Imperfect market timing still works -- if you stick with it

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
HomerJ
Posts: 12785
Joined: Fri Jun 06, 2008 12:50 pm

Re: Imperfect market timing still works -- if you stick with it

Post by HomerJ » Sun May 12, 2019 4:28 pm

The thing is, the historical 10% long-term return from the stock market?

That INCLUDES the crashes.

It's really not that necessary to avoid the crashes.

Sure, if you're close to retirement or just started retirement, sequence of returns risk matters... But you can easily fix that with a conservative stock/bond asset allocation.

Going in and out might make you more. But it might make you less... Staying the course with an appropriate allocation seems easier and safer to me. And so far, you still get rich... why shoot for more?
Last edited by HomerJ on Sun May 12, 2019 4:41 pm, edited 1 time in total.
The J stands for Jay

Forester
Posts: 79
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Imperfect market timing still works -- if you stick with it

Post by Forester » Sun May 12, 2019 4:31 pm

Relative strength market timing beat a 50-50 US-foreign allocation by 2% annually after 2010.

Timing https://www.portfoliovisualizer.com/tes ... 0&total1=0

Buy & hold https://www.portfoliovisualizer.com/bac ... 0&total3=0

User avatar
HomerJ
Posts: 12785
Joined: Fri Jun 06, 2008 12:50 pm

Re: Imperfect market timing still works -- if you stick with it

Post by HomerJ » Sun May 12, 2019 4:42 pm

Forester wrote:
Sun May 12, 2019 4:31 pm
Relative strength market timing beat a 50-50 US-foreign allocation by 2% annually after 2010.

Timing https://www.portfoliovisualizer.com/tes ... 0&total1=0

Buy & hold https://www.portfoliovisualizer.com/bac ... 0&total3=0
Your first link shows the Buy and Hold S&P 500 guy winning by 2%... Was that a typo?

But that doesn't prove much.. We haven't had a huge crash since 2010... A couple of whipsaws in 2011, 2015, and 2018 though.
The J stands for Jay

Forester
Posts: 79
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Imperfect market timing still works -- if you stick with it

Post by Forester » Sun May 12, 2019 4:47 pm

HomerJ wrote:
Sun May 12, 2019 4:42 pm
Forester wrote:
Sun May 12, 2019 4:31 pm
Relative strength market timing beat a 50-50 US-foreign allocation by 2% annually after 2010.

Timing https://www.portfoliovisualizer.com/tes ... 0&total1=0

Buy & hold https://www.portfoliovisualizer.com/bac ... 0&total3=0
Your first link shows the Buy and Hold S&P 500 guy winning by 2%... Was that a typo?

But that doesn't prove much.. We haven't had a huge crash since 2010... A couple of whipsaws in 2011, 2015, and 2018 though.
Yes but the point was that a dumb system solved the issue of "how much do I allocate to foreign stocks". It got much of the S&P return, while foreign stocks went sideways.

I'm not necessarily saying it's a good idea :)

User avatar
tadamsmar
Posts: 8241
Joined: Mon May 07, 2007 12:33 pm

Re: Imperfect market timing still works -- if you stick with it

Post by tadamsmar » Sun May 12, 2019 8:16 pm

The article cited in the OP sets a standard: "you never want to lose 20% of your money". Then he sets a standard for back-testing against that standard: The last 19 years and no more than that (the last 2 plus +20% downturns). This stuff is not worthy of serious consideration.

Also, he uses holding S&P 500 as a benchmark. But the mainline Boglehead method involves rebalancing. If market behavior potentiates trend-following to beat buy and hold then it is likely that it potentiates rebalancing to beat buy and hold. He uses a strawman for his benchmark!

Anyway, his stuff is very unconvincing.

But maybe a clearer-thinking advocate of trend-following can avoid such blunders and show that trend-following provide superior returns against a realistic benchmark when used in tax-deferred or Roth accounts (not in taxable).

I am very skeptical that +20% losses can be avoided because I see no evidence that trend-following works to avoid the 1987 crash. The triggers of 20% losses are varied. I think this is a false claim.

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Sun May 12, 2019 9:52 pm

HomerJ wrote:
Sun May 12, 2019 4:28 pm
The thing is, the historical 10% long-term return from the stock market?

That INCLUDES the crashes.

It's really not that necessary to avoid the crashes.
So then why is your AA 50/50?

In reality, none of us wants to experience a 50% or greater decline in our assets. The Great Depression's 89% drawdown would have been entirely unbearable for most investors, including most Bogleheads. Yes, we know with hindsight that it wasn't strictly necessary to avoid the crashes to get a decent return, but we don't know that that will be the case going forward.

Personally, I view trend following a bit like insurance. We don't buy insurance to get ahead; we buy it to avoid financial disasters. Yes, whipsaws are historically likely to happen over time, and it's also historically likely that my maximum drawdown will be significantly less than the market's. Some elect to do this with a significant allocation to fixed income, which is fine, but there are many roads to Dublin.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Sun May 12, 2019 9:55 pm

tadamsmar wrote:
Sun May 12, 2019 8:16 pm
I am very skeptical that +20% losses can be avoided because I see no evidence that trend-following works to avoid the 1987 crash. The triggers of 20% losses are varied. I think this is a false claim.
I agree that drawdowns of at least 20% are just part of the deal.

IIRC, using a 200 DMA or any shorter period would have avoided the 1987 crash, but longer periods would have experienced it. Regardless, if someone is skittish about a 20% loss, they probably have no business owning stocks outside of an all-in-one fund where the loss is somewhat hidden from their view.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
HomerJ
Posts: 12785
Joined: Fri Jun 06, 2008 12:50 pm

Re: Imperfect market timing still works -- if you stick with it

Post by HomerJ » Sun May 12, 2019 10:10 pm

willthrill81 wrote:
Sun May 12, 2019 9:52 pm
HomerJ wrote:
Sun May 12, 2019 4:28 pm
The thing is, the historical 10% long-term return from the stock market?

That INCLUDES the crashes.

It's really not that necessary to avoid the crashes.
So then why is your AA 50/50?
Because I'm close to retirement.
In reality, none of us wants to experience a 50% or greater decline in our assets.
Right, which is why I'm 50/50.

During the last 20% decline, you lost 20%, I lost 10%.

I got my 10% back, you didn't get your full 20% back.

During the next 50%-60% crash, you'll lose 20% (and get out), I'll lose 25%-30% holding all the way to the bottom.

Doesn't seem like your way is protecting you that much more than me.

It DOES give you more during the good times, since you're 100% stocks most of the time (except for the whipsaws).

My plan is nearly as conservative as yours, and I don't ever experience whipsaws (which would personally drive me crazy).
The J stands for Jay

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Sun May 12, 2019 10:24 pm

HomerJ wrote:
Sun May 12, 2019 10:10 pm
willthrill81 wrote:
Sun May 12, 2019 9:52 pm
HomerJ wrote:
Sun May 12, 2019 4:28 pm
The thing is, the historical 10% long-term return from the stock market?

That INCLUDES the crashes.

It's really not that necessary to avoid the crashes.
So then why is your AA 50/50?
Because I'm close to retirement.
In reality, none of us wants to experience a 50% or greater decline in our assets.
Right, which is why I'm 50/50.

During the last 20% decline, you lost 20%, I lost 10%.

I got my 10% back, you didn't get your full 20% back.

During the next 50%-60% crash, you'll lose 20% (and get out), I'll lose 25%-30% holding all the way to the bottom.

Doesn't seem like your way is protecting you that much more than me.

It DOES give you more during the good times, since you're 100% stocks most of the time (except for the whipsaws).

My plan is nearly as conservative as yours, and I don't ever experience whipsaws (which would personally drive me crazy).
Having 50% of my assets have a real return of around 1% for years on end would drive me crazy.

Again, many roads to Dublin!
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

skeptic42
Posts: 38
Joined: Mon Feb 11, 2019 5:27 pm

Re: Imperfect market timing still works -- if you stick with it

Post by skeptic42 » Mon May 13, 2019 4:02 am

willthrill81 wrote:
Sun May 12, 2019 10:24 pm
HomerJ wrote:
Sun May 12, 2019 10:10 pm
willthrill81 wrote:
Sun May 12, 2019 9:52 pm
HomerJ wrote:
Sun May 12, 2019 4:28 pm
The thing is, the historical 10% long-term return from the stock market?

That INCLUDES the crashes.

