apologies for being a bit outspoken. You can backtest what you want, but if you don’t take into account market regimes you get misled and you mislead (yes, inadvertently). All of your backtest time is one where disinflation prevailed with one exception (2022), and guess what? By picking assets that perform well under disinflationary conditions, and leveraging one of them a lot, you do well. You can’t generalise like that, that’s why backtests cannot be used to prove that something works, but only that in that sample path nothing ugly came up (yes, that’s all you can say). In 2022 the strategy suffered as inflation spiked. While in all other periods Treasuries were an exceptional asset class as the Fed splurged on monetary easing with repeated QEs, till there was QE4 with Covid. For the same reason the Nasdaq, the asset you want to leverage so much, outperformed being a long-duration asset. Can’t you see how unique those conditions were? Your backtest tells you only this: if long-duration assets will do well, i can do well with this simple model. And it is not even a rotation strategy, as you are always invested in long-duration assets. Rotating from red into pink is not much of a rotation. Wait and see when inflation comes back. You’ll tell me what happens. Can’t you see that dollar-centric assets have started to underperform for secular reasons, and that your strategy is basically an overweight on US assets that perform well when the dollar rises? Plot Nasdaq/MSCI World xUS against real trade-weighted US dollar. When the world flocks to Nasdaq and Treasuries it has to buy dollars to do that. Hence, dollar (real trade-weighted) and Nasdaq relative to other global markets tend to move together. Your bet is a huge one on disinflation and perpetual US exceptionalism. US exceptionalism is done with, as fiscal and monetary capacity are gone (too much debt and future stagflation with tariffs), and the stimulus impulse is shifting to Europe and China. Trump has upended US exceptionalism with tariffs, and does not want a strong dollar. Markets in the future will not be as US -centric, and your backtest takes you straight into that and you’ll be headed for heavy losses once the regime changes. Under the new regime you’ll have to consider inflation hedges (gold, commodities) and shorter-duration assets. At least you will have to be careful to use Nasdaq leverage only when disinflation is there. How do you measure it in real time? With an oil-to-gold ratio. Yes, now we have a short spell of disinflation according to that ratio … You get the gist. This comment is already way too long
I am simply sharing this idea with it’s performance. As with all trading and especially systematic you MUST trade multiple strategies. With this post I wanted to show the idea and foundation about a simple combination of two ETFs and what they could achieve, in reality a 50/50 between QQQ and BTAL might be a better choice but it’s the same idea.
You can replace BTAL to multiple other similar ETFs or even futures trading ETFs and also replace TQQQ with the non-leveraged or even VTI or similar, it’s the same idea and that is what I want to share.
With that said, I don’t trade this and I don’t recommend anyone to trade this but I found the idea interesting.
I understand. Mine was a suggestion as well. But great performance may let others think it can be generalized in the future. The oil-to-gold ratio (a rate of change thereof) could be considered as a filter to avoid wrong regimes
This is a great representation of the strength of BTAL. The problem is the short history. This 17 year period is an exceptionally strong market. We don't really know what it will do in an extended down market, as occurred in the 30's-40s, the 70's or even 2000-2008.
Your comments on why annual rebalancing works for this are spot on!
That is for sure the weakness of this strategy, data only goes back to 2012, but we do have 2018, 2020 and 2022 with weak periods and the strategy have proven to perform well in these periods.
Nasdaq index was in a hole for that period. QQQ is a good proxy for Nasdaq. So with just BTAL and QQQ we don't really know the behavior over an extended period of down behavior. I really like BTAL but I don't think it covers that long, or that deep, a bear.
so your concern then is that the general concept of being Long in low-beta stocks (less sensitive to market movements) and being Short in high-beta stocks (more sensitive to market movements) will not hold up in long bear markets?
Well, that's not exactly what I said. In response though, I don't think staying short for 15 years works. How would that have worked in Japan for the last 20 years?
