My trend following strategy and experience

[hilink73]

I finally added the last of the OP article indicators, Corporate Earnings Growth (Total Return EPS). The OP article gives it a 0.8% contribution to strategy return. The indicator triggers when TR EPS annualized growth rate is negative.

[...]

-B

Thanks for this detailed info.
Are those contributions of the indicators cumulative? Or just per chosen strategy?

[BlueEars]

I finally added the last of the OP article indicators, Corporate Earnings Growth (Total Return EPS). The OP article gives it a 0.8% contribution to strategy return. The indicator triggers when TR EPS annualized growth rate is negative.

Currently, this metric is 4.2% and rapidly falling, based on Shiller data as of Sep 2019. Data for the past 12 months:

• Oct-2018: 22.62%
• Nov-2018: 22.61%
• Dec-2018: 22.50%
• Jan-2019: 21.50%
• Feb-2019: 20.16%
• Mar-2019: 18.41%
• Apr-2019: 16.15%
• May-2019: 14.35%
• Jun-2019: 12.55%
• Jul-2019: 9.50%
• Aug-2019: 6.81%
• Sep-2019: 4.20%

Straight-line forecasting the next three months from the prior nine months:

• Oct-2019: 2.78%
• Nov-2019: 0.38%
• Dec-2019: -2.8% (triggered)

...

This is interesting to me. I have the latest Shiller CAPE report for Nov 15th. I'm not quite sure how you calculated the earnings growth rates above. My take on that would be to use the "real earnings" column and, for example, for Aug 2019 I would compute Aug_2019/Aug_2018 = 134.63/130.54 = 1.031 or 3.1% real earnings growth.

From the Nov 15th Shiller report we get the latest data for Sept 2019. So I would assume at least a 2 month lag in this data availability and possibly a 3 month lag. In other words, I could not get that Sept 2019 earnings data to be useful until maybe Nov 2019 if I grabbed it directly off the SP reporting (I am guessing here) or maybe not until Dec 2019 if depending on the Shiller data update. It turns out this latest Shiller report for Nov 15 was unusually tardy in that the previous report was issued in August (see viewtopic.php?f=10&t=294657&e=1&view=unread#unread).

I would guess the markets are very sensitive to the latest earnings data. Not sure that the SP500 price behavior hasn't already incorporated that information well ahead of waiting for a Shiller report (in contrast to, say, Treasury yield curve data). FWIW, my initial go at this did not seem to indicate my model would benefit from this earnings growth data series.

[Mactheriverrat]

My hats off to willthrill for throwing this thread out there. He's taken a few hits for his non-boglehead approach.

My ideas and charts are as follows.

On the right side of the green lines one should be in index funds
On the right side of the red lines one should be in bond funds.
https://stockcharts.com/h-sc/ui?s=%24%2 ... =703150627

and if one trades stocks - I use Guppy MMA's for finding trend changes.
https://www.youtube.com/watch?v=kjfW2BaeldU&t=291s

IMHO!

[Barsoom]

This is interesting to me. I have the latest Shiller CAPE report for Nov 15th. I'm not quite sure how you calculated the earnings growth rates above.

It is detailed in this article from Philosophical Economist, which is linked inside the first linked article in the OP. See Introducing the Total Return EPS Index: A New Tool for Analyzing Fundamental Equity Market Trends. The article includes Excel formulas to create the metric from the Shiller spreadsheet.

Note that Shiller just updated his spreadsheet to add his own Total Return Price and Total Return CAPE, but his method isn't the same as the Philosophical Economist and returns a slightly different result. Also, the Shiller method doesn't attempt to calculate an isolated share count to derive an EPS. There are similarities in Shiller's Total Price calculation, but he doesn't apply a "mean" factor that Philosophical Economist does.

For the growth rate, I used a simple percent change from the current value to the value one year ago ( (new-old)/old ).

After reading the article, I'd be interested in your insights regarding the difference in Shiller's Total Return Price and Philosophical Economist's.

-B

[Barsoom]

Sorry, I got busy this afternoon and only really addressed your question of how I calculated the numbers. I just reread your entire post.