It's really not that necessary to avoid the crashes.
So then why is your AA 50/50?
Because I'm close to retirement.
In reality, none of us wants to experience a 50% or greater decline in our assets.
Right, which is why I'm 50/50.

During the last 20% decline, you lost 20%, I lost 10%.

I got my 10% back, you didn't get your full 20% back.

During the next 50%-60% crash, you'll lose 20% (and get out), I'll lose 25%-30% holding all the way to the bottom.

Doesn't seem like your way is protecting you that much more than me.

It DOES give you more during the good times, since you're 100% stocks most of the time (except for the whipsaws).

My plan is nearly as conservative as yours, and I don't ever experience whipsaws (which would personally drive me crazy).
Having 50% of my assets have a real return of around 1% for years on end would drive me crazy.

Again, many roads to Dublin!
This is the old risk return trade-off. It is really that simple, if one don't want to take the risk of 100% stocks, one has to accept a lower expected return too. In contrast, trend followers want to avoid the risk of 100% stocks, but without expecting lower returns. Sounds like beating the market or wishful thinking or some hidden risk.

Theoretically, a bad sequence of whipsaws could create a larger drawdown for a trend following strategy than buy and hold 100% stocks. Maybe, trend followers are taking more risk than they are aware of, because the risk didn't show up in backtests. 8-)

klrjaa
Posts: 15
Joined: Sat Mar 23, 2019 10:55 am

Re: Imperfect market timing still works -- if you stick with it

Post by klrjaa » Mon May 13, 2019 5:58 am

tadamsmar wrote:
Sun May 12, 2019 8:16 pm
The article cited in the OP sets a standard: "you never want to lose 20% of your money". Then he sets a standard for back-testing against that standard: The last 19 years and no more than that (the last 2 plus +20% downturns). This stuff is not worthy of serious consideration.

Also, he uses holding S&P 500 as a benchmark. But the mainline Boglehead method involves rebalancing. If market behavior potentiates trend-following to beat buy and hold then it is likely that it potentiates rebalancing to beat buy and hold. He uses a strawman for his benchmark!

Anyway, his stuff is very unconvincing.

But maybe a clearer-thinking advocate of trend-following can avoid such blunders and show that trend-following provide superior returns against a realistic benchmark when used in tax-deferred or Roth accounts (not in taxable).

I am very skeptical that +20% losses can be avoided because I see no evidence that trend-following works to avoid the 1987 crash. The triggers of 20% losses are varied. I think this is a false claim.
https://allocatesmartly.com/antonaccis- ... -momentum/
https://allocatesmartly.com/meta-strate ... trategies/
https://allocatesmartly.com/new-strateg ... -balanced/
https://allocatesmartly.com/meb-fabers- ... olio-lite/
https://allocatesmartly.com/vigilant-as ... w-keuning/
https://allocatesmartly.com/taa-strateg ... following/

I could keep going. You guys with the no evidence mantra never seem to offer anything but opinions. If you took the time to read the source paper behind many of the TAA models, you could decide for yourself if you feel comfortable. Many systems have out of sample testing and then in sample. For some TAA strategies, even if the results are good, if it doesn't have a sound basis behind it I stay away, but there are enough solid strategies, that when combined to smooth the ride since you shouldn't rely on a single strategy https://blog.thinknewfound.com/2018/04/ ... following/

My personal model uses 6 different strategies, only 1 of which is included in the list above. It has historical returns of 11.5 with max drawdown (end of month) of -5.9%. So even if intramonth drawdown is more volatile, I'd expect to see a drawdown in maybe to 10-12% range. The strategies I use get generally are defensive fairly quickly. The model trades all the most liquid ETFs so no built in bias towards what was hot mutual fund performance at the time
Last edited by klrjaa on Mon May 13, 2019 6:31 am, edited 2 times in total.

klrjaa
Posts: 15
Joined: Sat Mar 23, 2019 10:55 am

Re: Imperfect market timing still works -- if you stick with it

Post by klrjaa » Mon May 13, 2019 6:18 am

skeptic42 wrote:
Sun May 12, 2019 4:03 pm
klrjaa wrote:
Sat May 11, 2019 1:36 pm
willthrill81 wrote:
Sat May 11, 2019 11:57 am
abc132 wrote:
Sat May 11, 2019 10:45 am
Forester wrote:
Sat May 11, 2019 2:42 am
Yes, most TF trades will lose money - that's the whole point! No one can predict for certain when a correction becomes a 50% bear market.
I was saying they (imperfect timing, etc) would produce less returns than just following the market.
Historically, you would be incorrect. Just from 1994 until now, which is all of the data for VTSMX in Portfolio Visualizer, using a 10 month moving average (~210 day moving average) strategy with just cash as the out of market asset, you would have had a .61% higher CAGR but with a maximum drawdown of -17.57% instead of -50.89%. And if you used any bond index fund instead of cash, the return would have been at least 1% higher than that. But there were significant periods where this strategy would have underperformed buy-and-hold, such as 1998-2000 and 2009-current. Will the trading algorithms stick with a strategy that underperforms buy-and-hold for a decade or more? More importantly, will the people operating the algorithms stick with it?

Others have analyzed such a simple trend following strategy going back to the 1940s and come to very similar conclusions.
It's curious to me that this can even still be a debate. I'm convinced throwing up facts and data is not going to change any BAH mind.
Maybe, trend followers are fooled by randomness and performance chasing the good performance of trend following in the past. One argument for trend following is to avoid big drawdowns compared to buy and hold, because backtests show that behavior for historical data. But it is possible that the future sequence of returns results in a larger drawdown for a trend following strategy compared to buy and hold. So, without a good facts and data based explanation why the future sequence of returns has to behave similar as in the past, I will not change my stubborn BAH mind. :twisted:
https://blog.thinknewfound.com/2018/03/ ... -momentum/

and then this on future expected returns of static portfolios that don't account for the current interest rate envt

https://betterbuyandhold.com/buy-hold-b ... trategies/

Forester
Posts: 79
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Imperfect market timing still works -- if you stick with it

Post by Forester » Mon May 13, 2019 7:13 am

How often are these strategies in and out of stocks? Aside from DM, they are hidden behind paywalls. I'd like to see the trades that have been made since 2010. It doesn't matter whether the lookback is 12 months or equal-weighted 3/6/9/12 or whatever, it's been a fully risk-on scenario since 2010 (arguably, 20% corrections are not what trend following is intended to sidestep, and if it does, it would be through luck).

User avatar
tadamsmar
Posts: 8241
Joined: Mon May 07, 2007 12:33 pm

Re: Imperfect market timing still works -- if you stick with it

Post by tadamsmar » Mon May 13, 2019 7:16 am

klrjaa wrote:
Mon May 13, 2019 5:58 am
tadamsmar wrote:
Sun May 12, 2019 8:16 pm
The article cited in the OP sets a standard: "you never want to lose 20% of your money". Then he sets a standard for back-testing against that standard: The last 19 years and no more than that (the last 2 plus +20% downturns). This stuff is not worthy of serious consideration.

Also, he uses holding S&P 500 as a benchmark. But the mainline Boglehead method involves rebalancing. If market behavior potentiates trend-following to beat buy and hold then it is likely that it potentiates rebalancing to beat buy and hold. He uses a strawman for his benchmark!

Anyway, his stuff is very unconvincing.

But maybe a clearer-thinking advocate of trend-following can avoid such blunders and show that trend-following provide superior returns against a realistic benchmark when used in tax-deferred or Roth accounts (not in taxable).

I am very skeptical that +20% losses can be avoided because I see no evidence that trend-following works to avoid the 1987 crash. The triggers of 20% losses are varied. I think this is a false claim.
https://allocatesmartly.com/antonaccis- ... -momentum/
https://allocatesmartly.com/meta-strate ... trategies/
https://allocatesmartly.com/new-strateg ... -balanced/
https://allocatesmartly.com/meb-fabers- ... olio-lite/
https://allocatesmartly.com/vigilant-as ... w-keuning/
https://allocatesmartly.com/taa-strateg ... following/

I could keep going. You guys with the no evidence mantra never seem to offer anything but opinions. If you took the time to read the source paper behind many of the TAA models, you could decide for yourself if you feel comfortable. Many systems have out of sample testing and then in sample. For some TAA strategies, even if the results are good, if it doesn't have a sound basis behind it I stay away, but there are enough solid strategies, that when combined to smooth the ride since you shouldn't rely on a single strategy https://blog.thinknewfound.com/2018/04/ ... following/

My personal model uses 6 different strategies, only 1 of which is included in the list above. It has historical returns of 11.5 with max drawdown (end of month) of -5.9%. So even if intramonth drawdown is more volatile, I'd expect to see a drawdown in maybe to 10-12% range. The strategies I use get generally are defensive fairly quickly. The model trades all the most liquid ETFs so no built in bias towards what was hot mutual fund performance at the time
Kindly show me where they avoid all the +20% losses that are evidenced in the historical record.

