Always treat others with respect and kindness, even if you disagree with them. Avoid making personal attacks or insulting others, and try to maintain a civil and constructive tone in your discussions.
- Manlobbi
Investment Strategies / Mechanical Investing
No. of Recommendations: 14
A lot time ago I suggested that those who couldn't decide between cap weight and equal weight could simply use momentum: pick the one that has done the best in the last month, and hold that one this month. This created a calendar-month-cycle timing system. Likely only suited to a tax sheltered account.
Here's a much simpler approach: Create a data series that is the ratio of the data total return series for RSP to the total return series for SPY. Do two EMAs on the ratio and use the crossovers as a timing signal for which to use.
I suggest maybe 20 days and 90 days. Using 20 and 50 worked better 1960-1999, and 20 and 120 worked better 2000 to date, but 20 and 90 seemed like a nice compromise that worked adequately in both sub-periods.
You'd think that 20 days for the "fast" EMA would lead to an extraordinary number of signals, but not really: an average of 1.37 round trips per year.
Ignoring commissions and tax, the 20-and-90 switching scheme would have beat SPY total return by 2.41%/year from Jan 1960 to Feb this year. Probably even more since then, as the megacaps have consistently soared.
That's not as fantastic as it sounds: simply holding RSP (and its historical index before introduction) would have beat SPY by 1.59%/year. Perhaps a better metric of the value added by the switch is how much it beat the average of the two choices: 1.62%/year. Still, over long periods that does add up.
I'm not saying this is the greatest system ever devised, I just thought I'd pass it on.
Some recent signals, market close on day shown.
2008-08-13 Buy RSP
2008-10-03 Buy SPY
2009-04-01 Buy RSP
2011-07-20 Buy SPY
2012-02-01 Buy RSP
2012-03-06 Buy SPY
2012-10-23 Buy RSP
2013-11-25 Buy SPY
2014-01-22 Buy RSP
2014-01-27 Buy SPY
2014-01-28 Buy RSP
2014-05-21 Buy SPY
2014-06-17 Buy RSP
2014-07-16 Buy SPY
2014-12-23 Buy RSP
2015-01-15 Buy SPY
2015-01-27 Buy RSP
2015-05-12 Buy SPY
2016-03-07 Buy RSP
2016-10-13 Buy SPY
2016-11-15 Buy RSP
2017-02-01 Buy SPY
2018-04-04 Buy RSP
2018-05-09 Buy SPY
2018-11-26 Buy RSP
2018-12-03 Buy SPY
2019-01-22 Buy RSP
2019-03-21 Buy SPY
2020-11-02 Buy RSP
2021-07-06 Buy SPY
2021-10-13 Buy RSP
2021-10-19 Buy SPY
2022-01-19 Buy RSP
2022-07-15 Buy SPY
2022-07-26 Buy RSP
2022-07-29 Buy SPY
2022-09-01 Buy RSP
2023-03-14 Buy SPY
2023-12-27 Buy RSP
2024-01-17 Buy SPY (still in effect)
Here's a quick link.
https://stockcharts.com/h-sc/ui?s=SPY%3ARSP&p=D&yr...Blue above red = hold SPY
Red above blue = hold RSP
Jim
No. of Recommendations: 0
Jim, any idea of drawdowns with this system?
No. of Recommendations: 6
Jim, any idea of drawdowns with this system?
Well, it's a long/long system, not long/cash. It only switches between "long the S&P cap weight" and "long the S&P equal weight", so the drawdowns will be extremely comparable to those.
S&P 500 nominal total return worst drawdown since 1960 = -54.85%
S&P 500 equal weight nominal total return worst drawdown since 1960 = -59.84%
Switch scheme -57.29%
As with any "long equities all the time", you'll lose half your money once in a while, on a mark-to-market basis. That's life.
Or worse.
The memento mori:
S&P 500 nominal total return worst drawdown since 1930 = -81.84%
S&P 500 equal weight nominal total return worst drawdown since 1930 = -82.87%
If you had a long/cash timing system you trusted, you could use this one to pick which thing to be long. Depending on how frequent your long/cash signals are, you might not even have to bother switching which ting you're long, just hold what you bought till the next bear signal, and it still might give you a few basis points of advantage over "only SPY when long" over the long run.
Jim
No. of Recommendations: 0
Jim,
Appreciate you sharing. Could be useful for other 401k bound workers like myself. I will look into it more in depth.
My employer happens to use Fidelity for 401k administration and added an option several years ago that allows employees to invest in a wide universe of mutual funds outside of the base options. There is a fund that mimics RSP and I have been using that for a portion of my portfolio. Adding some switching, if I can find an option to avoid the typical 90day holding period within the plan, could at least add some entertainment.