I have the latest Shiller CAPE report for Nov 15th. I'm not quite sure how you calculated the earnings growth rates above.

I didn't really calculate earnings growth, I calculated earnings per share growth based on the Philosophical Economist method. Philosophical Economist called the indicator "Real S&P 500 EPS Growth (yoy)" in his bullet lists, but referred to it as "corporate earnings growth" in the body of the article, and I probably used that shorthand reference instead of the fuller descriptive one. Sorry.

From the Nov 15th Shiller report we get the latest data for Sept 2019. So I would assume at least a 2 month lag in this data availability and possibly a 3 month lag.

Yes, that data seems to be quarterly. All I did was use the Excel Trend function to straight-line forecast the next three months using the prior nine months.

It turns out this latest Shiller report for Nov 15 was unusually tardy in that the previous report was issued in August (see viewtopic.php?f=10&t=294657&e=1&view=unread#unread).

I know. I posted to that thread.

I assume he either 1) lost his interns for the summer, or 2) needed the extra time to merge his "alternate CAPE" date set with this one, because it now provides both the original CAPE and the Total Return CAPE. I hope that now that this is done, he will go back to providing a monthly updated spreadsheet.

Philosophical Economist always preferred the Total Return CAPE.

-B

[Forester]

Emerging markets & US Agg bonds as "canary indicators" for US (& global) equities.

Using a fast momentum filter (1, 3, 6 and 12 months) on VWO (emerging markets) & BND (US agg bonds)

https://papers.ssrn.com/sol3/papers.cfm ... id=3212862

[Forester]

More on US agg bond negative momentum as a prelude to poor equity performance.

https://allocatesmartly.com/vigilant-a ... w-keuning/

[garlandwhizzer]

From the article:

A word of caution

This strategy only holds one asset at any given time, which historically has resulted in extremely high returns, but could also lead to extremely high portfolio volatility and potential for loss.

There is also the problem that 100% of the portfolio is changed as often as every month. This generates huge trading costs and frictions which are of course totally neglected in the backtesting model. Likewise in a taxable account it generates huge short term capital gains which carry the same high tax burden as ordinary income in contrast to long term capital gains which receive preferential tax treatment for those who hold assets longer than 12 months. Backtesting models carefully ignore these considerations. In addition they ignore the increase complexity and work involved in making 100% changes in the portfolio as often as a monthly basis. There is also the point that past strong returns on this or any other backtesting model may not reliably translate into future strong returns. The future of markets has a very limited degree of reliable predictably.

Backtesting models are intellectually fascinating and get those who derive them promotions in academia and huge income for those who create, market, and sell them in the form of funds/management. It is very easy to understand why they are generated so frequently. Literally hundreds of investment factors and countless trading strategies have been identified. Their major real effects of all this research have been to generate promotions for those academics who "discover" them from data mining and to generate exorbitant income for those who create, market, and sell them to the public.

It is quite simple to construct a winning portfolio with data mining given assumptions of no trading costs, no trading frictions, no management costs, no marketing costs, and especially in the case of factors cost-free long/short portfolios that automatically magnify positive results by 100%. Anyone who can do arithmetic can do that over any time period. The evidence that any of these models have worked consistently in practice rather than in these unrealistic models from which they are derived is scant. It seems to me that a bit of skepticism is in order when viewing each new piece of the newly discovered magic secret investing sauce which seems to appear with increasing frequency as time passes.

People very much prefer certainty even if it is wrong to admitting uncertainty up front. Romans and other ancients studied sacrificed sheep entrails the night before a great battle to see if the gods would lead them on to victory. Since the general or a priest knowing the general's will was doing the interpretation, the answer was usually yes if the general actually wanted the battle and no if he didn't want it. Still in spite of this ruse, soldiers maintained their faith in these omens year after year for centuries. Those who seek predictability about the future are often much happier with false certainty than with true uncertainty. IMO there is some degree of parallel between the reading of sheep entrails and the ever increasing production of investing secret sauces from academics and the fund industry. It is up to the Individual investor to separate the wheat from the chaff, but most of us should keep in mind that our brains are wired to magnify wheat and neglect chaff.

