Skip to main content
No. of Recommendations: 27
As the years go by I find myself more interested in steadier gains with lower volatility. This post uses the profit latch concept with screens to achieve this goal.

I decided that I would be content if my screens went up a minimum of 4% a year (not counting dividends). Therefore, beginning in January, at the end of each month if a screen has a gain of at least 4% for the year the proceeds are not reinvested but put into CDs/money market until the following January.

But what screens to use? You want one with a good CAGR in case you end up in the screen for more months than anticipated. And, since the profit latch is driven by the volatility pump, a screen with high GSD is also wanted.

Using the standard SIPro screens at keelix.com I ran the screens (monthly, five stocks) for the 12 years from 1998 through 2009. I then selected the six screens with the highest total CAGR + GSD. The six screens were:

LowPS+
RS-100
WKvoom
POG
PIH...Naked
Up5X3

Each month the year-to-date return was calculated. If 4% or over the screen went to cash. The average time in stocks over the 12 years was about three months per year.

Here are the annual returns, not counting dividends and interest on cash:

Year 98 99 00 01 02 03 04 05 06 07 08 09
LowPS+ 8 31 59 4 8 8 15 14 5 5 8 14
RS-100 16 31 5 6 11 6 10 17 12 9 6 13
WKvoom 10 6 -27 6 19 17 11 6 15 8 5 4
POG 5 30 27 13 11 6 5 4 10 10 5 -47
PIH_Naked 15 33 36 4 5 10 12 6 9 6 14 10
Up5X3 6 6 7 5 7 4 9 12 7 7 7 27

As can be seen, there is an occasional year where a screen goes down before it goes up and then stays down. However, a blend of the six screens (a reasonable maximum of 30 stocks at the beginning of the year) smoothes things out.

The annual returns for the six-screen blend:

Year 98 99 00 01 02 03 04 05 06 07 08 09
10 23 18 6 10 8 9 10 10 7 8 3

Add in 3% for dividends and interest and you have a CAGR of 13% with annual returns ranging from 6% to 26%.

A couple of other examples:

Applied to the Trout on Sale screen your average holding period was 2.8 months per year with a CAGR of 9% plus interest and dividends.

I tested a screen of volatile stocks in a volatile industry. Using the Semiconductor industry each month I screened for the five stocks with the highest beta. Over the last 12 years the average holding period was 3.25 months per year with a CAGR of 11% plus interest.

DB2
Print the post Back To Top
No. of Recommendations: 6
Interesting. So, we've arrived at the point where, instead of maximizing returns by staying in during bullish periods and getting out during bearish periods, we're looking to hit RBI singles and protect capital using positive volatile screens for a limited time.

Does it strike anyone else that we've come full circle?

FC
Print the post Back To Top
No. of Recommendations: 0
Interesting. I would take advantage of what I think I know about seasonality and start my profit latch year on Nov. 1 of each year instead of Jan. 1.

Tricky
Print the post Back To Top
No. of Recommendations: 0
DrBob,

It seems you chose the 4% arbitrarily....perhaps since it is often the recommended withdrawal rate for retirement nest eggs.

Did you try or consider a 'mound of toast' on this value?

Thanks,

-Ernie
Print the post Back To Top
No. of Recommendations: 4
nteresting. So, we've arrived at the point where, instead of maximizing returns by staying in during bullish periods and getting out during bearish periods, we're looking to hit RBI singles and protect capital using positive volatile screens for a limited time.

Does it strike anyone else that we've come full circle?


I'm not sure what you mean by 'full circle'. Have we been here before? At any rate, it's not often you can compound your money at 13% while being out of the market 75% of the time.

Hitting those home runs can be pretty sexy, but to quote Jimmy Soul:

"If you wanna be happy for the rest of your life
Never make a pretty woman your wife"

DB2
Print the post Back To Top
No. of Recommendations: 1
Interesting. I would take advantage of what I think I know about seasonality and start my profit latch year on Nov. 1 of each year instead of Jan. 1.

That would most likely also work. I wouldn't start in May. :-)

DB2
Print the post Back To Top
No. of Recommendations: 3
It seems you chose the 4% arbitrarily....perhaps since it is often the recommended withdrawal rate for retirement nest eggs.
Did you try or consider a 'mound of toast' on this value?


I started out with a 10% limit to keep the nest egg growing. Soon it became apparent that
1) one often exceeded the limit and
2) in difficult years such as 2002 or 2008 many screens never reached the 10% level so that one rode the screen for the entire year. Not a good way to reduce risk.

I didn't do a mound-of-toast; the 4% level was sort of eye-balled. I may look at some volatile VL screens and do a toast mound with them.

DB2
Who likes using volatility to reduce risk
Print the post Back To Top
No. of Recommendations: 1
it's not often you can compound your money at 13% while being out of the market 75% of the time.

I am more comfortable in the market for fear of losing out to a declining dollar and inflation and taxes. I find it difficult being our of the market 2 months during the summer let alone 9 months. To each his own, I guess.
Print the post Back To Top
No. of Recommendations: 1
Thanks Dr Bob,

Can I assume that you combine them to smooth out the returns as there are safety in numbers?

Otherwise, why not simply go with LowPS+ since the CAGR is 14.57% (at 4% monthly max) or 16.52% (at 5% monthly max)?

