HEDGEFUNDIE's excellent adventure Part II: The next journey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

lawyeredCLO wrote: Thu Feb 16, 2023 1:32 pm
hiddenpower wrote: Thu Feb 16, 2023 12:03 pm
lawyeredCLO wrote: Thu Feb 16, 2023 7:59 am With treasury yields where they are today and S&P 500 valuations where they are, this strategy should probably shift to 50% UPRO/50% TMF or 45% UPRO/55% TMF again.
Why? Rates could just stay flat and TMF could then decay. UPRO is the growth driver.
Because valuations of the s&p 500 are over 20 price earning ratio. Meanwhile long-term treasuries valuations are at 10 year lows and highest yields
It was over 20 when the strategy was started too.
Lawyered_
Posts: 176
Joined: Wed Feb 09, 2022 10:52 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Lawyered_ »

hiddenpower wrote: Thu Feb 16, 2023 1:43 pm
lawyeredCLO wrote: Thu Feb 16, 2023 1:32 pm
hiddenpower wrote: Thu Feb 16, 2023 12:03 pm
lawyeredCLO wrote: Thu Feb 16, 2023 7:59 am With treasury yields where they are today and S&P 500 valuations where they are, this strategy should probably shift to 50% UPRO/50% TMF or 45% UPRO/55% TMF again.
Why? Rates could just stay flat and TMF could then decay. UPRO is the growth driver.
Because valuations of the s&p 500 are over 20 price earning ratio. Meanwhile long-term treasuries valuations are at 10 year lows and highest yields
It was over 20 when the strategy was started too.
And when it started it was 44/55, correct?
bgf
Posts: 2085
Joined: Fri Nov 10, 2017 8:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by bgf »

lawyeredCLO wrote: Thu Feb 16, 2023 2:43 pm
hiddenpower wrote: Thu Feb 16, 2023 1:43 pm
lawyeredCLO wrote: Thu Feb 16, 2023 1:32 pm
hiddenpower wrote: Thu Feb 16, 2023 12:03 pm
lawyeredCLO wrote: Thu Feb 16, 2023 7:59 am With treasury yields where they are today and S&P 500 valuations where they are, this strategy should probably shift to 50% UPRO/50% TMF or 45% UPRO/55% TMF again.
Why? Rates could just stay flat and TMF could then decay. UPRO is the growth driver.
Because valuations of the s&p 500 are over 20 price earning ratio. Meanwhile long-term treasuries valuations are at 10 year lows and highest yields
It was over 20 when the strategy was started too.
And when it started it was 44/55, correct?
I don’t see how current PE could be of any timing help with this strategy.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"
Lawyered_
Posts: 176
Joined: Wed Feb 09, 2022 10:52 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Lawyered_ »

bgf wrote: Thu Feb 16, 2023 5:42 pm
lawyeredCLO wrote: Thu Feb 16, 2023 2:43 pm
hiddenpower wrote: Thu Feb 16, 2023 1:43 pm
lawyeredCLO wrote: Thu Feb 16, 2023 1:32 pm
hiddenpower wrote: Thu Feb 16, 2023 12:03 pm

Why? Rates could just stay flat and TMF could then decay. UPRO is the growth driver.
Because valuations of the s&p 500 are over 20 price earning ratio. Meanwhile long-term treasuries valuations are at 10 year lows and highest yields
It was over 20 when the strategy was started too.
And when it started it was 44/55, correct?
I don’t see how current PE could be of any timing help with this strategy.
Certainly you can imagine that you should not be leveraging an asset that is overvalued already, correct? Stocks are more likely to be overvalued when they have a PE of over 20 versus 15 for example. Of course we can take PE into account when analyzing the proper amount of equities to be holding when leveraged.
bgf
Posts: 2085
Joined: Fri Nov 10, 2017 8:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by bgf »

lawyeredCLO wrote: Thu Feb 16, 2023 8:16 pm
bgf wrote: Thu Feb 16, 2023 5:42 pm
lawyeredCLO wrote: Thu Feb 16, 2023 2:43 pm
hiddenpower wrote: Thu Feb 16, 2023 1:43 pm
lawyeredCLO wrote: Thu Feb 16, 2023 1:32 pm

Because valuations of the s&p 500 are over 20 price earning ratio. Meanwhile long-term treasuries valuations are at 10 year lows and highest yields
It was over 20 when the strategy was started too.
And when it started it was 44/55, correct?
I don’t see how current PE could be of any timing help with this strategy.
Certainly you can imagine that you should not be leveraging an asset that is overvalued already, correct? Stocks are more likely to be overvalued when they have a PE of over 20 versus 15 for example. Of course we can take PE into account when analyzing the proper amount of equities to be holding when leveraged.
This strategy is so path dependent that it’s almost not worth discussing, but my thought was that it’s just as likely to signal strong performance than reversal. At 20 PE people are willing to pay more for company earnings, perhaps due to low rates, expected growth, strong past performance, momentum, etc. all of which would signal likely outperformance by HFEA.