Garland Whizzer

[willthrill81]

There is also the problem that 100% of the portfolio is changed as often as every month. This generates huge trading costs and frictions which are of course totally neglected in the backtesting model.

Most months, I don't have to make any trades at all. When I do, it might take 10 minutes. It's not at all burdensome. And I encounter no trading costs or other frictions when I do so.

Likewise in a taxable account it generates huge short term capital gains which carry the same high tax burden as ordinary income in contrast to long term capital gains which receive preferential tax treatment for those who hold assets longer than 12 months.

In general, I would not recommend this strategy for a taxable account and have said so in this thread. I have also said that one's stock position could be effectively zeroed out through the use of futures contracts and no tax implications if an investors wanted to implement such a strategy with a taxable account. That involves some costs, but these are likely much less than capital gains taxes.

There is also the point that past strong returns on this or any other backtesting model may not reliably translate into future strong returns. The future of markets has a very limited degree of reliable predictably.

Of course. The future could look very different from the past. But that issue also applies to buy-and-hold and any other strategy we can devise. Buy-and-hold of index funds only 'guarantees' that the investor's returns will closely approximate those of the index less the expense ratio; it says nothing about what the performance of the underlying index will be. There are historic instances of stock markets falling to zero and others languishing for decades, and the could happen again, even on a global scale.

I'm not saying that you believe or are insinuating this, but there seems to be a perception among many that TSM or global stocks will somehow provide investors with the returns they need and that any deviation from that is a literal gamble. That is obviously false; stocks come with no such guarantee, and a given buy-and-hold investor may never be rewarded for taking on the risk associated with stocks.

As noted in the OP, part of my motivation for this strategy is to attempt to reduce 'deep tail risk' and use a strategy that I firmly believe that I can stick with. I fully recognize and accept that my returns may be significantly lower than if I implemented a buy-and-hold strategy of the 3-fund portfolio or something akin to it. Most investors are likely best served with the standard BH advice. But I am not convinced that such advice is appropriate for me. And I've never once recommended my strategy to anyone else.

[garlandwhizzer]

willthrill81 wrote:

As noted in the OP, part of my motivation for this strategy is to attempt to reduce 'deep tail risk' and use a strategy that I firmly believe that I can stick with. I fully recognize and accept that my returns may be significantly lower than if I implemented a buy-and-hold strategy of the 3-fund portfolio or something akin to it. Most investors are likely best served with the standard BH advice. But I am not convinced that such advice is appropriate for me. And I've never once recommended my strategy to anyone else.

1+

I totally agree with this. If it's right for willthrill81, that's the bottom line. If your primary goal is to reduce deep tail risk I believe this strategy will very likely do so. I've lived through a few deep and very trying bear markets and know quite well how frightening it can be. However my strong basic faith in the future of capitalism and especially of the US prevents me from panic selling. Of course having a big slug of quality bonds helps as well. So far I've made it through pretty well, but there's always a chance that the next one will get me. I'm willing to take that chance. That's what right for me.

A hidden benefit of such deep tail risk prevention strategy is that in the depths of a black swan event/deep bear market, unlike most investors, you have preserved your investment capital. If you're emotionally willing to buy risk assets at fire sale prices you can do so. Most investors at such moments of general panic either do not have the necessary liquidity to invest, or if they do have it, the courage to pull the trigger. As Graham said during the GD, those with initiative have no money and those with money have no initiative. These incredibly difficult moments however are precisely the times where opportunity for huge future investment outperformance is greatest. Warren Buffett often does this. When the market is fully priced like now he often hordes huge positions in cash/ST Treasury in anticipation of future opportunities to buy quality equity assets marked down massively in price during bad times. Buffett did that in a deal with Goldman Sachs which was desperate for cash in the GR 2008-9.

The choice is out there for reducing severe market risk--high allocation to totally liquid quality bonds kept in reserve, or willthrill81"s approach. I believe both will substantially reduce losses in severe downturns and preserve assets. These approaches may in turn offer a rare opportunity in the depths of severe bear markets to increase risk assets and very likely obtain substantial outsized returns going forward. As Warren Buffett says, you don't have to pick the exact bottom the bear market. All you have to do is to buy a quality asset, a company with sound finance and a long term sustainable advantage over competitors, at a price that is considerably less than it's actually worth. Lots of these opportunities are available at such times and few, if any, at other times.