Did you happen to look at adding in 99 day to the switching? I plan to look at it but thought I would ask.
Jeff
No. of Recommendations: 11
Did you happen to look at adding in 99 day to the switching? I plan to look at it but thought I would ask. \No, I didn't check that.
The whole idea behind the 99 day is that market tops are pretty rounded...you usually have quite a while to get out of the market because it's not usually all that much lower even a few months after "the" top.
Changes in underperformance versus outperformance among gigacaps seem to be pretty sharp, more turn-on-a-dime thing, so the EMA seems to be more suited. An even shorter EMA might backtest even better in terms of CAGR discrimination on average, but nobody needs a signal that changes state too often.
=============
Speaking of market tops, has anyone else noticed the remarkable string of over 20 fresh all time highs lately on the Nasdaq Composite, while Nasdaq breadth has been solidly bad and deteriorating? The advance/decline line and new highs / new lows have both been negative for a while and on a weakening trend since about mid May.
https://stockcharts.com/h-sc/ui?s=%24NAHL&p=D&yr=0...I'm no good at identifying market tops at the time, but this divergence something that is very typical at major market tops. It's one of the few bad omens that would have saved somebody in 87, though taken by itself it would also have cried "the sky is falling" tons of times when the sky did not in fact fall.
But once things have started to look bad for a little while, it's easier to have an idea what's going on. It might be one of those years that the "no longer looks like a bull market" signal from the 99 day rule might be worth watching for. A timeout of 3 months is even more conservative/chicken.
Jim
No. of Recommendations: 3
BCC = 5 (SMA and DBE positive, NHNL negative) is generally unrewarding, particularly on a risk-adjusted basis. Since 1980, when BCC = 5 (15% of the time) the S&P has barely outperformed the interest that GTR1 imputes to cash.
Baltassar
No. of Recommendations: 5
has anyone else noticed ... <?i> ==> Definitely. The biggest megacap techs keep going up, but that's about it. Breadth divergence as bad as I've seen it in years. Small caps underperformance bad and getting worse (relatively) daily. All indexing. And the 99 day will never go down because the indexes keep inching higher.
Is adding more allocation to megacap US a disaster-tempting exercise of FOMO? Or a savvy money-following move?
The hard right edge of the chart, indeed.
FC
No. of Recommendations: 0
No. of Recommendations: 5
I'm no good at identifying market tops at the time, but this divergence....
..... is the reason why I am currently holding more cash than ever.
If it is correct that tops are rounded and there is plenty of time -- a month or two or three -- to sell, why head for the door now?
“Far more money has been lost by investors preparing for corrections, or trying to anticipate corrections, than has been lost in corrections themselves.” – Peter Lynch
"More Money Has Been Lost Avoiding Risk Than at the Point of a Gun."
"Until you realize taking a beating is a normal part of long-term investing, you’ll hurt the overall performance of your portfolio.
Staying with your strategy during a pullback is difficult, and it never gets any easier."
"When the market finally corrects, observable shifts will begin to occur. There is no need to anticipate, predict, or forecast." -- Chris Ciovacco
No. of Recommendations: 4
If it is correct that tops are rounded and there is plenty of time -- a month or two or three -- to sell, why head for the door now?
A) I am not that sure about the (always?) rounded tops.
B) I think there is good evidence for the Presidential Cycle. If so we are close to the end --- plus with the current candidates/situation there is a chance for this cycle ending more dramatic than usual.
C) I wouldn't know what to invest in right now.
D) It doesn't really move the needle whether that huge stash of cash earns some % more or not than 5% Treasuries for 1/4 year or even 1/2 year --- and how great are currently the chances for it to earn 10%, 15% or 20%?
E) Having been self employed and worrying my whole life about financial survival I finally realized that unless I do very stupid things I don't have to have worry about money for the rest of my life.
A+B+C+D+especially E = There is nothing more important for me than Rule #1.
No. of Recommendations: 1
I forgot one important point: As so many investors losing money creates a huge psychological problem for me. I know me well enough that I am extremely unwilling to sell stuff with a loss. So even if "the top is rounded" and "there is 1 or 2 months or more time to get out" I would NOT easily get out on the way down, but against my better knowledge I would constantly hope for a counter reaction big enough that I could sell my positions without a loss (as is since quite a while the case with my little remaining BABA+KMX+HSY). So on the way down I would wait and wait and wait......
No. of Recommendations: 18
A) I am not that sure about the (always?) rounded tops.
Definitely not always.
The *average* major market top is pretty rounded. e.g., the average market return from the average price in the week of an absolute cyclical top in the S&P, to the average price in a later week, is about -8% five months after the peak. But an average can hide a lot of sins. Sometimes the market drops a lot, shortly after a recent top.