Sitting around on cash as a value investor is tough enough in unlevered simple stock portfolios… underperformance can last over a decade.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"
Lawyered_
Posts: 176
Joined: Wed Feb 09, 2022 10:52 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Lawyered_ »

bgf wrote: Sat Feb 18, 2023 9:04 am
lawyeredCLO wrote: Thu Feb 16, 2023 8:16 pm
bgf wrote: Thu Feb 16, 2023 5:42 pm
lawyeredCLO wrote: Thu Feb 16, 2023 2:43 pm
hiddenpower wrote: Thu Feb 16, 2023 1:43 pm

It was over 20 when the strategy was started too.
And when it started it was 44/55, correct?
I don’t see how current PE could be of any timing help with this strategy.
Certainly you can imagine that you should not be leveraging an asset that is overvalued already, correct? Stocks are more likely to be overvalued when they have a PE of over 20 versus 15 for example. Of course we can take PE into account when analyzing the proper amount of equities to be holding when leveraged.
This strategy is so path dependent that it’s almost not worth discussing, but my thought was that it’s just as likely to signal strong performance than reversal. At 20 PE people are willing to pay more for company earnings, perhaps due to low rates, expected growth, strong past performance, momentum, etc. all of which would signal likely outperformance by HFEA.

Sitting around on cash as a value investor is tough enough in unlevered simple stock portfolios… underperformance can last over a decade.
I agree sitting around on cash as a value investor is tough enough, it should be deployed for 1x value investors no doubt ASAP. What I am advocating, however, is that path dependency is so important for 3x investors (and underlying valuation strongly drives path) that perhaps we should be cognizant of the underlying positions and valuations of S&P 500 and LTTs in this strategy. Certainly 23+ PE ratio and 1% LTT rates, in retrospect, were a horrible time to get into this strategy in January 2022. Perhaps we can apply that lesson to the current equity valuations? For example, it may not be crazy to go 50/50 UPRO/TMF at this juncture, or 25% MIDU; 25% UPRO; 50% TMF, or 45/55 etc... while taking into account valuations of equities vs LTTs.
User avatar
OohLaLa
Posts: 459
Joined: Tue Mar 09, 2021 7:26 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by OohLaLa »

lawyeredCLO wrote: Mon Feb 20, 2023 2:08 pm I agree sitting around on cash as a value investor is tough enough, it should be deployed for 1x value investors no doubt ASAP. What I am advocating, however, is that path dependency is so important for 3x investors (and underlying valuation strongly drives path) that perhaps we should be cognizant of the underlying positions and valuations of S&P 500 and LTTs in this strategy. Certainly 23+ PE ratio and 1% LTT rates, in retrospect, were a horrible time to get into this strategy in January 2022. Perhaps we can apply that lesson to the current equity valuations? For example, it may not be crazy to go 50/50 UPRO/TMF at this juncture, or 25% MIDU; 25% UPRO; 50% TMF, or 45/55 etc... while taking into account valuations of equities vs LTTs.
***EDIT***
I misunderstood this: I had the bad recollection, for some reason, that the classic HFEA AA was different. It was 55/45 and you mention going 50/50 or 45/55. My bad.
***
Why would someone go lower on LTTs at this point, when we've reached a 4% rate on 20y? I understand your general point (to adjust to the environmental factors), but shouldn't that have been done when the rate was significantly lower?

Just to confirm, I don't support the general argument of timing the AA to market conditions. I would rather suggest people to set their AA in response to their risk tolerance, their age/ human capital, net worth VS. target(s) etc.
moneyflowin
Posts: 287
Joined: Sat Oct 16, 2021 6:14 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by moneyflowin »

What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Marseille07 »

moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I'm not a big fan of QQQ because you'd be concentrated in tech. Also EDV's duration is extremely long.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.

If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
Lawyered_
Posts: 176
Joined: Wed Feb 09, 2022 10:52 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Lawyered_ »

skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.

If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
I like the idea of toning down HFEA, not bucketing the strategy, and running it across a person's entire holdings. IMO, a solid toned down leveraged portfolio that you could run forever would be something like 12.5% UPRO; 27.5% NTSX; 40% NTSI; 20% NTSE. 116.25% stocks; 52.5% ITTs; rebalance quarterly. Gives you a bit of leverage on the equities, some treasuries, and international exposure in a capital efficient way. Draw downs should be MUCH smaller than HFEA.
moneyflowin
Posts: 287
Joined: Sat Oct 16, 2021 6:14 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by moneyflowin »

skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.
No, I meant QQQ. NDX isn't investable unless one wants to use options or NQ futures, but QQQ is superior for b&h over those.

UPRO and TMF have even worse Sharpes than QQQ and EDV but that doesn't stop people from using them. QQQ has a superior Sharpe/Sortino than any mix of LETF/ETF that produces the equivalent leverage
If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
Per PV, your 30/20/50 portfolio has a lower CAGR, Sharpe and Sortinos than HFEA-mini
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

lawyeredCLO wrote: Tue Feb 21, 2023 8:45 am
skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.