It would take someone with some huge stones to invest in one stock. Since at least ome of those screens, LowPS+, may only have one selection in a particular month.

This may be, stress MAY BE, a good choice for someone looking for relative safety, but yet still score some decent returns? Was that the original intent?

The ingenuity (sp?) on this board never ceases to amaze me.
Print the post Back To Top
No. of Recommendations: 10
Each month the year-to-date return was calculated. If 4% or over the screen went to cash. The average time in stocks over the 12 years was about three months per year.

Here are the annual returns, not counting dividends and interest on cash:


Year 98 99 00 01 02 03 04 05 06 07 08 09
LowPS+ 8 31 59 4 8 8 15 14 5 5 8 14
RS-100 16 31 5 6 11 6 10 17 12 9 6 13
WKvoom 10 6 -27 6 19 17 11 6 15 8 5 4
POG 5 30 27 13 11 6 5 4 10 10 5 -47
PIH_Naked 15 33 36 4 5 10 12 6 9 6 14 10
Up5X3 6 6 7 5 7 4 9 12 7 7 7 27



Am I the only one who is baffled by this? If a screen is sold once it has gained more than 4% for the year, how can any screen make much more than 4% in any year? How could LowPS+, for example, earn 59% in 2000 if you sold it when its gain was 4%?

Elan
Print the post Back To Top
No. of Recommendations: 3
Elan asked:
Am I the only one who is baffled by this? If a screen is sold once it has gained more than 4% for the year, how can any screen make much more than 4% in any year? How could LowPS+, for example, earn 59% in 2000 if you sold it when its gain was 4%?

DrBob wrote:
Each month the year-to-date return was calculated. If 4% or over the screen went to cash.

If it wasn't for the fact that he was only checking it monthly it would be impossible, but since he checks monthly this means it gained far than the requisite 4% between the end of the previous month and the end of the month he exited.

I could easily backtest this myself, but it is difficult for me to only look once a month. My profit latch will be based on daily data, and therefore, like you said, it will end up being very close to 4%. In other words, mine will be like someone set a limit order and exited when it hit the threshold which could be in the middle of the data.
Print the post Back To Top
No. of Recommendations: 1
Following this would certainly take a lot of discipline. As do all Mechanical Screens.

It's something I struggle with all the time.
Print the post Back To Top
No. of Recommendations: 0
...we're looking to hit RBI singles and protect capital using positive volatile screens for a limited time.

To a certain extent we've come full circle. But part of the "problem" is saving for retirement is so individual that no one system will fit everyone.

I realized a long time ago that I don't have to hit home runs but a bunch of singles and doubles to do what I want to do in retirement. Primarily because of the shear volume of money I can save and because of my lifestyle choices. So for me, its more important to save capital, or more precisely to decrease volatility/risk. Some one with 4 kids to put through college and late to the retirement savings game might have a different take.

JLC
Print the post Back To Top
No. of Recommendations: 1
Can I assume that you combine them to smooth out the returns as there are safety in numbers?
Otherwise, why not simply go with LowPS+ since the CAGR is 14.57% (at 4% monthly max) or 16.52% (at 5% monthly max)?


Blending is used to smooth things out. It also cushions the loss when one of your screens goes into a tailspin and ends up down 44% for the year.

It would take someone with some huge stones to invest in one stock. Since at least ome of those screens, LowPS+, may only have one selection in a particular month.

You are right, and one wouldn't want to select a screen that had a tendency have few picks.

This may be, stress MAY BE, a good choice for someone looking for relative safety, but yet still score some decent returns? Was that the original intent?

Yep. Since one ends up out of the market a majority of the time the volatility and risk drop considerably.

DB2
Print the post Back To Top
No. of Recommendations: 1
If a screen is sold once it has gained more than 4% for the year, how can any screen make much more than 4% in any year?

Zee has it right. You only make changes to the portfolio on a monthly basis. Thus, looking at the venerable RS26wk, in January 1986 the screen returned 12%. At rebalance time that portion of your funds would be transfered to cash (where it then proceeded to earn 12%, IIRC).

If the profit latch has not clicked, use the proceeds to buy the stock picks for February. At your next rebalance check to see what your change is for the year. Rinse and repeat.

For me this requires no more patience or fortitude than usual, since I never rebalance in the middle of a period.

DB2
Print the post Back To Top
No. of Recommendations: 1
Trying out the seasonal variation idea you might test VL blends you like in Jamies' backtester. It gives average monthly/quarterly returns
at the bottom right of the page. So I tried a Jan. and Nov. start getting an average of 21% in just two quarters using either start month
with EGPLOW-PE, 1-3, and PIH-CSO-simple, 1-3, both quarterlies.

Other blends with more stocks to lower volatility and rebalanced monthly might give better results.

rrjjgg
Print the post Back To Top
No. of Recommendations: 4
Here is a look at the screen profit latch with six high volatility VL screens over 23 years, 1986-2008. The screens used (all with high CAGR and GSD) were:

Olap_RS52keystone
PLOW_PE2MOD
PST_5-10
RSEGOL
RSWEPS
SOS_KJ

At the 4% cut-off the CAGR was 10%, not including dividends and interest. As the cut-off was raised the CAGR increased, to eventually converge on that of the straight blend.

Cut-off level CAGR
4% 10%
6 10
8 12
10 13
12 14
14 16

It should be noted that by the 10% level the return for 2008 was a -26%.

