If we have a long term expected return on the sp of about 8-10%(?) Shouldnt we increase the 19 years to counter the year of the simulated drawdown? Or am I just proving my misunderstanding 🙃
I'd agree with you because our basis for that expectation is historical returns which include the drawdowns. What's interesting is if you look at 12 month rolling returns using monthly data from https://www.officialdata.org/us/stocks/s-p-500/1900 less than 1% of the of the returns are >40% loss
The math in this post gives a better intuition for the impact of a tail move (as opposed to just the volatility drag) on a return stream which should be of concern to anyone for whom path matters (and I really don't believe anyone can put all their wealth in a box and emotionally ignore path).
But to address you directly, I agree that there's an apples to orange comparison issue in the way I framed it. I reran it with the drawdown occurring 1/100 instead of 1/20 and a single 50% drawdown knocks only .70% off the CAGR which makes putting the money in a box wise if you can do it...but again the counterbalance is how does the tail risk scale as you increase the observation periods...we've had 10-20% sell-offs in one day a couple of times in the past 40 years (ie an impossible number of standard devs...so tails are fatter for increases in sampling and that's what an investor will experience)
I appreciate you commenting, I had been thinking about this since I published and now getting around to tinkering with the sheet again.
Thank you for taking the time to answer, now I understand the post even better. I hate path, it really messes with my performance. :) As an engineer I usually don't include emotions in my calculations so naturally I aim for highest expected return... Then start doing strange things while path happens: selling during setbacks and adding leverage during runs...
And thank you for writing all together. This sub might be the most interesting text my phone is showing me.
As I'm messing with this a handy way to remember it would if you were earning 8% for 199 years and hit a 50% drawdown in 1 of the years then your 200 year CAGR is 7.5% (so shaves .50% off the total)
If that drawdown happened 1 in 100 years it would shave 1% off the total CAGR
Hey there! Just wanted to share that I recently managed to log in to my Substack account after a good 2 years – totally forgot I had it! On another note, I've been thinking about investment strategies, and it seems like going all in on %100 SPY might be more practical for folks in the U.S. due to custodial risks in offshore investments. It's just something to consider. Also, I've noticed that discussions on Twitter and American social media platforms sometimes miss the broader global perspective. There's a lot happening around the world that might not get the attention it deserves. Just my two cents – thanks for letting me share!
One of my favorite posts you've written
If we have a long term expected return on the sp of about 8-10%(?) Shouldnt we increase the 19 years to counter the year of the simulated drawdown? Or am I just proving my misunderstanding 🙃
Have a nice day!
I'd agree with you because our basis for that expectation is historical returns which include the drawdowns. What's interesting is if you look at 12 month rolling returns using monthly data from https://www.officialdata.org/us/stocks/s-p-500/1900 less than 1% of the of the returns are >40% loss
The math in this post gives a better intuition for the impact of a tail move (as opposed to just the volatility drag) on a return stream which should be of concern to anyone for whom path matters (and I really don't believe anyone can put all their wealth in a box and emotionally ignore path).
But to address you directly, I agree that there's an apples to orange comparison issue in the way I framed it. I reran it with the drawdown occurring 1/100 instead of 1/20 and a single 50% drawdown knocks only .70% off the CAGR which makes putting the money in a box wise if you can do it...but again the counterbalance is how does the tail risk scale as you increase the observation periods...we've had 10-20% sell-offs in one day a couple of times in the past 40 years (ie an impossible number of standard devs...so tails are fatter for increases in sampling and that's what an investor will experience)
I appreciate you commenting, I had been thinking about this since I published and now getting around to tinkering with the sheet again.
Thank you for taking the time to answer, now I understand the post even better. I hate path, it really messes with my performance. :) As an engineer I usually don't include emotions in my calculations so naturally I aim for highest expected return... Then start doing strange things while path happens: selling during setbacks and adding leverage during runs...
And thank you for writing all together. This sub might be the most interesting text my phone is showing me.
As I'm messing with this a handy way to remember it would if you were earning 8% for 199 years and hit a 50% drawdown in 1 of the years then your 200 year CAGR is 7.5% (so shaves .50% off the total)
If that drawdown happened 1 in 100 years it would shave 1% off the total CAGR
Hey there! Just wanted to share that I recently managed to log in to my Substack account after a good 2 years – totally forgot I had it! On another note, I've been thinking about investment strategies, and it seems like going all in on %100 SPY might be more practical for folks in the U.S. due to custodial risks in offshore investments. It's just something to consider. Also, I've noticed that discussions on Twitter and American social media platforms sometimes miss the broader global perspective. There's a lot happening around the world that might not get the attention it deserves. Just my two cents – thanks for letting me share!