If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
I like the idea of toning down HFEA, not bucketing the strategy, and running it across a person's entire holdings. IMO, a solid toned down leveraged portfolio that you could run forever would be something like 12.5% UPRO; 27.5% NTSX; 40% NTSI; 20% NTSE. 116.25% stocks; 52.5% ITTs; rebalance quarterly. Gives you a bit of leverage on the equities, some treasuries, and international exposure in a capital efficient way. Draw downs should be MUCH smaller than HFEA.
I definitely would want the international too I was just being lazy last night. I think that's pretty good. I'm still a little torn on whether the ER of NTSX is worth it. I personally don't have high expectations for leveraged bonds. The term premium by some estimates is below zero. So for the benefit of 60% leveraged bonds which may or may not have positive term premium, you're paying .2% on the whole investment. Those bonds are costing you .32% per dollar of bond. That's likely higher than their expected return. For NTSI you are paying. 5% per dollar of underlying bond exposure. That's almost certainly higher than the expected return of those bonds.

If you have access to TYA we can construct the exact same portfolio with lower cost.

15/20/25/40 UPRO/TYA/VTI/VXUS comes out to 110/60 AA. The total ER is .20%.

The NTSX implementation you suggested has an ER of .35%.

The only concern is if TYA is liquid enough for rebalancing (I don't know).
Lawyered_
Posts: 176
Joined: Wed Feb 09, 2022 10:52 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Lawyered_ »

skierincolorado wrote: Tue Feb 21, 2023 10:27 am
lawyeredCLO wrote: Tue Feb 21, 2023 8:45 am
skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.

If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
I like the idea of toning down HFEA, not bucketing the strategy, and running it across a person's entire holdings. IMO, a solid toned down leveraged portfolio that you could run forever would be something like 12.5% UPRO; 27.5% NTSX; 40% NTSI; 20% NTSE. 116.25% stocks; 52.5% ITTs; rebalance quarterly. Gives you a bit of leverage on the equities, some treasuries, and international exposure in a capital efficient way. Draw downs should be MUCH smaller than HFEA.
I definitely would want the international too I was just being lazy last night. I think that's pretty good. I'm still a little torn on whether the ER of NTSX is worth it. I personally don't have high expectations for leveraged bonds. The term premium by some estimates is below zero. So for the benefit of 60% leveraged bonds which may or may not have positive term premium, you're paying .2% on the whole investment. Those bonds are costing you .32% per dollar of bond. That's likely higher than their expected return. For NTSI you are paying. 5% per dollar of underlying bond exposure. That's almost certainly higher than the expected return of those bonds.

If you have access to TYA we can construct the exact same portfolio with lower cost.

15/20/25/40 UPRO/TYA/VTI/VXUS comes out to 110/60 AA. The total ER is .20%.

The NTSX implementation you suggested has an ER of .35%.

The only concern is if TYA is liquid enough for rebalancing (I don't know).
How do you calculate what the bonds are "costing" per dollar of bonds in the NTS products? Also, if you are going to use TYA, why not just use ZROZ rather than the 3x ITTs?
Last edited by Lawyered_ on Tue Feb 21, 2023 10:32 am, edited 1 time in total.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

moneyflowin wrote: Tue Feb 21, 2023 10:08 am
skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.
No, I meant QQQ. NDX isn't investable unless one wants to use options or NQ futures, but QQQ is superior for b&h over those.

UPRO and TMF have even worse Sharpes than QQQ and EDV but that doesn't stop people from using them. QQQ has a superior Sharpe/Sortino than any mix of LETF/ETF that produces the equivalent leverage
If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
Per PV, your 30/20/50 portfolio has a lower CAGR, Sharpe and Sortinos than HFEA-mini
With PV you can only go back to 2010. With simba you can go back to 1955 and the sharpe is much higher.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

lawyeredCLO wrote: Tue Feb 21, 2023 10:31 am
skierincolorado wrote: Tue Feb 21, 2023 10:27 am
lawyeredCLO wrote: Tue Feb 21, 2023 8:45 am
skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.

If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
I like the idea of toning down HFEA, not bucketing the strategy, and running it across a person's entire holdings. IMO, a solid toned down leveraged portfolio that you could run forever would be something like 12.5% UPRO; 27.5% NTSX; 40% NTSI; 20% NTSE. 116.25% stocks; 52.5% ITTs; rebalance quarterly. Gives you a bit of leverage on the equities, some treasuries, and international exposure in a capital efficient way. Draw downs should be MUCH smaller than HFEA.
I definitely would want the international too I was just being lazy last night. I think that's pretty good. I'm still a little torn on whether the ER of NTSX is worth it. I personally don't have high expectations for leveraged bonds. The term premium by some estimates is below zero. So for the benefit of 60% leveraged bonds which may or may not have positive term premium, you're paying .2% on the whole investment. Those bonds are costing you .32% per dollar of bond. That's likely higher than their expected return. For NTSI you are paying. 5% per dollar of underlying bond exposure. That's almost certainly higher than the expected return of those bonds.

If you have access to TYA we can construct the exact same portfolio with lower cost.

15/20/25/40 UPRO/TYA/VTI/VXUS comes out to 110/60 AA. The total ER is .20%.

The NTSX implementation you suggested has an ER of .35%.

The only concern is if TYA is liquid enough for rebalancing (I don't know).
How do you calculate what the bonds are "costing" per dollar of bonds in the NTS products? Also, if you are going to use TYA, why not just use ZROZ rather than the 3x ITTs?
Since you can get sp500 exposure for free, the cost of NTSX is for the benefit of the bond portion. For every $1 of ntsx you pay .02 in fees. That gets you .60 of bond exposure. Each dollar of bond exposure cost you .033 (.02/.6). It's unlikely that leveraged bonds will return more than that, giving you negative expected value. It's even worse for NTSE.