DB2
Print the post Back To Top
No. of Recommendations: 2
An update for this year's experience with the 4% screen profit latch. The returns for each screen were checked at the end of each month. When the YTD return was greater than 4% the screen went to cash for the rest of the year.

The results:

LowPS+ 8% 1 month
RS-100 8 3
WKvoom 7 3
POG 9 2
PIH...Naked 9 4
Up5x3 8 2

The average return was 8% while being in the market only 2.5 months.

DB2
Print the post Back To Top
No. of Recommendations: 8
An update for this year's experience with the 4% screen profit latch. The returns for each screen were checked at the end of each month. When the YTD return was greater than 4% the screen went to cash for the rest of the year.

The results:

LowPS+ 4% 1 month
RS-100 -13 12
WKvoom 17 2
POG 8 1
PIH...Naked 5 3
Up5x3 9 3

The average return was 5% while being in the market an average of only 3.6 months.

DB2
Print the post Back To Top
No. of Recommendations: 5
This year's market exposure with the screen profit latch is over. The returns for each screen were calculated at month's end. If greater than 4% the proceeds for that screen went to cash for the rest of the year.

LowPS+ 11% 1 month
RS-100 20 1
WKVoom 7 1
POG 5 3
PIH...Naked 13 1
Up5x3 8 1

The average return was 11% While being in the market an average of only 1.3 months.

DB2
Print the post Back To Top
No. of Recommendations: 6
Here's a summary of the post-discovery performance of the 4% screen profit latch. The returns for each screen were checked at the end of each month. When the YTD return was greater than 4% the screen went to cash for the rest of the year.

Ave months
in market
2010 9% 2.7
2011 18 2.3
2012 9 1.7
2013 8 1.8
2014 6 7.3
2015 -6 5.8
2016 8 5.8

One disappointing year. The CAGR during the seven-year bull market was 7.2% which underperformed SPY, but still met the goal for a retirement account of exceeding 4% growth while being out of the market a majority of the time. Average time-in market, 4 months/year) with a GSD(annual) of 7.

DB2
Print the post Back To Top
No. of Recommendations: 7
When the YTD return was greater than 4% the screen went to cash for the rest of the year.

Why would you do this? This is nuts.

This is bad for the same reason that Indexed Universal Life (IUL) policies are crappy. You put a ceiling on your gains but there is no floor on your losses. But you NEED the 10% & 15% and 20% gains to cover the 10% & 20% losses.
IUL's are even better than this--at least they have ceilings like 12% and not a measly 4%.

2010 9% 2.7
2011 18 2.3
2012 9 1.7
2013 8 1.8
2014 6 7.3
2015 -6 5.8
2016 8 5.8


Here's your problem. Your observation period is when the market is only a bull market.
In a bear market the loss is going to be ~20%, and it will take you FIVE up years to cover that loss at 4%/yr. Just in time to get hit with the next bear market.
Print the post Back To Top
No. of Recommendations: 3
Here's your problem. Your observation period is when the market is only a bull market.

Ray, in the post that started the thread we find these performance numbers for bear market years (same six screens):

S&P
2000 18% -11%
2001 6 -13
2002 10 -23
2008 8 -38
.
Return 49% -63%

And there's the reason for the profit latch -- not to mention that being out of the market three-quarters of the time reduces one's exposure to Black Swan events.

DB2
Print the post Back To Top
No. of Recommendations: 6
in the post that started the thread we find these performance numbers for bear market years (same six screens) ... And there's the reason for the profit latch

"The answer you like the most is the one you should trust the least."

Philosophically:
Look, the error here is Confirmation Bias. You are looking at data that confirms your thesis but not looking at or for anything that dis-confirms it. That's backwards. The thing that harms you financially is losses. So what you should first be looking at is "what would DISconfirm this thesis? What are the things that would make this lose my money?"

The very first thing to be looking for in a theory/screen/strategy is the potential for losses -- because losses are what kill you and take you out of the game.



Mathematically:
Put aside any emotional attachment you have for the thesis and just look at it dispassionately from the math standpoint.
We know that we get both losses and gains. We also know that losses in the 10%-20% range are unavoidable and happen once every year or two.

Back-of-envelope math is close enough.
If we take a 20% loss, how long will it take to recover from it? If we cap our gain at 4%, the simple math shows that it will take 5 years to get back to even. Alternatively, 5 good years in a row get wiped out by one bad(ish) year. Eight to ten good years in a row will get wiped out by one seriously bad year. (It's actually longer, because the simple math underestimates the true recovery time.)

That right there is a very strong argument that a profit-latch is not going to be a successful tactic. Still needs to be proved or disproved, of course, but it will take a strong rebuttal in the face of this simple math.

Frankly, just looking at this math is enough for me to reject any such thesis out of hand. Like the way I feel (as an engineer) about perpetual motion machine proposals --- the chances of "this proposed one" actually working is so vanishingly small that it's not worth the effort to look at it closely.


****************************
Ha! Made me look!
Historical data:

My IUL/SP500 spreadsheet comes in handy once again. (I am _so_ glad that Dave D. induced me to do this.) https://www.dropbox.com/s/cbzvg74iyeyfwt6/SPX-monthly-1950-2...

Note that it does all rolling 12-month periods, not just calendar year periods.