Garland Whizzer

[willthrill81]

willthrill81 wrote:

As noted in the OP, part of my motivation for this strategy is to attempt to reduce 'deep tail risk' and use a strategy that I firmly believe that I can stick with. I fully recognize and accept that my returns may be significantly lower than if I implemented a buy-and-hold strategy of the 3-fund portfolio or something akin to it. Most investors are likely best served with the standard BH advice. But I am not convinced that such advice is appropriate for me. And I've never once recommended my strategy to anyone else.

1+

I totally agree with this. If it's right for willthrill81, that's the bottom line. If your primary goal is to reduce deep tail risk I believe this strategy will very likely do so. I've lived through a few deep and very trying bear markets and know quite well how frightening it can be. However my strong basic faith in the future of capitalism and especially of the US prevents me from panic selling. Of course having a big slug of quality bonds helps as well. So far I've made it through pretty well, but there's always a chance that the next one will get me. I'm willing to take that chance. That's what right for me.

A hidden benefit of such deep tail risk prevention strategy is that in the depths of a black swan event/deep bear market, unlike most investors, you have preserved your investment capital. If you're emotionally willing to buy risk assets at fire sale prices you can do so. Most investors at such moments of general panic either do not have the necessary liquidity to invest, or if they do have it, the courage to pull the trigger. As Graham said during the GD, those with initiative have no money and those with money have no initiative. These incredibly difficult moments however are precisely the times where opportunity for huge future investment outperformance is greatest. Warren Buffett often does this. When the market is fully priced like now he often hordes huge positions in cash/ST Treasury in anticipation of future opportunities to buy quality equity assets marked down massively in price during bad times. Buffett did that in a deal with Goldman Sachs which was desperate for cash in the GR 2008-9.

The choice is out there for reducing severe market risk--high allocation to totally liquid quality bonds kept in reserve, or willthrill81"s approach. I believe both will substantially reduce losses in severe downturns and preserve assets. These approaches may in turn offer a rare opportunity in the depths of severe bear markets to increase risk assets and very likely obtain substantial outsized returns going forward. As Warren Buffett says, you don't have to pick the exact bottom the bear market. All you have to do is to buy a quality asset, a company with sound finance and a long term sustainable advantage over competitors, at a price that is considerably less than it's actually worth. Lots of these opportunities are available at such times and few, if any, at other times.

Garland Whizzer

A very fair answer and very well said!

[rascott]

I'm using a trend following system based upon price (9 month MA) to hold leveraged ETFs (mainly TQQQ).... so the info discussed here is very helpful.

[Forester]

It's funny that people on here chase their own tails over 3.5% vs 4% withdrawal rate, sequence of returns, bond tents etc etc, when there's strong evidence that even only 10% or 20% allocated to equity trend following would likely make their retirement plans more robust. There's no good reason for the concept of a "portfolio stop loss" not to gain traction other than superstitious beliefs.

[willthrill81]

It's funny that people on here chase their own tails over 3.5% vs 4% withdrawal rate, sequence of returns, bond tents etc etc, when there's strong evidence that even only 10% or 20% allocated to equity trend following would likely make their retirement plans more robust. There's no good reason for the concept of a "portfolio stop loss" not to gain traction other than superstitious beliefs.

Our views of the world are result of our paradigms. One of the paradigms of many here is that it's impossible to reliably 'time' the market in a favorable way at all. As such, they reject the notion of stop losses, trend following, target volatility, etc. In many cases, evidence to the contrary is to be dismissed because it doesn't fit the paradigm, and evidence in support of their paradigm is held to (i.e. confirmation bias).

There are some here who seem to be open to the idea of trend following but simply find the evidence not sufficiently persuasive. One logical person can look at a given set of information and reach one conclusion, while another logical person can look at the same information and come to a different conclusion. It doesn't mean that either is being illogical; it's a matter of differences in perception, psychology, and sometimes paradigms.