That's all sort of irrelevant in the end. In the end you can get the same AA using TYA for lower ER. I'm not sure which would be more tax efficient in taxable. I think people usually overestimate the tax drag and the tax drag occurs only if the market does very well. The ER occurs even if the market does poorly, increasing the variance of your final outcome.

I prefer leveraged ITT to unleveraged LTT. Historically they have much higher sharpe and there are theoretical reasons why this would remain the case. Remember when you buy $100 of TYA you are getting $100 of unleveraged ITT exposure plus $200 of leveraged ITT exposure. The curve is usually pretty flat between ITT and LTT so you aren't getting any extra return for buying LTT. You're only getting extra risk.
er999
Posts: 1343
Joined: Wed Nov 05, 2008 10:00 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by er999 »

12 month return UPRO -25%, VOO (s&p 500) -3%, more than 3x the losses. Volatility decay in action — hopefully will work in reverse if there’s a recovery before my position is ground down to zero!
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

er999 wrote: Mon Mar 06, 2023 9:21 am 12 month return UPRO -25%, VOO (s&p 500) -3%, more than 3x the losses. Volatility decay in action — hopefully will work in reverse if there’s a recovery before my position is ground down to zero!
TBF wouldn't it be worse if using futures and you happened to rebalance exactly on the bottoms? What we need is a loan :mrgreen:
er999
Posts: 1343
Joined: Wed Nov 05, 2008 10:00 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by er999 »

hiddenpower wrote: Mon Mar 06, 2023 9:27 am

TBF wouldn't it be worse if using futures and you happened to rebalance exactly on the bottoms? What we need is a loan :mrgreen:
Right, it really shows the benefit of low cost loans. I paid off early my 2.75% mortgage in 2018, started HFEA 12/2021 (with about 8% of my portfolio) and it would have been better to keep the mortgage and invest the difference rather than paying it off early and using extreme leverage for a small part of the portfolio.
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

er999 wrote: Mon Mar 06, 2023 10:30 am
hiddenpower wrote: Mon Mar 06, 2023 9:27 am

TBF wouldn't it be worse if using futures and you happened to rebalance exactly on the bottoms? What we need is a loan :mrgreen:
Right, it really shows the benefit of low cost loans. I paid off early my 2.75% mortgage in 2018, started HFEA 12/2021 (with about 8% of my portfolio) and it would have been better to keep the mortgage and invest the difference rather than paying it off early and using extreme leverage for a small part of the portfolio.
However UPRO is such bang for your buck compared to going heavier with SSO.

What about buying LEAPS at a fixed strike in the event you need to roll and it wasn't breached? That way you're only paying for time interest.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Marseille07 »

hiddenpower wrote: Mon Mar 06, 2023 1:51 pm However UPRO is such bang for your buck compared to going heavier with SSO.
I don't think this is substantiated. If you backtested this, be sure the backtest included 2008 GFC.
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

Marseille07 wrote: Mon Mar 06, 2023 6:59 pm
hiddenpower wrote: Mon Mar 06, 2023 1:51 pm However UPRO is such bang for your buck compared to going heavier with SSO.
I don't think this is substantiated. If you backtested this, be sure the backtest included 2008 GFC.
Not quite following can you elaborate what’s not substantiated? even if you buy more SSO at the start of the lottery ticket, it won’t behave the same
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Marseille07 »

hiddenpower wrote: Mon Mar 06, 2023 7:32 pm Not quite following can you elaborate what’s not substantiated? even if you buy more SSO at the start of the lottery ticket, it won’t behave the same
Based on my backtest, SSO's long-term CAGR was higher than UPRO. Higher leverage isn't necessarily optimal due to beta slippage.
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

Marseille07 wrote: Mon Mar 06, 2023 8:00 pm
hiddenpower wrote: Mon Mar 06, 2023 7:32 pm Not quite following can you elaborate what’s not substantiated? even if you buy more SSO at the start of the lottery ticket, it won’t behave the same
Based on my backtest, SSO's long-term CAGR was higher than UPRO. Higher leverage isn't necessarily optimal due to beta slippage.
Can you share a few over different timeframes? From what I remember UPRO/TMF typically crushed PSLDX.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Marseille07 »

hiddenpower wrote: Mon Mar 06, 2023 8:37 pm Can you share a few over different timeframes? From what I remember UPRO/TMF typically crushed PSLDX.
Never mind, I think I misinterpreted what you were saying. It seems like you were talking about leverage construction.

If you want to increase leverage then UPRO needs to be used, as SSO takes up more of your portfolio space with lower leverage.
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

Marseille07 wrote: Mon Mar 06, 2023 9:12 pm
hiddenpower wrote: Mon Mar 06, 2023 8:37 pm Can you share a few over different timeframes? From what I remember UPRO/TMF typically crushed PSLDX.
Never mind, I think I misinterpreted what you were saying. It seems like you were talking about leverage construction.