You can backtest a profit latch by going to the IUL section (AA1 thru AA5) and set the floor to -65%, cap to 4% (this is the profit latch), fee to .000001%, load to 0%. Then the inital values (T1 thru T7) $100,000 Initial, $0 add, W/D date 1/1/2033).
Set Start Date (R1) to whatever you like.

Then look at the final value (AA804). You can compute the CAGR from that. Or copy W817 to AD817 and change $W to $AC.

Start date 1/1/1975 grows (read: shrinks) $100,000 to $83,156, for CAGR of -0.5%.
Start 1/1/1950: -1.0% CAGR
Start 1/1/1995: -3.4% CAGR
Start 1/1/2000: -3.8% CAGR

You can get the same information from the "Return Distrib" tab, set the floor to -65% and the cap to 4% (profit latch) (E10 & F10).

Then look at the stats at row 795.
65% of all rolling 12-month periods have return greater than 4%. The profit patch caps all these to 4%.
If you *don't* do the latch, the average (not CAGR, arithmetic average) annual gain is 8.8%.
If you *do* do the latch, the average annual gain is 2.8%.
This is the stats on gains only, not counting the periods with a loss.
Print the post Back To Top
No. of Recommendations: 3
65% of all rolling 12-month periods have return greater than 4%. The profit patch caps all these to 4%....

Ray, please read the thread. You will see that gains are not capped at 4%. Here is the distribution of the annual returns (without dividends or interest, which would add another 2-3%) for the last 19 years. This was a period that included two bear markets, one lasting three years, where the S&P declined 35-40%.

20 to 25% x
15 to 19 xx
10 to 14 xxxx
5 to 9 xxxxxxxxxx
0 to 4 x
-5 to -1
-9 to -6 x

With 2% for interest and dividends, the CAGR is 11%. Over the same period the S&P was at only 6%.

Note that it does all rolling 12-month periods, not just calendar year periods....

I know you're not a fan of seasonal effects, but they have been shown to exist over many decades. This strategy does not start on "all months".

DB2
Print the post Back To Top
No. of Recommendations: 1
Ray, please read the thread. You will see that gains are not capped at 4%. Here is the distribution of the annual returns (without dividends or interest, which would add another 2-3%) for the last 19 years. This was a period that included two bear markets, one lasting three years, where the S&P declined 35-40%.


20 to 25% x
15 to 19 xx
10 to 14 xxxx
5 to 9 xxxxxxxxxx
0 to 4 x
-5 to -1
-9 to -6 x


I think I don't understand this. I'm assuming that you looks at the profit latch only at the end of each month. Thus, it's possible that on the month you breached the 4% profit mark for the year, the month actually ended with a 5, 6 or 7% profit. But you seem to claim that almost half the time the latch triggered with a profit of 10% or (a lot) more. That seems extremely unlikely.

Elan
Print the post Back To Top
No. of Recommendations: 3
But you seem to claim that almost half the time the latch triggered with a profit of 10% or (a lot) more. That seems extremely unlikely.

Remember that
- the screens were chosen for their volatility
- the cycle starts in January, a favorable time for stocks especially smaller ones favored by SIPro screens.

Here are the returns for 2011 where the strategy returned 18% while being in the market less than three months.

Jan Feb Mar
LowPS+ 3 15 18%
PIH Naked -1 1 43 43%
POG 9 9%
RS-100 -2 -2 11 7%
Up5x3 1 -2 15 14%
WKvoom -1 18 17%

Even without the PIH...Naked outstanding March results, the year far exceeds a 4% limit. For the curious, the PIH stock picks for that March were GMCR, ASGR, VRUS, CSU and BAS. A good month to be in biotech stocks (IBB, a biotech ETF was up over 7%) and that month Green Mountain Coffee Roasters announced a strategic alliance with Starbucks sending the stock up 58%.

DB2
Print the post Back To Top
No. of Recommendations: 1
Okay, you made me go back and look at the original post, so I was sure what the rule was.

So basically you aren't capping the gain at 4%, you are capping the gain to 0% for the remaining 12 months after some month gets to 4% YTD gain. Yeah, that's not something that would be easy to model in my spreadsheet. Could be done with a little bit of work, though, given the inclination.

I still have a couple of issues, though.

One is as Elan said, that it's unlikely to see a huge rise over and above 4%. Granted, a high volatility screen might well be able to throw a large up month that you can capture, but it seems unlikely for that to be multiples of 4%.
If you are, say, 9 months into the year and still are under 4% YTD, how likely is it that the next month will be massively more than 4% YTD? The more months there are in the denominator so far, the larger the next month must be to get over the 4% threshold.

If several months along you are at 3% YTD, you need 1% more to get to 4%. Which means that next month must have a 33% larger return than you got so far in the entire year-to-date. Not impossible, but would seem to be infrequent.

Another is some math that doesn't seem right to me. For example, if you get 6% for 6 months YTD and then 0% (cash) for the next 6 months, isn't the CAGR 3%? (I don't really know -- I could probably make an argument either way. I *do* know that the math gets tricky when computing on sequential series's of returns, and that it's very easy to screw it up and use the wrong math formulas.)

Another is the asymmetry of returns and the cap (bail out latch). A loss of 20% requires a gain of 25% to get back to even. It is not possible to cap the losses as much lower than 20%, so you can't avoid those. But you are capping your gain to something just above 4% YTD, maybe as high as 6%-8% if you are lucky. It is going to take a long time to recover from a largish loss.
I think this method only "works" because it hasn't yet been hit by a large loss that it must recover from. Every strategy appears to work great if it hasn't yet taken a loss.