Or, agree with me that the OP link is wrong since it makes this claim.
Last edited by tadamsmar on Mon May 13, 2019 12:56 pm, edited 2 times in total.

skeptic42
Posts: 38
Joined: Mon Feb 11, 2019 5:27 pm

Re: Imperfect market timing still works -- if you stick with it

Post by skeptic42 » Mon May 13, 2019 7:36 am

klrjaa wrote:
Mon May 13, 2019 5:58 am

...

https://allocatesmartly.com/antonaccis- ... -momentum/
https://allocatesmartly.com/meta-strate ... trategies/
https://allocatesmartly.com/new-strateg ... -balanced/
https://allocatesmartly.com/meb-fabers- ... olio-lite/
https://allocatesmartly.com/vigilant-as ... w-keuning/
https://allocatesmartly.com/taa-strateg ... following/

I could keep going. You guys with the no evidence mantra never seem to offer anything but opinions. If you took the time to read the source paper behind many of the TAA models, you could decide for yourself if you feel comfortable.

...
Creating well performing backtests is no evidence and suggesting the good performance will repeat in the future is just an opinion as well. And don't underestimate the due diligence of the bogleheads. Many bogleheads already investigated TAA and trend following and other fancy stuff, but decided to choose a simpler approach to investing after recognizing the dismal real money performance of many beat the market schemes. Search the forum, there is a mount of evidence that the market is hard to beat.

If you need to beat the market to achieve your investment goals, then good luck to you!
To each his own.

klrjaa
Posts: 15
Joined: Sat Mar 23, 2019 10:55 am

Re: Imperfect market timing still works -- if you stick with it

Post by klrjaa » Mon May 13, 2019 7:38 am

tadamsmar wrote:
Mon May 13, 2019 7:16 am
klrjaa wrote:
Mon May 13, 2019 5:58 am
tadamsmar wrote:
Sun May 12, 2019 8:16 pm
The article cited in the OP sets a standard: "you never want to lose 20% of your money". Then he sets a standard for back-testing against that standard: The last 19 years and no more than that (the last 2 plus +20% downturns). This stuff is not worthy of serious consideration.

Also, he uses holding S&P 500 as a benchmark. But the mainline Boglehead method involves rebalancing. If market behavior potentiates trend-following to beat buy and hold then it is likely that it potentiates rebalancing to beat buy and hold. He uses a strawman for his benchmark!

Anyway, his stuff is very unconvincing.

But maybe a clearer-thinking advocate of trend-following can avoid such blunders and show that trend-following provide superior returns against a realistic benchmark when used in tax-deferred or Roth accounts (not in taxable).

I am very skeptical that +20% losses can be avoided because I see no evidence that trend-following works to avoid the 1987 crash. The triggers of 20% losses are varied. I think this is a false claim.
https://allocatesmartly.com/antonaccis- ... -momentum/
https://allocatesmartly.com/meta-strate ... trategies/
https://allocatesmartly.com/new-strateg ... -balanced/
https://allocatesmartly.com/meb-fabers- ... olio-lite/
https://allocatesmartly.com/vigilant-as ... w-keuning/
https://allocatesmartly.com/taa-strateg ... following/

I could keep going. You guys with the no evidence mantra never seem to offer anything but opinions. If you took the time to read the source paper behind many of the TAA models, you could decide for yourself if you feel comfortable. Many systems have out of sample testing and then in sample. For some TAA strategies, even if the results are good, if it doesn't have a sound basis behind it I stay away, but there are enough solid strategies, that when combined to smooth the ride since you shouldn't rely on a single strategy https://blog.thinknewfound.com/2018/04/ ... following/

My personal model uses 6 different strategies, only 1 of which is included in the list above. It has historical returns of 11.5 with max drawdown (end of month) of -5.9%. So even if intramonth drawdown is more volatile, I'd expect to see a drawdown in maybe to 10-12% range. The strategies I use get generally are defensive fairly quickly. The model trades all the most liquid ETFs so no built in bias towards what was hot mutual fund performance at the time
Kindly show me where they avoid all the +20% losses are in evidence in the historical record.

Or, agree with me that the OP link is wrong since it makes this claim.
It's in the drawdown curve; if you scroll down each entry you can see the all the stats. Maybe I'm missing the gist of your question, thanks

klrjaa
Posts: 15
Joined: Sat Mar 23, 2019 10:55 am

Re: Imperfect market timing still works -- if you stick with it

Post by klrjaa » Mon May 13, 2019 7:41 am

skeptic42 wrote:
Mon May 13, 2019 7:36 am
klrjaa wrote:
Mon May 13, 2019 5:58 am

...

https://allocatesmartly.com/antonaccis- ... -momentum/
https://allocatesmartly.com/meta-strate ... trategies/
https://allocatesmartly.com/new-strateg ... -balanced/
https://allocatesmartly.com/meb-fabers- ... olio-lite/
https://allocatesmartly.com/vigilant-as ... w-keuning/
https://allocatesmartly.com/taa-strateg ... following/

I could keep going. You guys with the no evidence mantra never seem to offer anything but opinions. If you took the time to read the source paper behind many of the TAA models, you could decide for yourself if you feel comfortable.

...
Creating well performing backtests is no evidence and suggesting the good performance will repeat in the future is just an opinion as well. And don't underestimate the due diligence of the bogleheads. Many bogleheads already investigated TAA and trend following and other fancy stuff, but decided to choose a simpler approach to investing after recognizing the dismal real money performance of many beat the market schemes. Search the forum, there is a mount of evidence that the market is hard to beat.

If you need to beat the market to achieve your investment goals, then good luck to you!
To each his own.
same can be said for BAH. And agree 100% that backtessted results don't guarantee anything. That's why I spread my bets across strategies that have generally low correlation to eaachother.

klrjaa
Posts: 15
Joined: Sat Mar 23, 2019 10:55 am

Re: Imperfect market timing still works -- if you stick with it

Post by klrjaa » Mon May 13, 2019 8:02 am

Forester wrote:
Mon May 13, 2019 7:13 am
How often are these strategies in and out of stocks? Aside from DM, they are hidden behind paywalls. I'd like to see the trades that have been made since 2010. It doesn't matter whether the lookback is 12 months or equal-weighted 3/6/9/12 or whatever, it's been a fully risk-on scenario since 2010 (arguably, 20% corrections are not what trend following is intended to sidestep, and if it does, it would be through luck).
The average number of trades per year is shown in the summary statistics table for each strategy.

In terms of behind paywall/trades, the data is often available from other sources too. For example, all that is available from https://indexswingtrader.blogspot.com/2 ... asset.html for Vigilant Asset Allocation. Allocate Smartly tracks the Top 1/Bad 1 version. The strategy Signals Page shows the recent trades and current next months allocation if it were the end of May.

And you simply cannot make a blanket statement regarding risk on/off as each strategy makes that decision as part of it's rule set. For example GTAA aggressive 6 (tracked by allocate smartly and from meb fabers classic paper) will be in up to 6 assets based on relative strength and if the asset is above it's own 10 month moving average. So if of the top 6 only 3 are above their own 10 month moving average, the strategy would be in cash (total of 50%) for the other 3. Is risk on only when at least 6 are on buys? 5? And this is a very simple strategy; others tracked by AS....well you need to understand them and make your own decision. Allocate Smartly does provide an overall risk on/off % as a function of the current selected assets in aggregate for the strategies. Risk on has grown in the last 3 months for example which makes sense

The following shows performance in April but more important to this discussion is the risk on/off and allocations in aggregate. Scroll down to to data dump area where you can see the shorter term info as well as longer term by clicking on each of the 2 "click for a longer view". Note that all of the data I am providing is not behind a paywall, it's freely available from the blog.

https://allocatesmartly.com/tactical-as ... pril-2019/

Finally, one counter argument might be "costs too much" to get behind the scenes. A couple of thoughts...You could take a one month test drive, and the overall cost of making trades plus the yearly subscription is about 800 bucks which is 1/10 of one percent on an 800k bankroll. Everyone's mileage will vary of course. There is also consideration for taxes and there's plenty behind the curtain that provides that analysis.