If you want to increase leverage then UPRO needs to be used, as SSO takes up more of your portfolio space with lower leverage.
Imagine leveraging the initial deposit to SSO/TYA more than 1x to be say $150k at the start for example whereas HFEA starts at $100k. Same initial notional value
corpgator
Posts: 217
Joined: Wed Jul 22, 2015 12:00 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by corpgator »

Finally getting some bounce from TMF when UPRO goes down. Still not nearly enough to recover all the losses, but at least it's not broken entirely.
User avatar
firebirdparts
Posts: 4411
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by firebirdparts »

I don't expect to see a day like that one again. Instead of "finally" I was thinking "never again"
This time is the same
User avatar
firebirdparts
Posts: 4411
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by firebirdparts »

hiddenpower wrote: Mon Mar 06, 2023 1:51 pm
However UPRO is such bang for your buck compared to going heavier with SSO.

What about buying LEAPS at a fixed strike in the event you need to roll and it wasn't breached? That way you're only paying for time interest.
"What about" is a pretty open-ended question. It's sort of traditional to look at January, so if you did this (and I did) January of 2023 was an interesting moment where you lost half your money and the leverage is now 5 or 6 instead of 3. So you have to decide whether or not to stay at a leverage of 5 or 6 or go back to 3. You can do either one.
This time is the same
stimulacra
Posts: 1006
Joined: Wed Dec 21, 2016 2:50 pm
Location: Houston

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by stimulacra »

Logged into my M1 account today… ouch.

Overdue for a rebalance… but feel like I'll be locking it in when I do.

How often are people rebalancing nowadays?

How many people are staying the course?

The initial amount was all play money so I'm a bit dispassionate about it… but in terms of new inflows there seems to be plenty of squeaky wheels in my portfolio this year…
Hydromod
Posts: 1051
Joined: Tue Mar 26, 2019 10:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Hydromod »

stimulacra wrote: Sat Mar 11, 2023 11:11 am Logged into my M1 account today… ouch.

Overdue for a rebalance… but feel like I'll be locking it in when I do.

How often are people rebalancing nowadays?

How many people are staying the course?

The initial amount was all play money so I'm a bit dispassionate about it… but in terms of new inflows there seems to be plenty of squeaky wheels in my portfolio this year…
Staying but with a different approach that's more balanced.
bgf
Posts: 2085
Joined: Fri Nov 10, 2017 8:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by bgf »

skierincolorado wrote: Tue Feb 21, 2023 10:27 am
lawyeredCLO wrote: Tue Feb 21, 2023 8:45 am
skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.

If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
I like the idea of toning down HFEA, not bucketing the strategy, and running it across a person's entire holdings. IMO, a solid toned down leveraged portfolio that you could run forever would be something like 12.5% UPRO; 27.5% NTSX; 40% NTSI; 20% NTSE. 116.25% stocks; 52.5% ITTs; rebalance quarterly. Gives you a bit of leverage on the equities, some treasuries, and international exposure in a capital efficient way. Draw downs should be MUCH smaller than HFEA.
I definitely would want the international too I was just being lazy last night. I think that's pretty good. I'm still a little torn on whether the ER of NTSX is worth it. I personally don't have high expectations for leveraged bonds. The term premium by some estimates is below zero. So for the benefit of 60% leveraged bonds which may or may not have positive term premium, you're paying .2% on the whole investment. Those bonds are costing you .32% per dollar of bond. That's likely higher than their expected return. For NTSI you are paying. 5% per dollar of underlying bond exposure. That's almost certainly higher than the expected return of those bonds.

If you have access to TYA we can construct the exact same portfolio with lower cost.

15/20/25/40 UPRO/TYA/VTI/VXUS comes out to 110/60 AA. The total ER is .20%.

The NTSX implementation you suggested has an ER of .35%.

The only concern is if TYA is liquid enough for rebalancing (I don't know).
You shouldn’t assume a constant term premium over the life of a long term investment. If you choose to time your investment in NTSX then you’re stuck with the problem of timing interest rates, which is impossible to do.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

bgf wrote: Sat Mar 11, 2023 2:24 pm
skierincolorado wrote: Tue Feb 21, 2023 10:27 am
lawyeredCLO wrote: Tue Feb 21, 2023 8:45 am
skierincolorado wrote: Mon Feb 20, 2023 11:03 pm
moneyflowin wrote: Mon Feb 20, 2023 8:51 pm What do people think of this idea (I call it HFEA-mini):

Using daily percentage change data for NDX and SPY dating back to 1993, I did linear regression and found that NDX moves 1.21% for every 1% daily move of SPX, with an R-squared of 0.69. From 2015 to present, it moves 1.15x with R-squared of 0.87.

With that in mind, a portfolio of 70/30 QQQ/EDV should behave like 1.3x 60/40 but without the risk of ruin or margin calls and without the borrowing cost of LETFs. This portfolio backtests in PV very well. [Use PEDIX in place of EDV for the backtest because it has more historical data.] I've compared this to other forms of 1.3x leverage such as using SSO+SPY, and this beats them.

This would be suitable for people who aren't comfortable with 3:1 leverage but still want something more aggressive than 60/40.
I assume you meant NDX not QQQ. I don't like it becaues NDX is tech concentrated and has poor sharpe historically. EDV has horrible sharpe.