But the biggest issues I have are:

1) Cherry-picking the start calendar date. You are looking at only the starting month that gives the return you like, while ignoring all the other months. Well, guess what? Out of 12 months, there is ALWAYS one that will be better than the other twelve.
It's like me being a smart-ass at the grocery store when total bill comes out to be X dollars and zero cents. I always say "Wow! What are the chances of THAT happening?" (right before my wife smacks me on the head and apologies for me to the clerk. And says to them, "Don't laugh, it just encourages him." And then I say, "OWow, that's only a one in one hundred chance." Then she smacke me again.)

2) Cherry-picking the screens.
and
3) Data-dredging. You trained the strategy on the same data you tested with. If you backtest a strategy over the period which is the period that you tuned the parameters for, then Of course it will show outstanding returns.

4) Making the assumption that breaks will always go in your favor. It's never a good assumption to assume that Murphy's Law will not hit you.

5) "the average holding period was 3.25 months per year with a CAGR of 11%." This is too good to be true. It is simply not believable that there exists a simple methodology that will have returns like this. What is more likely is that there is a huge unseen risk that you just don't see, and that just hasn't happened yet. The internets are full of strategies that give enormous returns --- until they don't ... and lose all their money.
Print the post Back To Top
No. of Recommendations: 5
One is as Elan said, that it's unlikely to see a huge rise over and above 4%.

Depends upon your definition of 'huge' but I posted the distribution of returns for the past two (almost) decades.

If several months along you are at 3% YTD, you need 1% more to get to 4%. Which means that next month must have a 33% larger return than you got so far in the entire year-to-date. Not impossible, but would seem to be infrequent.

Not hard at all to have a good/great screen return in November and December. Heck, IWM (the whole freakin' Russell Index of 2000 small cap stocks) was up almost 14% this November and December.

Another is some math that doesn't seem right to me. For example, if you get 6% for 6 months YTD and then 0% (cash) for the next 6 months, isn't the CAGR 3%?

If you start the year with $100K and end it with $106K then your return is 6%, n'est-ce pas? Hint: figure out the CAGR for the first six months. It's a lot higher than 6%. :-)

I think this method only "works" because it hasn't yet been hit by a large loss that it must recover from. Every strategy appears to work great if it hasn't yet taken a loss.

Sure, everything is subject to failure. However, 2000-2003 and 2008 were navigated with success.

Cherry-picking the start calendar date. You are looking at only the starting month that gives the return you like, while ignoring all the other months. Well, guess what? Out of 12 months, there is ALWAYS one that will be better than the other twelve.

I never tested any other start month, so a January start may or may not be optimal. Who knows, there may be a mound of toast for the start month and this strategy would do better starting in December or some other month.

Cherry-picking the screens.

It wasn't cherry-picking in that I didn't/haven't tried other screens with this strategy. However, the whole idea behind the profit latch is that it is driven by what I called the volatility pump. The screens were chosen, using the Keelix backtester, for high GSD.

Data-dredging. You trained the strategy on the same data you tested with.

I'm not sure 'data dredging' is the right term here. At any rate, we have years of both pre- and post-discovery data.

Making the assumption that breaks will always go in your favor.

No assumptions were made or harmed in the making of this screen.

"the average holding period was 3.25 months per year with a CAGR of 11%." This is too good to be true.

We shall see over the next 18 years. I'll keep everyone posted.
[Geez, how many years have I been posting here?]

DB2
Print the post Back To Top
No. of Recommendations: 5
Geez, how many years have I been posting here?

18 years, if you're like me.

OMG!!!

Elan
Print the post Back To Top
No. of Recommendations: 6
Geez, how many years have I been posting here?
---
18 years, if you're like me.
OMG!!!


Me too. Of course, there are worse places to spend your time.

And, I have certainly enjoyed the journey.

DB2
Print the post Back To Top
No. of Recommendations: 5
The first month of 2017 (with SPY up 1.8%) is now in the books, and the screen profit latch had these results:

Low PS+ 1%
PIH...Naked -3
POG 9
RS-100 1
Up5x3 2
WKvoom 11

Thus, one-third of the strategy's money now goes to cash (POG and WKvoom).

For the curious, here are the stocks that were held in WKvoom:

KEM 5%
CC 20
PGNX 3
WLDN 19
TTMI 9

DB2
Print the post Back To Top
No. of Recommendations: 6
2017 is now in the books for the monthly screen profit latch system. Time to go play golf....

March was a good month for the Nasdaq and for these SIPro screens. Here are the monthly returns for the six screens

Jan Feb Mar End
Low PS+ 1.01 x 1.01 x 1.12 14%
PIH...Naked 0.97 x 0.98 x 1.15 9%
POG 1.09 9%
RS-100 1.01 x 1.01 x 1.04 6%
Up5x3 1.02 x 0.98 x 1.07 7%
WKvoom 1.11 11%
.
Average return 9.3%
Ave time in market 2.3 months

DB2
Print the post Back To Top
No. of Recommendations: 5
As the years go by I find myself more interested in steadier gains with lower volatility. This post uses the profit latch concept with screens to achieve this goal.

