Thank you for sharing, there is a lot of information in the comments, great. Your latest article mentions a sell signal, HYG seems to be a little bit away, but HYG has also fallen a lot recently.
It is disappointing that the high-yield bond system did not create a sell signal in time. HYG is less than 1% away from generating a sell. But apparently, in unusual times, SPY can slide down faster than bonds.
On the positive side, I think the system's robustness is indicated by the fact that it also works with other bond ETFs, with similar results (mostly not as good as HYG, but sometimes better). And other bond ETFs are also close to a "sell": JNK: only 0.4% over sell signal. FLOT: 1%. USIG: 0.5%. AGG: 1%. LQD: 0.3%.
I think the one danger of this strategy is that HYG does not drop when SPY drops. Thus keeping you in a bad trade.
There are times when SPY get very overvalued, especially with AI and tech dominating the index. For example, witness 16 July to 15 August 2024. HYG barely moved downward when SPY moved down from 558 to a low of 508. If you look at the 100 EMA for HYG at this time, it was still far below HYG.
So caution is warranted entering this strategy when HYG is considerably above it's EMA and perhaps there should be a sell rules added when SPY gets overextended - for example the RSI14 of SPY goes above 70.
just as an anecdotal side-note, the strategy has a similar (yet slightly weaker) performance when one uses JNK instead of HYG. I take that an an indicator of robustness.
Oddly though, when one enters JNK:HYG on StockCharts, there are some weird and violent divergences in March 2020 and in late 2008. I guess discovery is difficult when things are burning.
The question is whether this is a flaw, or a feature. My premise is that it's a feature: I like signals that separate noise from substance. HYG is calm unless things get really worrisome -- like in 2000, 2007, 2011, 2015, 2018, 2020 and 2022.
Following this logic, it was positive that HYG barely moved downward in 2024. You were better off not selling! (Unless you had a perfect entry signal for getting back into the market, which I don't).
But that's just my premise! It can be useful, or flawed. I certainly can't guarantee whether HYG will sniff out the next crisis! And anyway, my time to pull the ripcord needn't be identical to yours. A proper trader would have been in and out of the market several times in 2024, but doing such things is above my pay grade.
BTAL is a narrowly focused, equity-only, rules-based strategy betting on a specific stock characteristic (beta), while managed futures is a broader, futures-driven, often trend-chasing approach managed actively across diverse markets. BTAL’s like a precision tool for equity volatility arbitrage; managed futures is more like a Swiss Army knife for global market trends.
Many managed futures funds struggled during market downturns despite their design to thrive in such conditions because they often rely heavily on trend-following strategies that need sustained, identifiable price trends to profit—conditions that don’t always materialize in choppy or abrupt crashes. Rapid, trendless volatility (like in 2020’s COVID drop) or crowded positioning among CTAs can erode gains, while high leverage and transaction costs amplify losses when trades misfire. Unlike BTAL’s focus on a specific equity factor, their broader, futures-based approach can falter if global markets lack clear direction or if trends reverse unexpectedly.
Nusbaum is a useful source of information and analysis, thank you!
I'm not so sure whether managed futures are easily comparable to BTAL, but if they are, this is worrisome. Roger uses RYMFX and AQMIX, and while they share similar performance with BTAL from 2012 onwards (if not quite so good), things got quite bad in 2008 and 2011.
Hence, please beware using BTAL as a cure for volatility in all market situations. We don't know how it will perform in the next bear market -- if the managed future ones are an indication, it won't be good enough to mitigate a deflationary recession.
As previously posted: currently, I have no skin in the game with BTAL, and hold the usual reservations against a strategy that does not go to cash in seriously bad times. Be cautious, please.
Another fine article, Martin. Thanks so much for sharing your excellent work.