It seems to be uncommon for someone to see the value in multiple very distinct strategies. Most only seem to see one strategy as being plausible and see all others as implausible.

Buy-and-hold can work very well, but it has risks. Trend following can also work very well, but it too has risks. There is no magic bullet.

[tadamsmar]

It's funny that people on here chase their own tails over 3.5% vs 4% withdrawal rate, sequence of returns, bond tents etc etc, when there's strong evidence that even only 10% or 20% allocated to equity trend following would likely make their retirement plans more robust. There's no good reason for the concept of a "portfolio stop loss" not to gain traction other than superstitious beliefs.

Is there any reason to think that trend following in tax-deferred accounts is likely to make your retirement plan less robust?

[BlueEars]

It's funny that people on here chase their own tails over 3.5% vs 4% withdrawal rate, sequence of returns, bond tents etc etc, when there's strong evidence that even only 10% or 20% allocated to equity trend following would likely make their retirement plans more robust. There's no good reason for the concept of a "portfolio stop loss" not to gain traction other than superstitious beliefs.

Is there any reason to think that trend following in tax-deferred accounts is likely to make your retirement plan less robust?

There are some caveats to using trend following that could affect retirees. A few I can think of:
1) Don't adopt a poorly designed trend following approach. Any approach should take into account that sudden declines like October 1987 can happen.
2) Be sure you can implement the methodology for the long haul. You have to follow up even if you are ill or traveling or there is some disturbing life event going on that distracts you.
3) If you are having cognitive issues perhaps move to buy-hold and/or use an advisor.

[willthrill81]

It's funny that people on here chase their own tails over 3.5% vs 4% withdrawal rate, sequence of returns, bond tents etc etc, when there's strong evidence that even only 10% or 20% allocated to equity trend following would likely make their retirement plans more robust. There's no good reason for the concept of a "portfolio stop loss" not to gain traction other than superstitious beliefs.

Is there any reason to think that trend following in tax-deferred accounts is likely to make your retirement plan less robust?

There are some caveats to using trend following that could affect retirees. A few I can think of:
1) Don't adopt a poorly designed trend following approach. Any approach should take into account that sudden declines like October 1987 can happen.
2) Be sure you can implement the methodology for the long haul. You have to follow up even if you are ill or traveling or there is some disturbing life event going on that distracts you.
3) If you are having cognitive issues perhaps move to buy-hold and/or use an advisor.

To add to #2, you must be prepared to lag a buy-and-hold approach, perhaps for many years and especially in a strong bull market.

[Barsoom]

I finally added the last of the OP article indicators, Corporate Earnings Growth (Total Return EPS). The OP article gives it a 0.8% contribution to strategy return. The indicator triggers when TR EPS annualized growth rate is negative.

[...]

-B

Thanks for this detailed info.
Are those contributions of the indicators cumulative? Or just per chosen strategy?

Sorry, I just realized I never answered your question.

In the OP articles, the indicators are individual in the sense that any one of them triggering is all it takes to trigger the exit strategy when the S&P 500 is also below its 10MMA (in the original article - I've been using 9MMA as being closer to the 200DMA). Different recession exits may be triggered by different indicators, but in the recent past recessions, all the indicators were triggered before the S&P 500 crossed its trigger.

This applies only to the original seven indicators, not the full list of additional leading indicators I added.

This article analyzes each indicator separately.

This article has nice summary tables showing the contributions and sensitivities of the individual indicators (it's the second article linked in the OP).



-B

[Barsoom]

It's funny that people on here chase their own tails over 3.5% vs 4% withdrawal rate, sequence of returns, bond tents etc etc, when there's strong evidence that even only 10% or 20% allocated to equity trend following would likely make their retirement plans more robust. There's no good reason for the concept of a "portfolio stop loss" not to gain traction other than superstitious beliefs.

Is there any reason to think that trend following in tax-deferred accounts is likely to make your retirement plan less robust?