I decided that I would be content if my screens went up a minimum of 4% a year (not counting dividends). Therefore, beginning in January, at the end of each month if a screen has a gain of at least 4% for the year the proceeds are not reinvested but put into CDs/money market until the following January.

But what screens to use? You want one with a good CAGR in case you end up in the screen for more months than anticipated. And, since the profit latch is driven by the volatility pump, a screen with high GSD is also wanted.

Using the standard SIPro screens at keelix.com I ran the screens (monthly, five stocks) for the 12 years from 1998 through 2009. I then selected the six screens with the highest total CAGR + GSD.


For 2018 it was necessary to change the screen list to fit with jamie's backtester. The six screens with the highest CAGR + GSD:

78RPM
Gentle Screamers
PIH...Naked
POG
Turnarounds
Up5

The results this year, while not spectacular, did outperform SPY (-4.6%).

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec End
78RPM 0.99 x 1.21 21%
GS 1.01 x 1.06 7%
PIH...Naked 0.99 x 1.09 8%
POG 0.85 x 1.07 x 1.04 x 1.03 x 1.06 x 0.92 x 1.01 x 1.05 x 0.94 x 0.92 x 0.95 x 0.89 -26%
Turnarounds 0.82 x 1.10 x 0.96 x 1.00 x 1.03 x 1.00 x 1.06 6%
Up5 0.96 x 1.07 x 0.96 x 0.98 x 0.96 x 1.01 x 0.98 x 1.08 x 0.98 x 0.99 x 0.96 x 0.93 -14%
.
Average return 0.3%
Ave time in market 6.2 months

DB2
Print the post Back To Top
No. of Recommendations: 3
Here's a summary of the performance of the 4% screen profit latch since it was introduced at the end of 2009. The returns for each screen are checked at the end of each month. When the YTD return is greater than 4% the screen goes to cash for the rest of the year.

Ave months
in market
2010 9% 2.7
2011 18 2.3
2012 9 1.7
2013 8 1.8
2014 6 7.3
2015 -6 5.8
2016 8 5.8
2017 9 2.3
2018 0 6.2

One disappointing year. The CAGR during the nine years was 6.6% which underperformed SPY, but still met the goal for a retirement account of exceeding 4% growth while being out of the market a majority of the time. Average time in the market was 4 months/year.

DB2
Print the post Back To Top
No. of Recommendations: 16
The results this year, while not spectacular, did outperform SPY (-4.6%).

It's probably worth noting that this year was down a lot, fairly steadily and substantially, near the end.
"Down near the end" is the ideal situation to make such a strategy look particularly good.


If you want a 4% return, how's this for a strategy:
74% of portfolio in one year T-bills paying 2.58%
26% of portfolio in a particular stock with a yield of 8.04%, which I personally consider safe and currently underpriced/mispriced.

Well, sure, it's way riskier.
But hey, at least there's a cap on how much you can lose!

Jim

Uncle Jim's crazy pick: Brookfield Property, currently trading at $17.67.
NYSE ticker BPY.
If you're American, there is a REIT-registered economically equivalent ticker BPR. Easier paperwork.
Yeah, yeah, I know nobody here is interested in individual stock picks. I like messing with people's heads.
I talk about quant methods on the single-stock boards.
Print the post Back To Top
No. of Recommendations: 1
If you want a 4% return, how's this for a strategy:
74% of portfolio in one year T-bills paying 2.58%
26% of portfolio in a particular stock with a yield of 8.04%, which I personally consider safe and currently underpriced/mispriced.


Well, the screen profit latch strategy was on the equity side of things with the (unstated) assumption that one already had a fixed income allotment. How would you screen for particular stocks to meet your safe and underpriced/mispriced goals?

By the way, thanks for the BPY suggestion. I bought a position last month and it is currently up 12%.

DB2
Print the post Back To Top
No. of Recommendations: 5
. How would you screen for particular stocks to meet your safe and underpriced/mispriced goals?

I'm not sure such a screen is possible.
If it is, I'm not sure how you'd do it.
The closest I got was debt less than five times earnings, lots of earnings (low P/E), and a high ROE.

It took me several years of reading before I was comfortable opening a decent sized position in Brookfield Asset Management.
(the parent company which owns a huge fraction of Brookfield Property)

I own both at the moment, but mostly the parent.
BAM the parent grows faster and is the better long term bet, but BPY the kid is probably more undervalued at the moment.

Jim
Print the post Back To Top
No. of Recommendations: 1
BAM the parent grows faster and is the better long term bet, but BPY the kid is probably more undervalued at the moment.

I struggle with this all the time in my closed-end fund (CEF) investments. Two CEFs that have similar NAV performance, althoug maybe one has a bit better NAV performance....
But the other one is currently selling at a deeper discount (read: comparatively undervalued). So is it better to go with the undervalued one or not?

For those unfamiliar with CEFs, the NAV (net asset value) is the value of the holdings, and the price is whatever the CEF itself trades at.

Boulder Growth & Income: (BIF)
NAV is 12.43, current price is 10.36.

34% of BIF's holdings are BRK. Wouldn't you like to buy BRK at a 16% discount? OTOH, until and unless BIF liquidates, your P/L is detemined by the changes in price of BIF regardless of how BRK performs.
Print the post Back To Top
No. of Recommendations: 5
I own both at the moment, but mostly the parent.