If you want a toned back version of HFEA, something like 30/20/50 UPRO/VTI/VGIT would be good. Or 10/90 UPRO/NTSX
I like the idea of toning down HFEA, not bucketing the strategy, and running it across a person's entire holdings. IMO, a solid toned down leveraged portfolio that you could run forever would be something like 12.5% UPRO; 27.5% NTSX; 40% NTSI; 20% NTSE. 116.25% stocks; 52.5% ITTs; rebalance quarterly. Gives you a bit of leverage on the equities, some treasuries, and international exposure in a capital efficient way. Draw downs should be MUCH smaller than HFEA.
I definitely would want the international too I was just being lazy last night. I think that's pretty good. I'm still a little torn on whether the ER of NTSX is worth it. I personally don't have high expectations for leveraged bonds. The term premium by some estimates is below zero. So for the benefit of 60% leveraged bonds which may or may not have positive term premium, you're paying .2% on the whole investment. Those bonds are costing you .32% per dollar of bond. That's likely higher than their expected return. For NTSI you are paying. 5% per dollar of underlying bond exposure. That's almost certainly higher than the expected return of those bonds.

If you have access to TYA we can construct the exact same portfolio with lower cost.

15/20/25/40 UPRO/TYA/VTI/VXUS comes out to 110/60 AA. The total ER is .20%.

The NTSX implementation you suggested has an ER of .35%.

The only concern is if TYA is liquid enough for rebalancing (I don't know).
You shouldn’t assume a constant term premium over the life of a long term investment. If you choose to time your investment in NTSX then you’re stuck with the problem of timing interest rates, which is impossible to do.
Term premium has been low for decades and plenty of theoretical reasons to believe it will stay low or negative. I wouldn't be ok with a higher cost vehicle like ntsx just based on the hope that the returns will be so good that the costs are immaterial. Not suggesting market timing investments. Choose a vehicle where the costs are likely to be worth it or don't lever bonds at all.
User avatar
OohLaLa
Posts: 459
Joined: Tue Mar 09, 2021 7:26 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by OohLaLa »

stimulacra wrote: Sat Mar 11, 2023 11:11 am Logged into my M1 account today… ouch.

Overdue for a rebalance… but feel like I'll be locking it in when I do.

How often are people rebalancing nowadays?

How many people are staying the course?

The initial amount was all play money so I'm a bit dispassionate about it… but in terms of new inflows there seems to be plenty of squeaky wheels in my portfolio this year…
I am still in, with target 35% TQQQ, 25% UPRO, 15% TMF, 15% TYD, 10% VXZ. Keep contributing when I can. Rebalancing every quarter, so next one will take place come April. Historically-speaking, it's definitely a bad idea to just ride it out without rebalancing.
NewEnglandVolley
Posts: 6
Joined: Sun Mar 15, 2020 7:37 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by NewEnglandVolley »

https://finance.yahoo.com/news/silicon- ... 35121.html

Does SVB going insolvent prompt the Fed to alter it's timelines on interest rate increases. Powell signals that interest rates are likely to be higher than anticipated, whatever that means, but there's no doubt SVB and likely a lot of other large banks were heavily invested into long term treasuries during 2020/2021 when rates were quite low. Yes, their own management messed up but how pervasive will the issue with longer term maturity assets causing current insolvency in banking be? That's what drove the 10%+ TMF move on Friday and so far another 5%+ move today. It looks like the market may believe the Fed will change their tune soon. I'd certainly welcome it as I've been keeping up with re-balancing UPRO to TMF and have lots of upside with TMF moves.
User avatar
OohLaLa
Posts: 459
Joined: Tue Mar 09, 2021 7:26 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by OohLaLa »

NewEnglandVolley wrote: Mon Mar 13, 2023 10:53 am https://finance.yahoo.com/news/silicon- ... 35121.html

Does SVB going insolvent prompt the Fed to alter it's timelines on interest rate increases. Powell signals that interest rates are likely to be higher than anticipated, whatever that means, but there's no doubt SVB and likely a lot of other large banks were heavily invested into long term treasuries during 2020/2021 when rates were quite low. Yes, their own management messed up but how pervasive will the issue with longer term maturity assets causing current insolvency in banking be? That's what drove the 10%+ TMF move on Friday and so far another 5%+ move today. It looks like the market may believe the Fed will change their tune soon. I'd certainly welcome it as I've been keeping up with re-balancing UPRO to TMF and have lots of upside with TMF moves.
I am not convinced the treasuries rally is simply due to a belief the Fed will lower rate increases, or halt them, because of SVB. I am seeing defensive sectors going up (utilities, healthcare, consumer staples), as well. This might be a general concern about the future state of the economy and stock market.

Does anybody have a way of tracking analyst predictions on rate hikes? That would help answer this question.
bgf
Posts: 2085
Joined: Fri Nov 10, 2017 8:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by bgf »

Apparently there was a LOT of bearish positioning on long duration treasuries and walking that back may account for some of the big move.

I still think this fund will allow the fed to increase rates while sidestepping the risks to regional banks. Ultimately I don’t think these bank failures affect the path of rates.