Again, I'm no way affiliated with the site other than a paying member. It seems to me some of the TAA folks should consider taking a look as that was my real intent of providing all the data points as no single TAA approach should be held sacred IMO

Forester
Posts: 79
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Imperfect market timing still works -- if you stick with it

Post by Forester » Mon May 13, 2019 2:44 pm

If that sites subscription was less I'd be tempted. But I can already use Excel to calculate moving averages, the unemployment data can be downloaded from the BLS & LEI data as far as I know is freely available.

My point about risk on or off, is trend following is not the same as trend. Arguably it's been an unbroken bullmarket for US stocks. If a model has not been fully invested in the S&P 500, it's been slipping. I definitely believe there's something to the GEM framework, but I am sceptical about some of these other tactical schemes which do things like treat the Nasdaq or Emerging Markets as separate momentum candidates. I think ultimately the best course, whatever someone's allocation to trend is, is to go 50-50 Price - Price&LEI(s).

randomguy
Posts: 7509
Joined: Wed Sep 17, 2014 9:00 am

Re: Imperfect market timing still works -- if you stick with it

Post by randomguy » Mon May 13, 2019 3:29 pm

skeptic42 wrote:
Mon May 13, 2019 4:02 am

This is the old risk return trade-off. It is really that simple, if one don't want to take the risk of 100% stocks, one has to accept a lower expected return too. In contrast, trend followers want to avoid the risk of 100% stocks, but without expecting lower returns. Sounds like beating the market or wishful thinking or some hidden risk.

Theoretically, a bad sequence of whipsaws could create a larger drawdown for a trend following strategy than buy and hold 100% stocks. Maybe, trend followers are taking more risk than they are aware of, because the risk didn't show up in backtests. 8-)
It might be wishful thinking. It might also be the result of holding a better portfolio that generates a better risk/reward return than just holding stocks does. If you take it as a tennant of faith that you can't do better than holding total market, then trend following isn't for you. If you accept the possibility of beating the market, then trend following gets interesting. You can make up stories of why it works (i.e. people don't have the will to execute so the premium doesn't get grazed away) but they are hard to prove.

And to some extent we aren't talking about beating the market. We are talking about chopping off both ends (i.e. some of the outperformance and underperformance) in an attempt to get smoother results. For someone in the accumulation phase, that is pretty uninteresting as you don't care about volatility. Keep pouring the money in and in 20 years you will be fine. For a person in the decumulation phase, avoiding the bottom returns has a ton of value. You don't need very high real returns to generate a 5% SWR see (note holding 100% stocks and 100% trend following isn't a really great portfolio:))

https://www.portfoliovisualizer.com/tes ... 0&total1=0


Obviously the big question is how often it works. Same people retiring in 1985 results you being poorer pretty much all the time

https://www.portfoliovisualizer.com/tes ... 0&total1=0

Now maybe you don't care about the good cases (i.e. who cares if don't go broke) so trend following is a good trade off. Then you have to figure out if there are any cases where you really fail hard. Not sure if they are out there but there is no reason why they can't exist. Something variation of 1987 where you are in the market for the loss, you go to cash and then their is a rapid rally while you are in cash and you end up locking in some 50% loss.

And obviously this is only part of your strategy. Things like bonds and rebalancing also factor in but they are sort of independent of what you do with your stocks.

klrjaa
Posts: 15
Joined: Sat Mar 23, 2019 10:55 am

Re: Imperfect market timing still works -- if you stick with it

Post by klrjaa » Mon May 13, 2019 3:44 pm

Forester wrote:
Mon May 13, 2019 2:44 pm
If that sites subscription was less I'd be tempted. But I can already use Excel to calculate moving averages, the unemployment data can be downloaded from the BLS & LEI data as far as I know is freely available.

My point about risk on or off, is trend following is not the same as trend. Arguably it's been an unbroken bullmarket for US stocks. If a model has not been fully invested in the S&P 500, it's been slipping. I definitely believe there's something to the GEM framework, but I am sceptical about some of these other tactical schemes which do things like treat the Nasdaq or Emerging Markets as separate momentum candidates. I think ultimately the best course, whatever someone's allocation to trend is, is to go 50-50 Price - Price&LEI(s).
Hi Forester,

Yep, here you on the US bull market. Many of the models force diversification via various means so being in those creates a drag. One could switch between models if say overall market is in bull mode (say above 10 month moving average) or other proxy. One of the models at AS is Philosophical Economics Growth Trend Timing http://www.philosophicaleconomics.com/2016/01/gtt/ which gets to recession outlook (or not). It's performance frankly is pretty poor compared to most of the other TAA models so I don't use it. AS does like the approach and is on the lookout for other similar approaches based on published data (papers, other author websites). And to your point, not many have beat 60/40 over the last 10 years and those who have are generally simple strategies that only trade one or two assets. I stay away from these due to high drawdowns; all roughly 20% in 2016 whereas mine was 5%. No right or wrong answer. An interesting post from AS was https://allocatesmartly.com/timing-taa- ... -approach/ regrading chasing performance via constant strategy switching. Message is Buy and Hold your TAA strategies !!

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Mon May 13, 2019 4:00 pm

randomguy wrote:
Mon May 13, 2019 3:29 pm
skeptic42 wrote:
Mon May 13, 2019 4:02 am

This is the old risk return trade-off. It is really that simple, if one don't want to take the risk of 100% stocks, one has to accept a lower expected return too. In contrast, trend followers want to avoid the risk of 100% stocks, but without expecting lower returns. Sounds like beating the market or wishful thinking or some hidden risk.

Theoretically, a bad sequence of whipsaws could create a larger drawdown for a trend following strategy than buy and hold 100% stocks. Maybe, trend followers are taking more risk than they are aware of, because the risk didn't show up in backtests. 8-)
It might be wishful thinking. It might also be the result of holding a better portfolio that generates a better risk/reward return than just holding stocks does. If you take it as a tennant of faith that you can't do better than holding total market, then trend following isn't for you. If you accept the possibility of beating the market, then trend following gets interesting. You can make up stories of why it works (i.e. people don't have the will to execute so the premium doesn't get grazed away) but they are hard to prove.

And to some extent we aren't talking about beating the market. We are talking about chopping off both ends (i.e. some of the outperformance and underperformance) in an attempt to get smoother results.
Over the last 30 years, that's precisely what many trend following systems appear to have done. Take, for instance, the very simple 10 month moving average strategy using 1 month T-bills as the out-of-market asset. It underperformed BAH in the 1990s (12.86% vs. 18.07%), especially the last few years of that decade, then posted very respectable returns in the 2000s (7.82% vs. -1.03%), and then again underperformed BAH over the last decade (2010 - Apr. 2019; 7.5% vs. 13.14%). Over the entirety of the that period (1990 - Apr. 2019), the absolute returns were remarkably similar (9.41% vs. 9.68%), but the 10 MMA had much lower maximum drawdown (16.77% vs. 50.97%). And these results would have been even better had any indexed bond fund other than perhaps corporate bonds been used instead of 1 month T-bills (i.e. cash).

Now if you believe that that is a representative sampling of what this type of strategy will do in the future, it's still not at all a 'slam dunk' that everyone or even many investors would be able to stick with it. Why? It would have been very difficult for many investors to underperform BAH by over 5% annually for a decade, never knowing if their strategy would 'pay off' in the future. Just look at how many people, including Rick Ferri very recently, are questioning investors' ability to stick with SCV when it's only underperformed TSM over the last decade by about 1%. I'm a bit surprised that we haven't had many threads saying something like "10 MMA has been decimated by BAH, with the latter outperforming the former by a cumulative 61% over the last ten years. Trend following is dead, and BAH of the three-fund portfolio is the only reasonable strategy!"
randomguy wrote:
Mon May 13, 2019 3:29 pm
For someone in the accumulation phase, that is pretty uninteresting as you don't care about volatility. Keep pouring the money in and in 20 years you will be fine. For a person in the decumulation phase, avoiding the bottom returns has a ton of value. You don't need very high real returns to generate a 5% SWR see (note holding 100% stocks and 100% trend following isn't a really great portfolio:))
Here's where I disagree. What if 2000-2009 is the last decade I have before retirement, meaning that it has the biggest impact of any on my final portfolio balance? Experiencing an annual return of -1% might prohibit me from retiring on time, whereas getting almost 8% would likely be better than I was planning for.

But I agree with you that if we could 'chop off' the left side of the tail risk in the withdrawal phase, the SWR could improve significantly. This academic paper analyzes the impact of the 200 DMA strategy on SWRs and found that it would have significantly improved the latter.