I got lucky that way either in the late 1960s or early 1970s. A salesman tried to high-pressure me into buying shares in a couple of mutual funds. They came to my door in the early evening in those days -- fortunately they gave that up.

I found it difficult to evaluate those mutual funds in those days (no World Wide Web, etc.), so I did not buy any of them. The wife of my boss was a stock broker in those days and I asked her what I should do. She said to buy the parent company who sold those mutual funds, but to hold it only for a short time. I did so, and the stock price of the parent company went up much more than any of their mutual funds. Then she told me to sell, which I did. That was when the investigation broke that these management companies were churning the portfolios of their mutual funds, and getting the commissions for that. Then the parent company I had dropped a lot; fortunately after I sold it. It does not suit my style for investing, so I did not do that again.

She was an interesting broker though. She had me buy shares in a southeast asia gold mining company for $2 or $3 a share, it doubled or something like that and she sold it. I was disappointed when it went up to something like $15, but a day or two later, it dropped to pennies. What had happened was that the company had made a one-time sale of a lot of properties that they had given up on, and that was a revenue spike that many people who saw the revenue spike, but did not know what caused it, thought meant they had struck gold. They did not. And my broker hit that one right.

Another was the time she had me buy Playboy stock and then sell it a week or two later. I bought the stock, the company announced they had just gotten a license to open a Casino in Atlantic city, it jumped in price, and then sold it before it had a chance to drop again. Hugh Hefner may have been a better businessman than Donald Trump, but not a whole lot better.
Print the post Back To Top
No. of Recommendations: 1
It's probably worth noting that this year was down a lot, fairly steadily and substantially, near the end. "Down near the end" is the ideal situation to make such a strategy look particularly good.

One never knows what the future will bring, but over the last 21 years the strategy had only one down year (and that was a -6%). It actually looks best during down markets.

SPY
2000 18% -10%
2001 6 -12
2002 10 -22
2008 8 -37
2018 0 - 4

DB2
Print the post Back To Top
No. of Recommendations: 0
<It took me several years of reading before I was comfortable opening a decent sized position in Brookfield Asset Management.
(the parent company which owns a huge fraction of Brookfield Property)

I own both at the moment, but mostly the parent.
BAM the parent grows faster and is the better long term bet, but BPY the kid is probably more undervalued at the moment.

Jim>

...of probably no interest whatsoever is that BAM shows up deep on both the LOWDV and SHORT_ALTMAN_Z screens.
:-)Shawn
Print the post Back To Top
No. of Recommendations: 0
<...of probably no interest whatsoever is that BAM shows up deep on both the LOWDV and SHORT_ALTMAN_Z screens.>
Ooops--sorry! About a year ago [but not now].
:-)Shawn
Print the post Back To Top
No. of Recommendations: 3
...of probably no interest whatsoever is that BAM shows up deep on both the LOWDV and SHORT_ALTMAN_Z screens.

It's probably worth noting that the Altman Z score isn't meaningful for non-manufacturing firms, and was never intended to be.

Perhaps also worth noting is that LOWDV isn't a short screen.
It's short for "low double value": low price to book and low price to cash flow.
It's not actually a very good long screen, but I don't imagine a stock showing up on it is a sign of a problem.

Jim
Print the post Back To Top
No. of Recommendations: 0
Jim, any other conservative ideas like this?

If you want a 4% return, how's this for a strategy:
74% of portfolio in one year T-bills paying 2.58%
26% of portfolio in a particular stock with a yield of 8.04%, which I personally consider safe and currently underpriced/mispriced.

Well, sure, it's way riskier.
But hey, at least there's a cap on how much you can lose!

Jim

Uncle Jim's crazy pick: Brookfield Property, currently trading at $17.67.
NYSE ticker BPY.
If you're American, there is a REIT-registered economically equivalent ticker BPR. Easier paperwork.
Yeah, yeah, I know nobody here is interested in individual stock picks. I like messing with people's heads.
I talk about quant methods on the single-stock boards.
Print the post Back To Top
No. of Recommendations: 4
"If you want a 4% return, how's this for a strategy:
74% of portfolio in one year T-bills paying 2.58%
26% of portfolio in a particular stock with a yield of 8.04%, which I personally consider safe and currently underpriced/mispriced."
Uncle Jim's crazy pick: Brookfield Property, currently trading at $17.67. NYSE ticker BPY.



Well, I hope that BPY is safer and under-valued better than KHC (Kraft Heinz).
BPY: P/E ratio 10.12 Yield 6.62%
KHC: P/E ratio 4.09 Yield 5.2% (At yesterday's close).

Because right now:
KHC 34.55 -13.63 (-28.29%)


Big danger in one-stock portfolios.
Print the post Back To Top
No. of Recommendations: 2
[OT]
Greetings

Kraft Heinz downgraded multiple times after earnings announcement filled with bad news

https://www.marketwatch.com/story/kraft-heinz-downgraded-mul...


Warren Buffett’s Berkshire Hathaway stock falls as big bet on Kraft Heinz sours

https://www.marketwatch.com/story/warren-buffetts-berkshire-...

... re: some recently discussed companies on the MI board.

:-)Shawn
Print the post Back To Top
No. of Recommendations: 2
This post uses the profit latch concept with screens....