There are some big rapid dips shortly after fresh recent highs, but the good news is that they tend to be particularly short lived: 1987 and 2020, for example, so it's arguably best to simply ignore those surprise bears. You didn't see it coming, you were long through the drop, but it probably won't last, so just stay long for the bounce. (on average again!) A "no new highs lately" signal will keep you long through the sudden drop, but also long during at least the first part of the strong rebound. After that, it should be pretty obvious you're in a strong market with or without any timing signal.
A tip:
If you do find yourself still long at a big pointy bear market bottom, whether it was a slow predictable bear or a sudden surprise plunge, as soon as you think the bottom is in it's a fabulous time to overweight small caps till about 14 months from the bottom.
e.g., smallest 120 stocks in the VL 1700 set. Not relying on a few lucky picks.
CAGR for 14 months starting Octob 2002: 130% versus 33% for SPY.
CAGR for 14 months starting March 2009: 444% versus 62% for SPY.
CAGR for 14 months starting April 2020: 200% versus 51% for SPY.
To increase the pleasantness, I might limit those 120 to the 60 with the highest ROE, just so it leans to the better quality firms. Not much difference on this measure, but higher on average through the years, better on almost any metric when considering not just the rebounds.
Bookmark this post for next time there's metaphorical blood in the streets and capitulation in every dentist's portfolio : )
Jim
No. of Recommendations: 10
Jim wrote:
If you do find yourself still long at a big pointy bear market bottom, whether it was a slow predictable bear or a sudden surprise plunge, as soon as you think the bottom is in it's a fabulous time to overweight small caps till about 14 months from the bottom.This is good advice... most of the time. There is a caveat. The amount of forward returns are very much tied to the degree of drawdown in the previous bear market.
These are the dates I employed for testing this:
10/12/1990 BULL
11/11/1991 BEAR
10/8/1998 BULL
11/7/1999 BEAR
10/9/2002 BULL
11/8/2003 BEAR
3/9/2009 BULL
4/8/2010 BEAR
3/23/2020 BULL
4/22/2021 BEAR
10/12/2022 BULL
11/11/2023 BEAR
This shows the returns and stats for this whole period with a hold of from ten to one-hundred and twenty stocks:
# to Hold ROI CAGR GSD Sharpe DDD3 BBW Stock % of Time Drawdown
10 47098.89% 20.03% 22.64 0.84 3.27% 6.35 72% 19.28% -35.78%
20 53841.65% 20.50% 20.15 0.94 3.89% 5.55 74% 19.28% -40.81%
30 50326.01% 20.26% 18.89 0.98 2.87% 7.37 73% 19.28% -38.62%
40 56227.95% 20.66% 18.16 1.03 2.00% 10.45 76% 19.28% -37.85%
50 55959.08% 20.64% 17.8 1.05 1.82% 11.49 76% 19.28% -34.72%
60 54282.88% 20.53% 17.52 1.06 1.74% 11.98 76% 19.28% -32.33%
70 51616.77% 20.35% 17.25 1.06 1.72% 12.04 77% 19.28% -32.03%
80 45528.37% 19.91% 17.07 1.05 1.81% 11.16 77% 19.28% -31.90%
90 41070.91% 19.54% 16.86 1.04 1.84% 10.83 76% 19.28% -31.94%
100 46220.05% 19.96% 17.56 1.02 1.89% 10.8 76% 19.28% -31.60%
110 45941.41% 19.94% 17.22 1.04 1.90% 10.76 77% 19.28% -30.61%
120 42208.94% 19.64% 16.9 1.04 1.89% 10.62 77% 19.28% -29.59%
The best CAGR goes to the portfolio of forty stocks. Here are the returns for each hold:
Date Signal P % G/L Dow ROI
10/12/1990 BUY L 38.98% 26.87%
10/8/1998 BUY L 55.70% 38.63%
10/9/2002 BUY L 177.72% 33.90%
3/9/2009 BUY L 831.67% 66.90%
3/23/2020 BUY L 337.23% 81.88%
10/12/2022 BUY L 6.40% 17.55%
Notice that from the most recent bear market bottom small caps have significantly under-performed.
No. of Recommendations: 5
As so many investors losing money creates a huge psychological problem for me. I know me well enough that I am extremely unwilling to sell stuff with a loss. Yeah, that is a problem. A big behavioral problem.
Funny, I am the exact opposite. When I am looking to sell something to raise cash or whatever, I prefer to sell something that is showing a loss.
Actually, the way to avoid that behavioral problem is to just invest in an index fund.
"
You should be concerned about portfolio results, not position results."
"Markowitz long ago told investors that the primary focus should be on designing and managing the portfolio. That’s a message that still resonates and it reminds us not to obsess over the parts."