Those who are convinced that it's impossible to experience the market's upside without a one-to-one exposure to its downside will never be convinced by any analyses and arguments because it's simply incompatible with their view of how the markets work.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
HomerJ
Posts: 12785
Joined: Fri Jun 06, 2008 12:50 pm

Re: Imperfect market timing still works -- if you stick with it

Post by HomerJ » Mon May 13, 2019 4:55 pm

willthrill81 wrote:
Mon May 13, 2019 4:00 pm
Those who are convinced that it's impossible to experience the market's upside without a one-to-one exposure to its downside will never be convinced by any analyses and arguments because it's simply incompatible with their view of how the markets work.
Incompatible with my view of how human nature works maybe :)

Paul Merriman himself doesn't have the discipline to faithfully follow his own trend algorithm.

Anyway, I don't believe in a free lunch. You want to be invested during the good times, and not invested during the bad times, by predicting the bad times based on signals from back-testing. But if the future isn't the same as the past, the signals may not work as well. No one knows.

And since no one knows, most humans will question their strategy the longer it seems to be doing poorly.

I'll just take all of the upside and all of the downside, and change my risk profile with an allocation to bonds. (I'm 50/50 for other readers).

Our risk profiles are remarkably similar on the downside in general. You'll probably get out around halfway through a drop. I'll experience the full drop, but with only half my money. Basically the same result, if your signals line up.

But you do risk the chance of a black swan event where markets drop very fast before your signals (like unemployment) show up. You only check at the end of the month, and the Market could indeed drop 40% in 6-11 weeks, and then finally your signal at the end of the 2nd or 3rd month is triggered, but only after experiencing most (or even all) of the entire full drop.

Since I'm 50/50, I'll never experience the full drop on all of my portfolio. But it could happen to you. Maybe not likely. But you could have bad luck with end of month timing and signals.

On other side, you get most (nearly all) of the upside. Beating a 100% stocks buy and hold guy would be hard because of the whipsaws and false positives, but you'll very likely make more than a 50% stocks guy like me, because I concentrated on protecting myself on the downside, which cost me heavily on the upside.

But I don't need that much upside to make my number... I'm concentrated on protecting myself from the downside, which I believe I've done better than you.

The funny thing is, you also don't need the extra upside. Even a horrible bad luck downside incident won't ruin you. You've got good job security and great pensions and SS coming, and you'll be fine no matter what.

Maybe that's why I think you'll be able to stick with this plan. Because you won't get that hurt if it doesn't work out that well.

But others thinking trend-following is a way to avoid the downside and get the upside may be a lot more nervous about their money. I think it may be harder for them to stick with it. It would be hard for me to follow it. Having 100% of my money in stocks when I might retire any time in the next 1-5 years would give me ulcers.

Again, even Paul Merriman can't do it. Even with only half his money.

I'm not following a strategy from a guy who can't stomach doing it himself.
The J stands for Jay

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Mon May 13, 2019 6:06 pm

HomerJ wrote:
Mon May 13, 2019 4:55 pm
willthrill81 wrote:
Mon May 13, 2019 4:00 pm
Those who are convinced that it's impossible to experience the market's upside without a one-to-one exposure to its downside will never be convinced by any analyses and arguments because it's simply incompatible with their view of how the markets work.
Incompatible with my view of how human nature works maybe :)

Paul Merriman himself doesn't have the discipline to faithfully follow his own trend algorithm.

Anyway, I don't believe in a free lunch. You want to be invested during the good times, and not invested during the bad times, by predicting the bad times based on signals from back-testing. But if the future isn't the same as the past, the signals may not work as well. No one knows.

And since no one knows, most humans will question their strategy the longer it seems to be doing poorly.

I'll just take all of the upside and all of the downside, and change my risk profile with an allocation to bonds. (I'm 50/50 for other readers).

Our risk profiles are remarkably similar on the downside in general. You'll probably get out around halfway through a drop. I'll experience the full drop, but with only half my money. Basically the same result, if your signals line up.

But you do risk the chance of a black swan event where markets drop very fast before your signals (like unemployment) show up. You only check at the end of the month, and the Market could indeed drop 40% in 6-11 weeks, and then finally your signal at the end of the 2nd or 3rd month is triggered, but only after experiencing most (or even all) of the entire full drop.

Since I'm 50/50, I'll never experience the full drop on all of my portfolio. But it could happen to you. Maybe not likely. But you could have bad luck with end of month timing and signals.

On other side, you get most (nearly all) of the upside. Beating a 100% stocks buy and hold guy would be hard because of the whipsaws and false positives, but you'll very likely make more than a 50% stocks guy like me, because I concentrated on protecting myself on the downside, which cost me heavily on the upside.

But I don't need that much upside to make my number... I'm concentrated on protecting myself from the downside, which I believe I've done better than you.

The funny thing is, you also don't need the extra upside. Even a horrible bad luck downside incident won't ruin you. You've got good job security and great pensions and SS coming, and you'll be fine no matter what.

Maybe that's why I think you'll be able to stick with this plan. Because you won't get that hurt if it doesn't work out that well.

But others thinking trend-following is a way to avoid the downside and get the upside may be a lot more nervous about their money. I think it may be harder for them to stick with it. It would be hard for me to follow it. Having 100% of my money in stocks when I might retire any time in the next 1-5 years would give me ulcers.

Again, even Paul Merriman can't do it. Even with only half his money.

I'm not following a strategy from a guy who can't stomach doing it himself.
That's a well rounded take on the issue, and I won't dispute any of it, although I will not receive a pension. I get a very good match on a DC plan, and my employer contributes $1,400 annually to our family HSA but no pension.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

randomguy
Posts: 7509
Joined: Wed Sep 17, 2014 9:00 am

Re: Imperfect market timing still works -- if you stick with it

Post by randomguy » Mon May 13, 2019 8:50 pm

HomerJ wrote:
Mon May 13, 2019 4:55 pm

I'll just take all of the upside and all of the downside, and change my risk profile with an allocation to bonds. (I'm 50/50 for other readers).

Our risk profiles are remarkably similar on the downside in general. You'll probably get out around halfway through a drop. I'll experience the full drop, but with only half my money. Basically the same result, if your signals line up.

But you do risk the chance of a black swan event where markets drop very fast before your signals (like unemployment) show up. You only check at the end of the month, and the Market could indeed drop 40% in 6-11 weeks, and then finally your signal at the end of the 2nd or 3rd month is triggered, but only after experiencing most (or even all) of the entire full drop.

Since I'm 50/50, I'll never experience the full drop on all of my portfolio. But it could happen to you. Maybe not likely. But you could have bad luck with end of month timing and signals.

On other side, you get most (nearly all) of the upside. Beating a 100% stocks buy and hold guy would be hard because of the whipsaws and false positives, but you'll very likely make more than a 50% stocks guy like me, because I concentrated on protecting myself on the downside, which cost me heavily on the upside.

Sure but in your case wouldn't a fair comparison be 50% trend following and 50% bonds? Same performance and even less risk. Well if it works:) I am not sure anyone recommends trend following to be your whole portfolio. You use it to handle the money you want exposed to equity risk.n You might up your AA because of the expected lower volatility but you want some bonds.

To some extent we all have a reaction to the words of "market timing". Call it a momentum factor investing and people don't get anywhere near as up in arms about it. But like factor investing there is an element of faith that you have to have. We know small value investing has been the winning strategy for pretty much all 20 year periods for the last 150 or so years. But there are plenty of people who think that strategy will not work going forward.

As far as execution, it is a know yourself issue. Some people can't buy and hold. They feel better taking action. A lot of people on the other hand only want to look at their account once/quarter to rebalance. Pick whatever you want. The fact the one of the people pushing this trend has execution issues a personal one that may not apply to you.

Theoretical
Posts: 1474
Joined: Tue Aug 19, 2014 10:09 pm

Re: Imperfect market timing still works -- if you stick with it

Post by Theoretical » Mon May 13, 2019 11:55 pm

This is a good resource: https://www.toptradersunplugged.com/res ... ket-score/

It's oriented towards a traditional managed futures type trend following portfolio, and it's run by a guy who works at one of the earliest Managed Futures shops - Dunn Capital, with about a 40 year history.

User avatar
steve roy
Posts: 1635
Joined: Thu May 13, 2010 5:16 pm

Re: Imperfect market timing still works -- if you stick with it

Post by steve roy » Tue May 14, 2019 12:15 am

Brain waves come, and brain waves go.