I rebalanced a little of TMF into UPRO this morning after the big pop.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"
NewEnglandVolley
Posts: 6
Joined: Sun Mar 15, 2020 7:37 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by NewEnglandVolley »

OohLaLa wrote: Mon Mar 13, 2023 12:29 pm
NewEnglandVolley wrote: Mon Mar 13, 2023 10:53 am https://finance.yahoo.com/news/silicon- ... 35121.html

Does SVB going insolvent prompt the Fed to alter it's timelines on interest rate increases. Powell signals that interest rates are likely to be higher than anticipated, whatever that means, but there's no doubt SVB and likely a lot of other large banks were heavily invested into long term treasuries during 2020/2021 when rates were quite low. Yes, their own management messed up but how pervasive will the issue with longer term maturity assets causing current insolvency in banking be? That's what drove the 10%+ TMF move on Friday and so far another 5%+ move today. It looks like the market may believe the Fed will change their tune soon. I'd certainly welcome it as I've been keeping up with re-balancing UPRO to TMF and have lots of upside with TMF moves.
I am not convinced the treasuries rally is simply due to a belief the Fed will lower rate increases, or halt them, because of SVB. I am seeing defensive sectors going up (utilities, healthcare, consumer staples), as well. This might be a general concern about the future state of the economy and stock market.

Does anybody have a way of tracking analyst predictions on rate hikes? That would help answer this question.
Don't have any specific tracker but various articles seem to show the market expecting a .5% increase now expecting a .25%
https://www.cnbc.com/2023/03/13/somethi ... hikes.html

Of course after my post today TMF ends on less than 1% increase. We'll see how this may shake out as I doubt we'll know how much banks are relying on bonds to boost balance sheets. If it's actually a systemic problem, well... no quicker way to end inflation than to freeze withdrawals / payroll from large swaths of your population.
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

firebirdparts wrote: Sat Mar 11, 2023 6:41 am
hiddenpower wrote: Mon Mar 06, 2023 1:51 pm
However UPRO is such bang for your buck compared to going heavier with SSO.

What about buying LEAPS at a fixed strike in the event you need to roll and it wasn't breached? That way you're only paying for time interest.
"What about" is a pretty open-ended question. It's sort of traditional to look at January, so if you did this (and I did) January of 2023 was an interesting moment where you lost half your money and the leverage is now 5 or 6 instead of 3. So you have to decide whether or not to stay at a leverage of 5 or 6 or go back to 3. You can do either one.
How did you lose so much since opening a LEAPs position in January? We are basically flat since then. Is this a position opened from long ago? I'd stay. Chance of going bust though.
User avatar
firebirdparts
Posts: 4411
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by firebirdparts »

I'm sorry, I'm not trying to confuse anybody. I thought everybody in HFEA had lost half their money from the peak, but that's not what I intended to flag as important. Just ignore that comment.

The important thing is that with LEAPS you aren't adjusting leverage every day, so there's no volatility decay at all. Instead, you have theta decay. So once a year, let's say, you have a decision to make about how much volatility decay you're going to accept.

Again, 'what about' is a really open-ended question and I was just throwing out the most interesting thing I could think of.
This time is the same
User avatar
hiddenpower
Posts: 564
Joined: Tue Nov 17, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hiddenpower »

firebirdparts wrote: Wed Mar 15, 2023 12:36 pm I'm sorry, I'm not trying to confuse anybody. I thought everybody in HFEA had lost half their money from the peak, but that's not what I intended to flag as important. Just ignore that comment.

The important thing is that with LEAPS you aren't adjusting leverage every day, so there's no volatility decay at all. Instead, you have theta decay. So once a year, let's say, you have a decision to make about how much volatility decay you're going to accept.

Again, 'what about' is a really open-ended question and I was just throwing out the most interesting thing I could think of.
I'd love to see how to do this strategy with LEAPS and the likelihood of going bust. Maybe it's historically better than HFEA with LETFs. That's what I was getting at.
User avatar
firebirdparts
Posts: 4411
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by firebirdparts »

Right. We would have to predict so much in this case to choose between these two. You could backtest, but option price history is a little bit painful to get. I'm sure somebody has a big pile of data somewhere.

Just a basic look at it, let's say today that SPY is about 390 and so a 3X leverage would be a strike price of 260. Time value of that option today, expiring March 15 2024, is $10 a share. So we know for sure we're losing that in a year. This is an inferior position because the leap is price only, where UPRO is based on total return. In a down market, the LEAP leverage is increasing, and in an up market, it's decreasing, so you don't have a very stable situation. In a 33% downturn, you would lose all the equity portion, but in HFEA that wouldn't be too bad. In an up market, the releveraging every day of UPRO is a really nice feature.
This time is the same
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

firebirdparts wrote: Thu Mar 16, 2023 9:26 am Right. We would have to predict so much in this case to choose between these two. You could backtest, but option price history is a little bit painful to get. I'm sure somebody has a big pile of data somewhere.

