaluated as part of the monthly signal. This is to reduce the risk of false alarms and maintain the simplicity of only having to deal with the strategy once a month.
Great article and inspiring, Question ? I’m an Australian investor , prefer to use ETFs domiciled in AU , tax reasons etc what support do you provide if we join ?
Thank you! Tax is one thing I can’t help with, as rules differ so much from one country to another. I always recommend reaching out to a local tax advisor.
Sure, ETFs domiciled in AU can be just as good as those in US or Europe, as long as they track the same or similar indices.
Was there anything else you needed support for? :)
Thks for responded so quickly , just considering the subscription,concerned if we use the domiciled ETFs can’t really compare with yours , filters ,or indices you mentioned not sure
The majority of my subscribers actually live outside of EU, meaning their returns are different and affected by currency fluctuations. They will never see the exact same numbers as me. What matters is finding ETFs that track the right (or similar) indices as those listed here: https://www.marketfighter.com/p/the-watchlist-see-the-15-tickers