A few years back at this very forum, many were extolling the virtues of "The Permanent Portfolio" (25% GLD, 25% Total Stock Market; 25% long term Treausries; 25% Short Term Treasuries).

For a while, this portfolio did outstandingly well. The "oohs" and "aahhs" went on and on. Financial Nirvana had been reached. But then, funny thing, other portfolios began out-performing it, and continued to our-perform. And the "oohs" and "aahs" stopped.

Nothing works forever. Market-timing is just another breed of portfolio, and falls in and out of favor like every other style of investing. Fidelity Investments investigated what worked best, and discovered that their FORGETFUL investors (the people who FORGOT they had investments at Fidelity ... or were dead) happened to be the most successful investors.

So. You want to excel at investing? Build a diversified portfolio. And then, get amnesia. Or die.

skeptic42
Posts: 38
Joined: Mon Feb 11, 2019 5:27 pm

Re: Imperfect market timing still works -- if you stick with it

Post by skeptic42 » Tue May 14, 2019 4:56 am

willthrill81 wrote:
Mon May 13, 2019 4:00 pm
Those who are convinced that it's impossible to experience the market's upside without a one-to-one exposure to its downside will never be convinced by any analyses and arguments because it's simply incompatible with their view of how the markets work.
It is not impossible to experience the market's upside without the downside. The people believing in efficient markets or that the market is hard to beat think it is just random luck for most active investors, because no one wants the downside and everyone wants the upside, but someone has to take the downside. Certainly, there are some sophisticated market participants (I am not one of them) who can beat the total market of stocks and bonds on a risk-adjusted basis with some fancy and secret trading scheme. I am skeptical that such a simple and public trading strategy like trend following produces more than random noise when implemented with real money and real costs. And I think there lurks more risk than visible in backtests, e.g. a deeper drawdown than the market. That's why these backtest analyses and arguments will not convince me, but the advocates of market timing will never be convinced by arguments against it either.
:sharebeer

Forester
Posts: 79
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Imperfect market timing still works -- if you stick with it

Post by Forester » Tue May 14, 2019 6:19 am

skeptic42 wrote:
Tue May 14, 2019 4:56 am

It is not impossible to experience the market's upside without the downside.
Trend following has been painful since the GFC. My understanding is that one of the Cambria trend following funds had to close down due to poor performance.

The controversial question for me is, not does trend following avoid severe drawdowns & participate in bull markets - it does. TF is guaranteed to sail through 2008/09 because it operates on price and price determines the severity of a bear market. But CAN whipsaws be managed. Are the attempts to do this foolhardy "curve fitting". Should we be suspicious of trend models which navigated 2015/16 by using LEIs or other methods.

LittleD
Posts: 194
Joined: Sun Jul 31, 2011 7:23 am

Re: Imperfect market timing still works -- if you stick with it

Post by LittleD » Tue May 14, 2019 6:22 am

There will always be whipsaws when using trend following and moving average crossovers to signal buys and sells in the market. I am including what I think is a very good analysis of how you can trade indexes just using one index, the SPX and apply those signals to other indexes to improve your overall return over time. Trend following works and if followed can show good results over an investment lifetime. Now let me say that there is absolutely nothing wrong with B&H with proper allocation to bonds and cash. Many roads to success...enjoy!

https://www.youtube.com/watch?v=h-xLsvij-04

30minutes of your time to learn what works.

FireProof
Posts: 660
Joined: Thu May 05, 2011 12:15 pm

Re: Imperfect market timing still works -- if you stick with it

Post by FireProof » Tue May 14, 2019 7:06 am

This thread illustrates perfectly why the average investor gets like 5% below the market return.

User avatar
Tycoon
Posts: 1452
Joined: Wed Mar 28, 2012 7:06 pm

Re: Imperfect market timing still works -- if you stick with it

Post by Tycoon » Tue May 14, 2019 7:31 am

FireProof wrote:
Tue May 14, 2019 7:06 am
This thread illustrates perfectly why the average investor gets like 5% below the market return.
I was thinking the same thing. Some will never succeed.
“To know what you know and what you do not know, that is true knowledge.” Confucius

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Tue May 14, 2019 10:01 am

FireProof wrote:
Tue May 14, 2019 7:06 am
This thread illustrates perfectly why the average investor gets like 5% below the market return.
I'm not sure why you think this thread is representative of the average investor, who is constantly trying to predict what direction the market is going next. Contrary to what many believe, rules-based trend following strategies are based on a historical expectancy, not a vain attempt at prediction.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

LittleD
Posts: 194
Joined: Sun Jul 31, 2011 7:23 am

Re: Imperfect market timing still works -- if you stick with it

Post by LittleD » Tue May 14, 2019 10:08 am

I'm not sure why you think this thread is representative of the average investor, who is constantly trying to predict what direction the market is going next. Contrary to what many believe, rules-based trend following strategies are based on a historical expectancy, not a vain attempt at prediction.

You are just whistling past the graveyard. Many here are just wedded and wholly invested in the B&H mantra which is good for them and not a terrible strategy for those who just want to set it for extended periods of time. If they sleep well at night then great for them. The only negative is that many just want to rag in one thread that explores the topic in detail. They could just as easily unmark the thread and move to something more their liking. Keep on commenting...I love it!!! :sharebeer

Topic Author
CULater
Posts: 1916
Joined: Sun Nov 13, 2016 10:59 am

Re: Imperfect market timing still works -- if you stick with it

Post by CULater » Tue May 14, 2019 10:09 am

Can't say that I'm all that enamored of the 50/50 BAH retirement portfolio. This ignores the risk of unexpected inflation, which kills both stocks and bonds. One example: a person retiring in 1973 with a 50/50 portfolio using the 4% rule for withdrawals ended up 9 1/2 years later with just 41% of their 50/50 portfolio left in purchasing power. Nice that they got lucky and could cut back on the dogfood diet before going bust. If they'd been 100% in stocks they would have had 32% left. Bonds helped but not hugely. Unable to determine the result for trend-following, but I venture that it would have been a LOT better than 50/50 BAH during that period of time.
Faced with the choice between changing one’s mind and proving that there is no need to do so, almost everyone gets busy on the proof. ~ John Kenneth Galbraith

FireProof
Posts: 660
Joined: Thu May 05, 2011 12:15 pm

Re: Imperfect market timing still works -- if you stick with it

Post by FireProof » Tue May 14, 2019 12:22 pm

willthrill81 wrote:
Tue May 14, 2019 10:01 am
FireProof wrote:
Tue May 14, 2019 7:06 am
This thread illustrates perfectly why the average investor gets like 5% below the market return.
I'm not sure why you think this thread is representative of the average investor, who is constantly trying to predict what direction the market is going next. Contrary to what many believe, rules-based trend following strategies are based on a historical expectancy, not a vain attempt at prediction.
Everybody thinks they're better than average. And every market timer relies on back-testing. And yet, the vast majority of those unsung geniuses, many of them professional fund managers, underperform the market.

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Tue May 14, 2019 12:33 pm

FireProof wrote:
Tue May 14, 2019 12:22 pm
willthrill81 wrote:
Tue May 14, 2019 10:01 am
FireProof wrote:
Tue May 14, 2019 7:06 am
This thread illustrates perfectly why the average investor gets like 5% below the market return.
I'm not sure why you think this thread is representative of the average investor, who is constantly trying to predict what direction the market is going next. Contrary to what many believe, rules-based trend following strategies are based on a historical expectancy, not a vain attempt at prediction.
Everybody thinks they're better than average. And every market timer relies on back-testing. And yet, the vast majority of those unsung geniuses, many of them professional fund managers, underperform the market.
We should be careful to distinguish the merits of a strategy from those who claim to practice it. It wouldn't be entirely fair to say that because many people panic-sold in 2008 buy-and-hold isn't a good strategy.
In my view, the reason that market timing is so heavily discouraged is two-fold:

(1) Market timing requires big choices, and big choices create big stress, especially when so much is at stake. Large amounts of stress usually lead to poor outcomes, not only in investing, but in everything.

(2) The most vocal practitioners of market timing tend to perform poorly as investors.