"you are generally best off if you cut your losses and let your winners run. "
"The irregularity of the markets exists to shake out market players that cannot handle losses. Those that cannot handle losses had unrealistic expectations. Markets are perverse, and they suck in amateurs near peaks, and the amateurs leave near troughs. They help provide the excess performance of the best."
========================================
"Risk control is the best route to loss avoidance. Risk avoidance, on the other hand, is likely to lead to return avoidance as well." -- Howard Marks
Better to sell at a 10% loss than to hang on through a 30% or 40% loss.
One way to do simple risk control is to mechanically follow a timing rule. There are regular posts here ("Recent BCC Signals") on one set.
A simple 10 or 12 month Simple Moving Average of the S&P500 is a well-known and readily available timing signal which will let you avoid most large bear markets.
This is simple to calculate on your own, or there are plenty of web sites that publish it. For example:
https://www.advisorperspectives.com/dshort/updates...
No. of Recommendations: 5
Jim wrote: If you do find yourself still long at a big pointy bear market bottom, whether it was a slow predictable bear or a sudden surprise plunge, as soon as you think the bottom is in it's a fabulous time to overweight small caps till about 14 months from the bottom.
...
This is good advice... most of the time. There is a caveat. The amount of forward returns are very much tied to the degree of drawdown in the previous bear market.
Very much so. I did say "big" pointy bear market bottom, but I did not emphasize it. You're right. This is the sort of thing that might make sense only after a *major* market bottom. Very big fall. In recent decades, only the three times I mentioned.
And don't switch to small caps until good confirmation signs that the bottom has been seen. You don't really want to hold those small caps starting at a "false" bottom on the way down the bear.
What surprised me in this particular effect is the relatively longevity of the outperformance. 14 months is a nice stretch.
For example, the other well known rule of what to hold just after a major bottom is to buy the absolute worst junk that has fallen the most: the stuff you'd normally short. But that works for only a matter of weeks. If the small cap rule works for longer, it's that much easier to wait for really good confirmation after the bottom, you can take your time to make sure that you are climbing the second stroke of the "V".
Oddly enough I forgot to test 1987, so that's an out of sample validation of my "about 14 months" thesis. The big outperformance didn't last as long that time: 9 or 10 months would have been optimal in terms of total return. The 120 VL smallest caps beat S&P CAGR by 20-25% for 9-10 months. The top 60 of those small caps by ROE beat SPY by a CAGR of 39-47%. The aggregate advantage was still there at the 14 month mark but was eroding because the small caps were underperformers around months 10-14. (in fact steadily underperformers until from 10 to 38 months after the '87 bottom, end 1990).
Jim
No. of Recommendations: 7
# to Hold ROI CAGR GSD Sharpe DDD3 BBW Stock % of Time Drawdown
10 47098.89% 20.03% 22.64 0.84 3.27% 6.35 72% 19.28% -35.78%
20 53841.65% 20.50% 20.15 0.94 3.89% 5.55 74% 19.28% -40.81%
30 50326.01% 20.26% 18.89 0.98 2.87% 7.37 73% 19.28% -38.62%
40 56227.95% 20.66% 18.16 1.03 2.00% 10.45 76% 19.28% -37.85%
50 55959.08% 20.64% 17.8 1.05 1.82% 11.49 76% 19.28% -34.72%
60 54282.88% 20.53% 17.52 1.06 1.74% 11.98 76% 19.28% -32.33%
70 51616.77% 20.35% 17.25 1.06 1.72% 12.04 77% 19.28% -32.03%
80 45528.37% 19.91% 17.07 1.05 1.81% 11.16 77% 19.28% -31.90%
90 41070.91% 19.54% 16.86 1.04 1.84% 10.83 76% 19.28% -31.94%
100 46220.05% 19.96% 17.56 1.02 1.89% 10.8 76% 19.28% -31.60%
110 45941.41% 19.94% 17.22 1.04 1.90% 10.76 77% 19.28% -30.61%
120 42208.94% 19.64% 16.9 1.04 1.89% 10.62 77% 19.28% -29.59%
All the returns (CAGR) are within +-0.5%, which to me seems statistically meaningless.
Elan
No. of Recommendations: 3
Elan wrote: All the returns (CAGR) are within +-0.5%, which to me seems statistically meaningless.
I'd agree which was my point in sharing it. I believe one of the key reasons for this is the percent of time invested - just 19%. The rest of the time it is in cash and that is the same return no matter how many you are holding.
No. of Recommendations: 0
All the returns (CAGR) are within +-0.5%, which to me seems statistically meaningless.
# to hold...
Sorry, the returns from what?
Jim