Agreed that high yield is a good indicator. Personally I use ‘BofA US High Yield Index…’ in StockCharts - $$HYIOAS. Another market indicator I’m using is ‘NYSE Percent of Stocks Above 200 Day Moving Average’ - $NYA200R. Take a look at it during the period of late ‘07-late ‘09. It took off like a rocket when the market bottomed out in March ‘09. There’s a couple of others I like if you’re interested.
thank you very much, Vance! Always great to get your fee-back.
Yes, I am interested in the other indicators you could share.
I actually use $$HYIOAS myself. What I find intriguiing -- and I have put into my toolbox -- is when you compare this one to treasuries, such as the short-term IEI. So, in Stockcharts: IEI:$$HYIOAS , which is an excellent indicator of market risk appetite. That said, it couldn't make up its mind in 2018, was not very useful in 2015, and was too late for 2020. Still good, though!
I also like $NYA200R, thank you! This one seems to be good for identifying market bottoms, right? Inasmuch as anything below 15 can well be a bottom, unless it isn't, and continues on down to 10 (2018), 4 (2020) or gosh, 2.13 (2008). Or how do you use it? One more approach I could imagine using would be to watch out for a $NYA200R thrust, in other words a strong reversal.
one more thing I should add is that I try to be particularly careful before I actually invest in a strategy that does not go to cash in bad times. As pretty as BTAL looks, we don't know how it will perform in a deflationary bear market, in a quick '87ish crash, or in some godforsaken new kind of crisis. We *do* know what cash does in such circumstances, which is why Taleb calls it the ultimate hedge.
One question - how does this signal fit in with the strategies that you are employing? Do you add it to your Benign Neglect strategies or do you run as an additional strategy? Something that would be helpful for a future post is to share how you manage all of these strategies. How automated (or manual) is your process? Thanks again. Steve
Good question! My process is: I find something I find investable. I write about it. I try to integrate comments and suggestions from my readers. I do some more thinking. I then possibly add it to my portfolio of strategies.
From March 1 onwards, I have added the Bitcoin as a Signal strategy, about which I wrote in January, to my portfolio of strategies, with a single weighting. HYG and BTAL: not there yet. We'll see!
Martin, Excellent post, as always. I am a fan of $BTAL; it is a personal holding. To me, I like the negative correlation to the SPY. In my opinion, it is best used as a 10% portfolio diversifier. There is a good blogger, Roger Nusbaum, who writes about $BTAL and managed futures and other non-correlated return streams. He rails against bonds as he believes they no longer provide diversification; they are too correlated to stocks. I differ from him on bonds, but I appreciate his work on uncorrelated return streams. Here is a post specifically about BTAL, you will notice he writes about it regularly. https://rogersplanning.blogspot.com/2024/04/the-power-of-owning-something-that.html
Hi, Martin. Appreciate the shout-out and another great find! Here is a QuantMage adaptation: https://quantmage.app/grimoire/5f8a6a274ab72ee051604db21a18f507 I find that checking the condition for two consecutive days makes it more robust and also reduces the number of trades (from 122 to 88) for the period.
I think it's closely related to the high yield spreads (so called "the best macro indicator" https://www.1nve.st/p/the-best-macro-indicator-round-two). HYG trending up basically means its yields going down, which in turn means a tight spread, which is a bullish signal.
thank you very much, JJ! I suspected the "two consecutive days" rule would be an improvement, but Portfoliovisualizer doesn't have that option. Your QuantMage software is certainly superior in that respect. PV better look out!
HYG is certainly related to hi-yield spreads, inasmuch as the market calculates risk similarly to insurance. For me of course, using HYG as a signal is easier, since PV hasn't integrated such macro indicators and mostly has only ETFs and funds in its database.
Have you tried using similar parameters as mine (including BTAL) for your high yield spread algorithm? If so, are the results pretty close?
Yeah, I was quite surprised by the fact that BTAL is a much better hedge in this case, compared to bond / dollar / gold. I briefly tried replacing risk-off assets in some of my high-yield spreads based strategies with BTAL, but it made the perf worse.