Looking at (2) specifically, why do the most vocal practitioners of market timing tend to perform poorly as investors? The answer is not that they are poor market timers per se. Rather, the answer is that they tend to always be underinvested. By nature, they’re usually more risk-averse to begin with, which is what sends them down the path of identifying problems in the market and stepping aside. Once they do step aside, they find it difficult to get back in, especially when the market has gone substantially against them. It’s painful to sell something and then buy it back at a higher price, locking in a loss. It’s even more difficult to admit that the loss was the result of one’s being wrong. And so instead of doing that, the practitioners entrench. They come up with reasons to stay on the course they’re on–a course that ends up producing a highly unattractive investment outcome.
https://www.philosophicaleconomics.com/ ... ngaverage/
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
greg24
Posts: 3499
Joined: Tue Feb 20, 2007 10:34 am

Re: Imperfect market timing still works -- if you stick with it

Post by greg24 » Tue May 14, 2019 1:09 pm

willthrill81 wrote:
Mon May 13, 2019 4:00 pm
Those who are convinced that it's impossible to experience the market's upside without a one-to-one exposure to its downside will never be convinced by any analyses and arguments because it's simply incompatible with their view of how the markets work.
Are you claiming to experience the market's upside without experiencing its downside?

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Tue May 14, 2019 2:36 pm

greg24 wrote:
Tue May 14, 2019 1:09 pm
willthrill81 wrote:
Mon May 13, 2019 4:00 pm
Those who are convinced that it's impossible to experience the market's upside without a one-to-one exposure to its downside will never be convinced by any analyses and arguments because it's simply incompatible with their view of how the markets work.
Are you claiming to experience the market's upside without experiencing its downside?
I'm saying that, historically, trend following strategies have experienced most of the market's upside with relatively limited exposure to the downside, with long-term returns being very similar to buy-and-hold but with a smoother ride. I've posted the results of one of the most popular TF strategies already in this thread.

Now whether you believe that decades of this kind of performance is a fluke or a consistent effect is another matter entirely.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
HomerJ
Posts: 12785
Joined: Fri Jun 06, 2008 12:50 pm

Re: Imperfect market timing still works -- if you stick with it

Post by HomerJ » Tue May 14, 2019 2:51 pm

CULater wrote:
Tue May 14, 2019 10:09 am
Can't say that I'm all that enamored of the 50/50 BAH retirement portfolio. This ignores the risk of unexpected inflation, which kills both stocks and bonds. One example: a person retiring in 1973 with a 50/50 portfolio using the 4% rule for withdrawals ended up 9 1/2 years later with just 41% of their 50/50 portfolio left in purchasing power. Nice that they got lucky and could cut back on the dogfood diet before going bust. If they'd been 100% in stocks they would have had 32% left. Bonds helped but not hugely. Unable to determine the result for trend-following, but I venture that it would have been a LOT better than 50/50 BAH during that period of time.
Some of the 50% in bonds can be iBonds or TIPs these days.

Trend following would not have helped with inflation in the 1970s.

That's a separate risk that affects both buy and hold and trend following systems.

You said so yourself..
This ignores the risk of unexpected inflation, which kills both stocks and bonds.
So how is switching between stocks and bonds going to help?
The J stands for Jay

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Tue May 14, 2019 3:06 pm

HomerJ wrote:
Tue May 14, 2019 2:51 pm
CULater wrote:
Tue May 14, 2019 10:09 am
Can't say that I'm all that enamored of the 50/50 BAH retirement portfolio. This ignores the risk of unexpected inflation, which kills both stocks and bonds. One example: a person retiring in 1973 with a 50/50 portfolio using the 4% rule for withdrawals ended up 9 1/2 years later with just 41% of their 50/50 portfolio left in purchasing power. Nice that they got lucky and could cut back on the dogfood diet before going bust. If they'd been 100% in stocks they would have had 32% left. Bonds helped but not hugely. Unable to determine the result for trend-following, but I venture that it would have been a LOT better than 50/50 BAH during that period of time.
Some of the 50% in bonds can be iBonds or TIPs these days.

Trend following would not have helped with inflation in the 1970s.

That's a separate risk that affects both buy and hold and trend following systems.

You said so yourself..
This ignores the risk of unexpected inflation, which kills both stocks and bonds.
So how is switching between stocks and bonds going to help?
I haven't personally done an in-depth analysis of trend following in the 1970s, but a quick observation of the returns of stocks in the 1970s in PV reveals that it clearly wasn't a straight downhill march. Stocks were up anywhere from 8.45% to 37.82% in 1972, 1975, 1976, 1978, and 1979. 1973 and 1974 were the rough years for stocks during that decade, with respective losses of -18.18% and -27.81%, which would likely have been largely avoided by most trend following strategies.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Topic Author
CULater
Posts: 1916
Joined: Sun Nov 13, 2016 10:59 am

Re: Imperfect market timing still works -- if you stick with it

Post by CULater » Tue May 14, 2019 3:57 pm

HomerJ wrote:
Tue May 14, 2019 2:51 pm
CULater wrote:
Tue May 14, 2019 10:09 am
Can't say that I'm all that enamored of the 50/50 BAH retirement portfolio. This ignores the risk of unexpected inflation, which kills both stocks and bonds. One example: a person retiring in 1973 with a 50/50 portfolio using the 4% rule for withdrawals ended up 9 1/2 years later with just 41% of their 50/50 portfolio left in purchasing power. Nice that they got lucky and could cut back on the dogfood diet before going bust. If they'd been 100% in stocks they would have had 32% left. Bonds helped but not hugely. Unable to determine the result for trend-following, but I venture that it would have been a LOT better than 50/50 BAH during that period of time.
Some of the 50% in bonds can be iBonds or TIPs these days.

Trend following would not have helped with inflation in the 1970s.

That's a separate risk that affects both buy and hold and trend following systems.

You said so yourself..
This ignores the risk of unexpected inflation, which kills both stocks and bonds.
So how is switching between stocks and bonds going to help?
Well, trend following often switches between stocks and T-Bills, not bonds. And you can also time your bond investments to switch into T-bills as well.
Faced with the choice between changing one’s mind and proving that there is no need to do so, almost everyone gets busy on the proof. ~ John Kenneth Galbraith

User avatar
HomerJ
Posts: 12785
Joined: Fri Jun 06, 2008 12:50 pm

Re: Imperfect market timing still works -- if you stick with it

Post by HomerJ » Tue May 14, 2019 4:00 pm

willthrill81 wrote:
Tue May 14, 2019 3:06 pm
HomerJ wrote:
Tue May 14, 2019 2:51 pm
CULater wrote:
Tue May 14, 2019 10:09 am
Can't say that I'm all that enamored of the 50/50 BAH retirement portfolio. This ignores the risk of unexpected inflation, which kills both stocks and bonds. One example: a person retiring in 1973 with a 50/50 portfolio using the 4% rule for withdrawals ended up 9 1/2 years later with just 41% of their 50/50 portfolio left in purchasing power. Nice that they got lucky and could cut back on the dogfood diet before going bust. If they'd been 100% in stocks they would have had 32% left. Bonds helped but not hugely. Unable to determine the result for trend-following, but I venture that it would have been a LOT better than 50/50 BAH during that period of time.
Some of the 50% in bonds can be iBonds or TIPs these days.

Trend following would not have helped with inflation in the 1970s.

That's a separate risk that affects both buy and hold and trend following systems.

You said so yourself..
This ignores the risk of unexpected inflation, which kills both stocks and bonds.
So how is switching between stocks and bonds going to help?
I haven't personally done an in-depth analysis of trend following in the 1970s, but a quick observation of the returns of stocks in the 1970s in PV reveals that it clearly wasn't a straight downhill march. Stocks were up anywhere from 8.45% to 37.82% in 1972, 1975, 1976, 1978, and 1979. 1973 and 1974 were the rough years for stocks during that decade, with respective losses of -18.18% and -27.81%, which would likely have been largely avoided by most trend following strategies.
The 50/50 person only experienced 9% and 14% losses from those drops. Are you sure the trend following strategies (OF THE TIME) would have gotten out before losing 10%-20% on a 100% stocks portfolio? AND gotten back in again quick enough after the bottom to make a difference?