There are some caveats to using trend following that could affect retirees. A few I can think of:
1) Don't adopt a poorly designed trend following approach. Any approach should take into account that sudden declines like October 1987 can happen.
2) Be sure you can implement the methodology for the long haul. You have to follow up even if you are ill or traveling or there is some disturbing life event going on that distracts you.
3) If you are having cognitive issues perhaps move to buy-hold and/or use an advisor.

To add to #2, you must be prepared to lag a buy-and-hold approach, perhaps for many years and especially in a strong bull market.

Correct me if I'm wrong, but would you say that the retiree world is actually very different from the saver world?

That is, the saver follows a "buy and hold" strategy for the 30+ years they are earning an income, but the retiree follows a "hold and sell" strategy during the 30 years they must live off of their savings?

As a retiree, to me "hold" also means "protect" and sell. A trend following strategy is about protecting my portfolio from anticipated drops when long-term indicators suggest a looming downturn in the economic environment. "Protecting" is important because I don't have a salary anymore to insulate me from the market recovery time.

To me, the risk as a retiree is that the downturn doesn't come and reentering the market costs the retiree the small period of growth between exit and reentry. This is mitigated to some degree by the portion of the retiree portfolio that is in equities vs bonds, meaning that the retiree my have less equity exposure than the B&H investor has.

-B

[Forester]

Pacer ETFs charge 0.60% for a straightforward equity trend strategy. Perfect to slot into a lazy Boglehead portfolio https://www.paceretfs.com/products/ptlc

[Barsoom]

Let me add another retiree thought:

Another aspect of this is what I'll call "lifestyle." As an accumulator, the question comes down to how much lifestyle one wishes to "buy" in retirement. For me, my goal was to be able to replace 100% of my income, essentially sustaining my current lifestyle into retirement without having to work for it anymore. I did this. As others choose their "number," I'd think a determinant is what kind of lifestyle they are acceptable with during retirement.

As a "protect and sell" investor, I'm motivated to protect my lifestyle against downturns, since I no longer have a salary to sustain me through market downturns. A B&H investor can accept the downturn knowing that the recovery will make them whole and the market will continue to grow. A retiree can also keep to their SWR and hope that the recovery period is short. I think the goal of a P&S investor is the hope of avoiding the downturn and then gain in the recovery to "buy" additional lifestyle.

-B

[willthrill81]

It's funny that people on here chase their own tails over 3.5% vs 4% withdrawal rate, sequence of returns, bond tents etc etc, when there's strong evidence that even only 10% or 20% allocated to equity trend following would likely make their retirement plans more robust. There's no good reason for the concept of a "portfolio stop loss" not to gain traction other than superstitious beliefs.

Is there any reason to think that trend following in tax-deferred accounts is likely to make your retirement plan less robust?

There are some caveats to using trend following that could affect retirees. A few I can think of:
1) Don't adopt a poorly designed trend following approach. Any approach should take into account that sudden declines like October 1987 can happen.
2) Be sure you can implement the methodology for the long haul. You have to follow up even if you are ill or traveling or there is some disturbing life event going on that distracts you.
3) If you are having cognitive issues perhaps move to buy-hold and/or use an advisor.

To add to #2, you must be prepared to lag a buy-and-hold approach, perhaps for many years and especially in a strong bull market.

Correct me if I'm wrong, but would you say that the retiree world is actually very different from the saver world?

That is, the saver follows a "buy and hold" strategy for the 30+ years they are earning an income, but the retiree follows a "hold and sell" strategy during the 30 years they must live off of their savings?

As a retiree, to me "hold" also means "protect" and sell. A trend following strategy is about protecting my portfolio from anticipated drops when long-term indicators suggest a looming downturn in the economic environment. "Protecting" is important because I don't have a salary anymore to insulate me from the market recovery time.

To me, the risk as a retiree is that the downturn doesn't come and reentering the market costs the retiree the small period of growth between exit and reentry. This is mitigated to some degree by the portion of the retiree portfolio that is in equities vs bonds, meaning that the retiree my have less equity exposure than the B&H investor has.

-B

It's commonly accepted that, at least in the first ten years or so of retirement, that retirees' portfolios should generally be more conservative than an accumulator's, especially a young one. Standard BH advice is to achieve this via a significant allocation to fixed income. The approach laid out here is to attempt to avoid exposure to stocks when they seem to be at greatest risk of a significant decline. So the goal might be the same, though the means are very distinct.