It's worth noting that BTAL can be quite volatile sometimes. For instance, it had multiple months where it recorded a high single-digit or even a double-digit loss.
BTAL has good liquidity. According to MarketWatch the average volume is 469,440 shares.
Today's bid/ask is $18.20 X $18.22
BTAL's CAGR is -1.68 and has a 35.4% MDD which is more than 10% worse than SPY! It is therefore not an ETF held by the common retail investor, but rather is traded by smart money that uses this ETF for hedging and in conjunction with other ETF's.
The only thing I don't like about using moving average strategies with PV, or strategies that use a stop loss is that you have to check the PV signal every single day : (
Perhaps email PV to suggest they send an alert email when a signal fires up. If enough users do so, then maybe they will do something about it!
I think the results and logic of the SPY/BTAL strategy speak for themselves! Look at the yearly “annual returns" bar charts; 9 out of 13 years the strategy has beat SPY, and overall the chart looks pretty good.
In the “drawdowns for the moving average model” section the longest time of “underwater period” was only 10 months, followed by 8 months then it drops to 4 months etc. That’s impressive!
Can someone please talk me off the ledge, because I’m about to start trading this strategy!
On another note, I never used the EMA in Portfolio Visualizer, but rather used a simple moving average. Now I see that I need to pay more attention to the EMA and consider it for my next round of testing. Nice work Martin!
CTA is quite new, hence can only be backtested to 2023. BTAL is bad enough in respect to its only short history but at least we can determine it was useful in both deflationary bull and inflationary bear markets.
Thank you for sharing, there is a lot of information in the comments, great. Your latest article mentions a sell signal, HYG seems to be a little bit away, but HYG has also fallen a lot recently.
thank you, Tianshen!
It is disappointing that the high-yield bond system did not create a sell signal in time. HYG is less than 1% away from generating a sell. But apparently, in unusual times, SPY can slide down faster than bonds.
On the positive side, I think the system's robustness is indicated by the fact that it also works with other bond ETFs, with similar results (mostly not as good as HYG, but sometimes better). And other bond ETFs are also close to a "sell": JNK: only 0.4% over sell signal. FLOT: 1%. USIG: 0.5%. AGG: 1%. LQD: 0.3%.
I think the one danger of this strategy is that HYG does not drop when SPY drops. Thus keeping you in a bad trade.
There are times when SPY get very overvalued, especially with AI and tech dominating the index. For example, witness 16 July to 15 August 2024. HYG barely moved downward when SPY moved down from 558 to a low of 508. If you look at the 100 EMA for HYG at this time, it was still far below HYG.
So caution is warranted entering this strategy when HYG is considerably above it's EMA and perhaps there should be a sell rules added when SPY gets overextended - for example the RSI14 of SPY goes above 70.
just as an anecdotal side-note, the strategy has a similar (yet slightly weaker) performance when one uses JNK instead of HYG. I take that an an indicator of robustness.
Oddly though, when one enters JNK:HYG on StockCharts, there are some weird and violent divergences in March 2020 and in late 2008. I guess discovery is difficult when things are burning.
good comment, Simon. Thanks!
The question is whether this is a flaw, or a feature. My premise is that it's a feature: I like signals that separate noise from substance. HYG is calm unless things get really worrisome -- like in 2000, 2007, 2011, 2015, 2018, 2020 and 2022.
Following this logic, it was positive that HYG barely moved downward in 2024. You were better off not selling! (Unless you had a perfect entry signal for getting back into the market, which I don't).
But that's just my premise! It can be useful, or flawed. I certainly can't guarantee whether HYG will sniff out the next crisis! And anyway, my time to pull the ripcord needn't be identical to yours. A proper trader would have been in and out of the market several times in 2024, but doing such things is above my pay grade.
BTAL is a narrowly focused, equity-only, rules-based strategy betting on a specific stock characteristic (beta), while managed futures is a broader, futures-driven, often trend-chasing approach managed actively across diverse markets. BTAL’s like a precision tool for equity volatility arbitrage; managed futures is more like a Swiss Army knife for global market trends.