Just a basic look at it, let's say today that SPY is about 390 and so a 3X leverage would be a strike price of 260. Time value of that option today, expiring March 15 2024, is $10 a share. So we know for sure we're losing that in a year. This is an inferior position because the leap is price only, where UPRO is based on total return. In a down market, the LEAP leverage is increasing, and in an up market, it's decreasing, so you don't have a very stable situation. In a 33% downturn, you would lose all the equity portion, but in HFEA that wouldn't be too bad. In an up market, the releveraging every day of UPRO is a really nice feature.
Dividends are implicit in LEAPs and all options. They are total return just like UPRO, not price only. Otherwise arbitrage opportunity would exist.
Ramjet
Posts: 1464
Joined: Thu Feb 06, 2020 10:45 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Ramjet »

Everyone, please pull out your crystal balls and humor me

1) What do your foresee happening with the economy/interest rates
2) How does the above affect this strategy
3) What would have to happen to convince you that this strategy is essentially dead
Last edited by Ramjet on Thu Mar 16, 2023 11:36 am, edited 2 times in total.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

Ramjet wrote: Thu Mar 16, 2023 11:32 am Everyone, please pull out your crystal balls and humor me

1) What do your foresee happening with the economy/interest rates
2) How does the above affect this strategy
3) What would have to happen to convince you that this strategy is essentially dead
No idea.

Diversification and leverage will never be dead. But the strategy was always a bit overleveraged and bond heavy. And it ignores lifecycle investing.
Ramjet
Posts: 1464
Joined: Thu Feb 06, 2020 10:45 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Ramjet »

skierincolorado wrote: Thu Mar 16, 2023 11:35 am
Ramjet wrote: Thu Mar 16, 2023 11:32 am Everyone, please pull out your crystal balls and humor me

1) What do your foresee happening with the economy/interest rates
2) How does the above affect this strategy
3) What would have to happen to convince you that this strategy is essentially dead
No idea.

Diversification and leverage will never be dead. But the strategy was always a bit overleveraged and bond heavy. And it ignores lifecycle investing.
Not even a guess?

Could be fun.
er999
Posts: 1343
Joined: Wed Nov 05, 2008 10:00 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by er999 »

Ramjet wrote: Thu Mar 16, 2023 11:32 am Everyone, please pull out your crystal balls and humor me

1) What do your foresee happening with the economy/interest rates
2) How does the above affect this strategy
3) What would have to happen to convince you that this strategy is essentially dead
I think the strategy will continue to succeed in the future, it's just whether those who bought in 2021 or earlier (like myself) will survive enough to see any of the future returns. I think buying long term bonds now is way better than in 2020 or 2021. I'm already at a 50-60% drawdown, at some point the drawdown will be so great hard to recover. I don't think the strategy can last 30 years, though, so maybe have an exit point once you've made a certain amount (or at least take some profits out of the strategy).

Interestingly if you started 2/2019 with 55% upro / 45% tmf (when first posted on here although at a different ratio) 100% VTI would be at 1.5x but the 55/45 UPRO/TMF would be at 1.03x. HFEA would have gotten to 2.7x at its peak so perhaps need to look at this strategy like gambling on individual stocks -- when it doubles sell half.

I also think that lump sum isn't as good as monthly contributions as leverage magnifies the start date sensitivity.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

er999 wrote: Thu Mar 16, 2023 1:48 pm
Ramjet wrote: Thu Mar 16, 2023 11:32 am Everyone, please pull out your crystal balls and humor me

1) What do your foresee happening with the economy/interest rates
2) How does the above affect this strategy
3) What would have to happen to convince you that this strategy is essentially dead
I think the strategy will continue to succeed in the future, it's just whether those who bought in 2021 or earlier (like myself) will survive enough to see any of the future returns. I think buying long term bonds now is way better than in 2020 or 2021. I'm already at a 50-60% drawdown, at some point the drawdown will be so great hard to recover. I don't think the strategy can last 30 years, though, so maybe have an exit point once you've made a certain amount (or at least take some profits out of the strategy).

Interestingly if you started 2/2019 with 55% upro / 45% tmf (when first posted on here although at a different ratio) 100% VTI would be at 1.5x but the 55/45 UPRO/TMF would be at 1.03x. HFEA would have gotten to 2.7x at its peak so perhaps need to look at this strategy like gambling on individual stocks -- when it doubles sell half.

I also think that lump sum isn't as good as monthly contributions as leverage magnifies the start date sensitivity.
Gambling is bad. Need a strategy that's better than gambling.
er999
Posts: 1343
Joined: Wed Nov 05, 2008 10:00 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by er999 »

skierincolorado wrote: Thu Mar 16, 2023 2:29 pm
Gambling is bad. Need a strategy that's better than gambling.
We already have one — that’s VTI or VT. Leverage will always be a gamble (unless special situations like you have a fixed rate mortgage below the current treasury rate).
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by skierincolorado »

er999 wrote: Thu Mar 16, 2023 2:44 pm
skierincolorado wrote: Thu Mar 16, 2023 2:29 pm
Gambling is bad. Need a strategy that's better than gambling.
We already have one — that’s VTI or VT. Leverage will always be a gamble (unless special situations like you have a fixed rate mortgage below the current treasury rate).
Leverage isn't gambling. 1x leverage is arbitrary. For some 1.2x makes more sense. For others 0.7x. Even at 1x leverage you have the opportunity cost of risk free gains on stock. For every $1 I invest I must choose between stock and risk free return on cash. That's true at 0.7x leverage, 1x leverage, and 1.2x leverage.

Extremely high leverage past the Kelly Criterion is gambling.
Post Reply