[Galton's Ghost]

Exactly. The first 10 years of retirement are the danger zone. Wade Pfau (I think) and others have argued that you should start retirement overwhelmingly in bonds and then slowly move to stocks as you age to protect yourself against sequence risk. But almost no retirees are going to accept moving to 70% or 80% stocks at 75 or 80. It's ridiculous.

Trend following accomplishes the same thing but in a more realistic way.

[Galton's Ghost]

From the article:

A word of caution

This strategy only holds one asset at any given time, which historically has resulted in extremely high returns, but could also lead to extremely high portfolio volatility and potential for loss.

There is also the problem that 100% of the portfolio is changed as often as every month. This generates huge trading costs and frictions which are of course totally neglected in the backtesting model. Likewise in a taxable account it generates huge short term capital gains which carry the same high tax burden as ordinary income in contrast to long term capital gains which receive preferential tax treatment for those who hold assets longer than 12 months. Backtesting models carefully ignore these considerations. In addition they ignore the increase complexity and work involved in making 100% changes in the portfolio as often as a monthly basis. There is also the point that past strong returns on this or any other backtesting model may not reliably translate into future strong returns. The future of markets has a very limited degree of reliable predictably.

Backtesting models are intellectually fascinating and get those who derive them promotions in academia and huge income for those who create, market, and sell them in the form of funds/management. It is very easy to understand why they are generated so frequently. Literally hundreds of investment factors and countless trading strategies have been identified. Their major real effects of all this research have been to generate promotions for those academics who "discover" them from data mining and to generate exorbitant income for those who create, market, and sell them to the public.

It is quite simple to construct a winning portfolio with data mining given assumptions of no trading costs, no trading frictions, no management costs, no marketing costs, and especially in the case of factors cost-free long/short portfolios that automatically magnify positive results by 100%. Anyone who can do arithmetic can do that over any time period. The evidence that any of these models have worked consistently in practice rather than in these unrealistic models from which they are derived is scant. It seems to me that a bit of skepticism is in order when viewing each new piece of the newly discovered magic secret investing sauce which seems to appear with increasing frequency as time passes.

People very much prefer certainty even if it is wrong to admitting uncertainty up front. Romans and other ancients studied sacrificed sheep entrails the night before a great battle to see if the gods would lead them on to victory. Since the general or a priest knowing the general's will was doing the interpretation, the answer was usually yes if the general actually wanted the battle and no if he didn't want it. Still in spite of this ruse, soldiers maintained their faith in these omens year after year for centuries. Those who seek predictability about the future are often much happier with false certainty than with true uncertainty. IMO there is some degree of parallel between the reading of sheep entrails and the ever increasing production of investing secret sauces from academics and the fund industry. It is up to the Individual investor to separate the wheat from the chaff, but most of us should keep in mind that our brains are wired to magnify wheat and neglect chaff.

Garland Whizzer

First, regarding trading costs and frictions:

1. The number of trades per year under this system are small.
2. The system uses some of the most liquid ETFs in the world, i.e. the bid-ask spread is basically zero

Therefore, this is an extremely cheap system to implement - if done in a tax-deferred account - probably similar to B&H.

Second, I think that you're missing a very big point. Because of the combination of economic indicator and trend screen, this strategy is in stocks ~85% of the time. So, let's assume that the whole strategy is baloney. The market is random and trend doesn't work. That would mean that Will's returns will be the same as though he invested in 85% stocks and 15% bonds. Also, since his trading costs are close to zero and he's using extremely cheap mutual funds/ETFs, he won't lose practically any more than a typical Boglehead. Again, he'll earn very close what a typical Boglehead would earn on an 85% stock, 15% bond portfolio. Not bad.

What I'm saying is that Will's "Strategy Risk" i.e. the danger that his economic indicator/trend screen is baloney, is very, very low. Maybe he doesn't want the risk level of an 85/15 portfolio, but other than that, his system is cheap and simple, much like a typical Boglehead portfolio.