Many managed futures funds struggled during market downturns despite their design to thrive in such conditions because they often rely heavily on trend-following strategies that need sustained, identifiable price trends to profit—conditions that don’t always materialize in choppy or abrupt crashes. Rapid, trendless volatility (like in 2020’s COVID drop) or crowded positioning among CTAs can erode gains, while high leverage and transaction costs amplify losses when trades misfire. Unlike BTAL’s focus on a specific equity factor, their broader, futures-based approach can falter if global markets lack clear direction or if trends reverse unexpectedly.
excellent comment Simon, much appreciated!
Are you aware of any fund that has similar characteristics to BTAL, yet has a longer trading history?
I asked Grok.
BTAL seems unique.
The others have manager discretion or use future.
Making them unpredictable.
good to know!
One more reason to own BTAL is its nice yield. According to StockCharts, 5.52% forward; Stockanalysis.com: 3.42%.
Nusbaum is a useful source of information and analysis, thank you!
I'm not so sure whether managed futures are easily comparable to BTAL, but if they are, this is worrisome. Roger uses RYMFX and AQMIX, and while they share similar performance with BTAL from 2012 onwards (if not quite so good), things got quite bad in 2008 and 2011.
Hence, please beware using BTAL as a cure for volatility in all market situations. We don't know how it will perform in the next bear market -- if the managed future ones are an indication, it won't be good enough to mitigate a deflationary recession.
As previously posted: currently, I have no skin in the game with BTAL, and hold the usual reservations against a strategy that does not go to cash in seriously bad times. Be cautious, please.
Another fine article, Martin. Thanks so much for sharing your excellent work.
Agreed that high yield is a good indicator. Personally I use ‘BofA US High Yield Index…’ in StockCharts - $$HYIOAS. Another market indicator I’m using is ‘NYSE Percent of Stocks Above 200 Day Moving Average’ - $NYA200R. Take a look at it during the period of late ‘07-late ‘09. It took off like a rocket when the market bottomed out in March ‘09. There’s a couple of others I like if you’re interested.
thank you very much, Vance! Always great to get your fee-back.
Yes, I am interested in the other indicators you could share.
I actually use $$HYIOAS myself. What I find intriguiing -- and I have put into my toolbox -- is when you compare this one to treasuries, such as the short-term IEI. So, in Stockcharts: IEI:$$HYIOAS , which is an excellent indicator of market risk appetite. That said, it couldn't make up its mind in 2018, was not very useful in 2015, and was too late for 2020. Still good, though!
I also like $NYA200R, thank you! This one seems to be good for identifying market bottoms, right? Inasmuch as anything below 15 can well be a bottom, unless it isn't, and continues on down to 10 (2018), 4 (2020) or gosh, 2.13 (2008). Or how do you use it? One more approach I could imagine using would be to watch out for a $NYA200R thrust, in other words a strong reversal.
one more thing I should add is that I try to be particularly careful before I actually invest in a strategy that does not go to cash in bad times. As pretty as BTAL looks, we don't know how it will perform in a deflationary bear market, in a quick '87ish crash, or in some godforsaken new kind of crisis. We *do* know what cash does in such circumstances, which is why Taleb calls it the ultimate hedge.
One question - how does this signal fit in with the strategies that you are employing? Do you add it to your Benign Neglect strategies or do you run as an additional strategy? Something that would be helpful for a future post is to share how you manage all of these strategies. How automated (or manual) is your process? Thanks again. Steve
Good question! My process is: I find something I find investable. I write about it. I try to integrate comments and suggestions from my readers. I do some more thinking. I then possibly add it to my portfolio of strategies.
From March 1 onwards, I have added the Bitcoin as a Signal strategy, about which I wrote in January, to my portfolio of strategies, with a single weighting. HYG and BTAL: not there yet. We'll see!