Sure, using today's trend-following strategies, you may have gotten out super quick and missed most of the drop, and then gotten back in super quick as well to catch the rebound before it climbed back up again. but that's because today's strategies were DERIVED using the data from 1973-1974. So, of course, today's strategies would do well against past drops.
The J stands for Jay

User avatar
HomerJ
Posts: 12785
Joined: Fri Jun 06, 2008 12:50 pm

Re: Imperfect market timing still works -- if you stick with it

Post by HomerJ » Tue May 14, 2019 4:05 pm

CULater wrote:
Tue May 14, 2019 3:57 pm
HomerJ wrote:
Tue May 14, 2019 2:51 pm
CULater wrote:
Tue May 14, 2019 10:09 am
Can't say that I'm all that enamored of the 50/50 BAH retirement portfolio. This ignores the risk of unexpected inflation, which kills both stocks and bonds. One example: a person retiring in 1973 with a 50/50 portfolio using the 4% rule for withdrawals ended up 9 1/2 years later with just 41% of their 50/50 portfolio left in purchasing power. Nice that they got lucky and could cut back on the dogfood diet before going bust. If they'd been 100% in stocks they would have had 32% left. Bonds helped but not hugely. Unable to determine the result for trend-following, but I venture that it would have been a LOT better than 50/50 BAH during that period of time.
Some of the 50% in bonds can be iBonds or TIPs these days.

Trend following would not have helped with inflation in the 1970s.

That's a separate risk that affects both buy and hold and trend following systems.

You said so yourself..
This ignores the risk of unexpected inflation, which kills both stocks and bonds.
So how is switching between stocks and bonds going to help?
Well, trend following often switches between stocks and T-Bills, not bonds. And you can also time your bond investments to switch into T-bills as well.
Short-term bonds is also an answer to inflation. Which bonds are you using in that 1973 example above? 10-year treasuries? So a 50/50 buy and hold person with shorter-term bonds would not have done as bad you say.

You guys are looking backwards at past years, using trend-following strategies devised to protect against those exact years. You're fooling yourselves if you think this will for sure protect you going forward. It might, but it might not.

You don't know that "imperfect market timing still works". It may work. I hope it does for you.
The J stands for Jay

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Tue May 14, 2019 4:28 pm

HomerJ wrote:
Tue May 14, 2019 4:00 pm
willthrill81 wrote:
Tue May 14, 2019 3:06 pm
HomerJ wrote:
Tue May 14, 2019 2:51 pm
CULater wrote:
Tue May 14, 2019 10:09 am
Can't say that I'm all that enamored of the 50/50 BAH retirement portfolio. This ignores the risk of unexpected inflation, which kills both stocks and bonds. One example: a person retiring in 1973 with a 50/50 portfolio using the 4% rule for withdrawals ended up 9 1/2 years later with just 41% of their 50/50 portfolio left in purchasing power. Nice that they got lucky and could cut back on the dogfood diet before going bust. If they'd been 100% in stocks they would have had 32% left. Bonds helped but not hugely. Unable to determine the result for trend-following, but I venture that it would have been a LOT better than 50/50 BAH during that period of time.
Some of the 50% in bonds can be iBonds or TIPs these days.

Trend following would not have helped with inflation in the 1970s.

That's a separate risk that affects both buy and hold and trend following systems.

You said so yourself..
This ignores the risk of unexpected inflation, which kills both stocks and bonds.
So how is switching between stocks and bonds going to help?
I haven't personally done an in-depth analysis of trend following in the 1970s, but a quick observation of the returns of stocks in the 1970s in PV reveals that it clearly wasn't a straight downhill march. Stocks were up anywhere from 8.45% to 37.82% in 1972, 1975, 1976, 1978, and 1979. 1973 and 1974 were the rough years for stocks during that decade, with respective losses of -18.18% and -27.81%, which would likely have been largely avoided by most trend following strategies.
The 50/50 person only experienced 9% and 14% losses from those drops. Are you sure the trend following strategies (OF THE TIME) would have gotten out before losing 10%-20% on a 100% stocks portfolio? AND gotten back in again quick enough after the bottom to make a difference?

Sure, using today's trend-following strategies, you may have gotten out super quick and missed most of the drop, and then gotten back in super quick as well to catch the rebound before it climbed back up again. but that's because today's strategies were DERIVED using the data from 1973-1974. So, of course, today's strategies would do well against past drops.
Well the simple 200 DMA, basically the same as a 10 MMA, strategy was being used back in the 1950s, so it was definitely a strategy being used by some in the 1970s. I don't have the exact data on how a 10 MMA strategy would have performed back then handy, but this chart from Philosophical Economics displays its charted performance (green line) relative to a static 85/15 AA (black line). It's not easy to read, and my apologies if you go blind trying to do so, but it appears that the green line was above the black line throughout all of the 1970s. I'm not sure what the maximum drawdown of each were during that decade, but it appears that 10 MMA did not dip in the 1973-1974 period like the 85/15 did. Perhaps someone else can more readily provide a detailed analysis of that period.

Image
https://www.philosophicaleconomics.com/2016/02/uetrend/
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Forester
Posts: 79
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Imperfect market timing still works -- if you stick with it

Post by Forester » Tue May 14, 2019 4:56 pm

I think it's normal for such strategies to give up 10% or so before exiting stocks. The tyranny of negative compounding means that it only takes an 11% rise to recover a 10% loss, but a bear market 40% loss requires a recovery of 67%.

FlyerJack
Posts: 19
Joined: Fri Feb 27, 2015 4:56 pm

Re: Imperfect market timing still works -- if you stick with it

Post by FlyerJack » Tue May 14, 2019 5:08 pm

steve roy wrote:
Tue May 14, 2019 12:15 am
Fidelity Investments investigated what worked best, and discovered that their FORGETFUL investors (the people who FORGOT they had investments at Fidelity ... or were dead) happened to be the most successful investors.
I agree with your message, but check this out regarding the “legend of the Fidelity study”: https://www.google.com/amp/s/www.nytime ... s.amp.html
That was the conclusion of a study by the investment giant Fidelity, according to a 2014 article on Business Insider. The article relayed the transcript of a Bloomberg program in which the well-known money manager Jim O’Shaughnessy said that people who had forgotten that they had accounts outperformed everyone else.

Fidelity, which has received inquiries about the study ever since, without knowing why, told me this week that it had never produced such a study.
Apocryphal study aside, the article ends with a great quote:
Even if it’s not true, it’s actually true.

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Tue May 14, 2019 5:27 pm

FlyerJack wrote:
Tue May 14, 2019 5:08 pm
steve roy wrote:
Tue May 14, 2019 12:15 am
Fidelity Investments investigated what worked best, and discovered that their FORGETFUL investors (the people who FORGOT they had investments at Fidelity ... or were dead) happened to be the most successful investors.
I agree with your message, but check this out regarding the “legend of the Fidelity study”: https://www.google.com/amp/s/www.nytime ... s.amp.html
That was the conclusion of a study by the investment giant Fidelity, according to a 2014 article on Business Insider. The article relayed the transcript of a Bloomberg program in which the well-known money manager Jim O’Shaughnessy said that people who had forgotten that they had accounts outperformed everyone else.

Fidelity, which has received inquiries about the study ever since, without knowing why, told me this week that it had never produced such a study.
Apocryphal study aside, the article ends with a great quote:
Even if it’s not true, it’s actually true.
So the moral of the story is that false stories that fit the desired narrative are fine, but decades of real data are inadequate to demonstrate that an alternative strategy is at least plausible?
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

FlyerJack
Posts: 19
Joined: Fri Feb 27, 2015 4:56 pm

Re: Imperfect market timing still works -- if you stick with it

Post by FlyerJack » Tue May 14, 2019 5:39 pm

No, I don’t believe that’s the intended moral. I think the moral would be something like: Obsessing over market fluctuations encourages tinkering, and tinkerers have generally done worse over the long term than those who just kept contributing (and rebalancing) in accordance with their investment policy statement. :happy

But mostly I just wanted to point out a fun fact I’d discovered regarding the oft-mentioned Fidelity study.

User avatar
willthrill81
Posts: 9668
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Imperfect market timing still works -- if you stick with it

Post by willthrill81 » Tue May 14, 2019 5:46 pm

FlyerJack wrote:
Tue May 14, 2019 5:39 pm
No, I don’t believe that’s the intended moral. I think the moral would be something like: Obsessing over market fluctuations encourages tinkering, and tinkerers have generally done worse over the long term than those who just kept contributing (and rebalancing) in accordance with their investment policy statement. :happy

But mostly I just wanted to point out a fun fact I’d discovered regarding the oft-mentioned Fidelity study.
Thanks for relaying that this study was fictitious. I'm really not surprised.

I agree that trend followers as a group tend to be tinkerers, but the same can be said of many buy-and-hold advocates on this forum as well, 'permanently' adding this asset class and subtracting that one, which they do on a fairly regular basis. This seems to be especially the case among those whom I refer to as the 'die-hard modern-portfolio theory advocates'. But again, we must carefully distinguish a strategy's merits from those who purportedly adopt it.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Post Reply