My point was simply that we don't have a long enough history to test BTAL/QQQ or TQQQ through a full range likely market conditions including long periods of stock market underperformance.
apologies for being a bit outspoken. You can backtest what you want, but if you don’t take into account market regimes you get misled and you mislead (yes, inadvertently). All of your backtest time is one where disinflation prevailed with one exception (2022), and guess what? By picking assets that perform well under disinflationary conditions, and leveraging one of them a lot, you do well. You can’t generalise like that, that’s why backtests cannot be used to prove that something works, but only that in that sample path nothing ugly came up (yes, that’s all you can say). In 2022 the strategy suffered as inflation spiked. While in all other periods Treasuries were an exceptional asset class as the Fed splurged on monetary easing with repeated QEs, till there was QE4 with Covid. For the same reason the Nasdaq, the asset you want to leverage so much, outperformed being a long-duration asset. Can’t you see how unique those conditions were? Your backtest tells you only this: if long-duration assets will do well, i can do well with this simple model. And it is not even a rotation strategy, as you are always invested in long-duration assets. Rotating from red into pink is not much of a rotation. Wait and see when inflation comes back. You’ll tell me what happens. Can’t you see that dollar-centric assets have started to underperform for secular reasons, and that your strategy is basically an overweight on US assets that perform well when the dollar rises? Plot Nasdaq/MSCI World xUS against real trade-weighted US dollar. When the world flocks to Nasdaq and Treasuries it has to buy dollars to do that. Hence, dollar (real trade-weighted) and Nasdaq relative to other global markets tend to move together. Your bet is a huge one on disinflation and perpetual US exceptionalism. US exceptionalism is done with, as fiscal and monetary capacity are gone (too much debt and future stagflation with tariffs), and the stimulus impulse is shifting to Europe and China. Trump has upended US exceptionalism with tariffs, and does not want a strong dollar. Markets in the future will not be as US -centric, and your backtest takes you straight into that and you’ll be headed for heavy losses once the regime changes. Under the new regime you’ll have to consider inflation hedges (gold, commodities) and shorter-duration assets. At least you will have to be careful to use Nasdaq leverage only when disinflation is there. How do you measure it in real time? With an oil-to-gold ratio. Yes, now we have a short spell of disinflation according to that ratio … You get the gist. This comment is already way too long
I am simply sharing this idea with it’s performance. As with all trading and especially systematic you MUST trade multiple strategies. With this post I wanted to show the idea and foundation about a simple combination of two ETFs and what they could achieve, in reality a 50/50 between QQQ and BTAL might be a better choice but it’s the same idea.
You can replace BTAL to multiple other similar ETFs or even futures trading ETFs and also replace TQQQ with the non-leveraged or even VTI or similar, it’s the same idea and that is what I want to share.
With that said, I don’t trade this and I don’t recommend anyone to trade this but I found the idea interesting.
I understand. Mine was a suggestion as well. But great performance may let others think it can be generalized in the future. The oil-to-gold ratio (a rate of change thereof) could be considered as a filter to avoid wrong regimes
This is a great representation of the strength of BTAL. The problem is the short history. This 17 year period is an exceptionally strong market. We don't really know what it will do in an extended down market, as occurred in the 30's-40s, the 70's or even 2000-2008.
Your comments on why annual rebalancing works for this are spot on!
That is for sure the weakness of this strategy, data only goes back to 2012, but we do have 2018, 2020 and 2022 with weak periods and the strategy have proven to perform well in these periods.
What is your main concern for the period of 2000-2008. That the negative correlation will not hold up?
Nasdaq index was in a hole for that period. QQQ is a good proxy for Nasdaq. So with just BTAL and QQQ we don't really know the behavior over an extended period of down behavior. I really like BTAL but I don't think it covers that long, or that deep, a bear.
so your concern then is that the general concept of being Long in low-beta stocks (less sensitive to market movements) and being Short in high-beta stocks (more sensitive to market movements) will not hold up in long bear markets?
Well, that's not exactly what I said. In response though, I don't think staying short for 15 years works. How would that have worked in Japan for the last 20 years?
My point was simply that we don't have a long enough history to test BTAL/QQQ or TQQQ through a full range likely market conditions including long periods of stock market underperformance.
no one made money in japan
What would it look like if this strategy was initiated after a sell-off? Now is a good time to find out. Thanks for the heads-up!