Martin, Excellent post, as always. I am a fan of $BTAL; it is a personal holding. To me, I like the negative correlation to the SPY. In my opinion, it is best used as a 10% portfolio diversifier. There is a good blogger, Roger Nusbaum, who writes about $BTAL and managed futures and other non-correlated return streams. He rails against bonds as he believes they no longer provide diversification; they are too correlated to stocks. I differ from him on bonds, but I appreciate his work on uncorrelated return streams. Here is a post specifically about BTAL, you will notice he writes about it regularly. https://rogersplanning.blogspot.com/2024/04/the-power-of-owning-something-that.html
Thanks Steve! I'll look into what Nusbaum writes and get back to you.
Hi, Martin. Appreciate the shout-out and another great find! Here is a QuantMage adaptation: https://quantmage.app/grimoire/5f8a6a274ab72ee051604db21a18f507 I find that checking the condition for two consecutive days makes it more robust and also reduces the number of trades (from 122 to 88) for the period.
I think it's closely related to the high yield spreads (so called "the best macro indicator" https://www.1nve.st/p/the-best-macro-indicator-round-two). HYG trending up basically means its yields going down, which in turn means a tight spread, which is a bullish signal.
thank you very much, JJ! I suspected the "two consecutive days" rule would be an improvement, but Portfoliovisualizer doesn't have that option. Your QuantMage software is certainly superior in that respect. PV better look out!
HYG is certainly related to hi-yield spreads, inasmuch as the market calculates risk similarly to insurance. For me of course, using HYG as a signal is easier, since PV hasn't integrated such macro indicators and mostly has only ETFs and funds in its database.
Have you tried using similar parameters as mine (including BTAL) for your high yield spread algorithm? If so, are the results pretty close?
Yeah, I was quite surprised by the fact that BTAL is a much better hedge in this case, compared to bond / dollar / gold. I briefly tried replacing risk-off assets in some of my high-yield spreads based strategies with BTAL, but it made the perf worse.
It's worth noting that BTAL can be quite volatile sometimes. For instance, it had multiple months where it recorded a high single-digit or even a double-digit loss.
BTAL has good liquidity. According to MarketWatch the average volume is 469,440 shares.
Today's bid/ask is $18.20 X $18.22
BTAL's CAGR is -1.68 and has a 35.4% MDD which is more than 10% worse than SPY! It is therefore not an ETF held by the common retail investor, but rather is traded by smart money that uses this ETF for hedging and in conjunction with other ETF's.
Exactly.
I've used HYG and VWEHX before as timing tools and they worked well for me!
good to know, Bill!
The only thing I don't like about using moving average strategies with PV, or strategies that use a stop loss is that you have to check the PV signal every single day : (
Perhaps email PV to suggest they send an alert email when a signal fires up. If enough users do so, then maybe they will do something about it!
Good idea, will do.
(Of course, the weekly and monthly strategies are less of a hassle).
(If I ever turn this Substack into a paid subscription newsletter, it will include email action alerts).
I think the results and logic of the SPY/BTAL strategy speak for themselves! Look at the yearly “annual returns" bar charts; 9 out of 13 years the strategy has beat SPY, and overall the chart looks pretty good.
In the “drawdowns for the moving average model” section the longest time of “underwater period” was only 10 months, followed by 8 months then it drops to 4 months etc. That’s impressive!
Can someone please talk me off the ledge, because I’m about to start trading this strategy!
On another note, I never used the EMA in Portfolio Visualizer, but rather used a simple moving average. Now I see that I need to pay more attention to the EMA and consider it for my next round of testing. Nice work Martin!
Thank you, Bill! Your opinion has considerable weight and is all the more appreciated.
Please check using CTA (up 8% YTD) instead of BTAL. Thanks
CTA is quite new, hence can only be backtested to 2023. BTAL is bad enough in respect to its only short history but at least we can determine it was useful in both deflationary bull and inflationary bear markets.
HYG is up about 1% last 1 month means signal risk on now? Thanks for interesting article.
yessir!