I decided that I would be content if my screens went up a minimum of 4% a year (not counting dividends). Therefore, beginning in January, at the end of each month if a screen has a gain of at least 4% for the year the proceeds are not reinvested but put into CDs/money market until the following January.

But what screens to use? You want one with a good CAGR in case you end up in the screen for more months than anticipated. And, since the profit latch is driven by the volatility pump, a screen with high GSD is also wanted.

Using the standard SIPro screens at keelix.com I ran the screens (monthly, five stocks) for the 12 years from 1998 through 2009. I then selected the six screens with the highest total CAGR + GSD.


In 2018 it was necessary to change the screen list to fit with jamie's backtester. The six screens with the highest CAGR + GSD:

78RPM
Gentle Screamers
PIH...Naked
POG
Turnarounds
Up5

The results for 2019, while not spectacular, did return more than the 4% goal.

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec End
78RPM 1.00 x 1.14 14%
GS 1.02 x 1.05 7%
PIH...Naked 1.06 6%
POG 1.10 10%
Turnarounds 1.02 x 0.98 x 0.92 x 0.96 x 0.95 x 1.06 x 0.86 x 0.98 x 1.10 x 1.08 x 0.95 x 1.07 -10%
Up5 1.07 7%
.
Average return 5.7%
Ave time in market 3.2 months

DB2
Print the post Back To Top
No. of Recommendations: 1
This post uses the profit latch concept with screens....

I decided that I would be content if my screens went up a minimum of 4% a year (not counting dividends). Therefore, beginning in January, at the end of each month if a screen has a gain of at least 4% for the year the proceeds are not reinvested but put into CDs/money market until the following January.

But what screens to use? You want one with a good CAGR in case you end up in the screen for more months than anticipated. And, since the profit latch is driven by the volatility pump, a screen with high GSD is also wanted.

Using the standard SIPro screens at keelix.com I ran the screens (monthly, five stocks) for the 12 years from 1998 through 2009. I then selected the six screens with the highest total CAGR + GSD.


In 2018 it was necessary to change the screen list to fit with jamie's backtester. The six screens with the highest CAGR + GSD:

78RPM
Gentle Screamers
PIH...Naked
POG
Turnarounds
Up5

For 2020 the strategy returned 2%, with Gentle Screamers being weak all year.

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec End
78RPM 1.07 7%
GS 0.94 x 0.96 x 0.88 x 0.90 x 0.99 x 1.04 x 1.00 x 1.05 x 0.99 x 0.97 x 1.09 x 0.96 -23%
PIH...Naked 1.01 x 1.05 6%
POG 1.14 14%
Turnarounds 0.92 x 0.97 x 0.66 x 1.10 x 1.16 x 1.09 x 1.00 x 1.06 x 1.04 x 1.01 x 1.15 5%
Up5 0.90 x 0.98 x 0.77 x 1.19 x 0.97 x 1.15 x 1.00 x 0.97 x 1.19 4%
.
Average return 2.2%
Ave time in market 6.0 months

DB2
Print the post Back To Top
No. of Recommendations: 3
Here's a summary of the post-discovery performance of the 4% screen profit latch. The returns for each screen were checked at the end of each month. When the YTD return was greater than 4% the screen went to cash for the rest of the year.

Ave months
in market
2010 9% 2.7
2011 18 2.3
2012 9 1.7
2013 8 1.8
2014 6 7.3
2015 -6 5.8
2016 8 5.8
2017 9 2.3
2018 0 6.2
2019 6 3.2
2020 2 6.0

The 11-year CAGR is 6.1% and the average time in the market was 4.1 months per year.

DB2
Print the post Back To Top
No. of Recommendations: 4
I decided that I would be content if my screens went up a minimum of 4% a year (not counting dividends). Therefore, beginning in January, at the end of each month if a screen has a gain of at least 4% for the year the proceeds are not reinvested but put into CDs/money market until the following January.

That's a dumb idea. Notwithstanding the fact that the market doesn't care what you are content with.

The implicit assumption here is that the market always goes up.

Some years the market goes down. Sometimes a lot. With this scheme you take all the losses but chop off your gains at 4%.

To illustrate:
The market (S&P 500) lost 19% in 2002. If you chop off at 4% annual it would take 5 years to get back to even.
In 2003 the market was up 23%. So just holding on you would be more than even. If you take the -19% loss and limit yourself to the 4% gain, you are way behind.
Print the post Back To Top
No. of Recommendations: 0
Here's a summary of the post-discovery performance of the 4% screen profit latch. [starting in 2010]

Neatly showing my point. Screens always look great if your backtest omits downturns like 2000 and 2008.
Print the post Back To Top
No. of Recommendations: 3
FWIW DB2 I might suggest ditching this approach. Without mining the thread, one problematic question is how long do you let a losing year go to try to come back and make that 4%? All the way? Without a parallel stop-loss in place it seems... unbalanced.

I've made 29% in 6 months with my 3 screen blend - which has brought my 3-YEAR return to 25%. And it got crushed 15% Jan-Mar around the pandemic - so, net 13% for the year. This is a special year; it took following timing signals, watching small cap asset class momentum and general market sentiment to stay determined to stay in with the blend.

I frequently take a profit latch on an MI *stock* above 25%. But if I had cut off the YTD return at 4%, this year that goalpost was reached after an *incredible* 5 month recovery in the market by late October. And then would have missed November.

As a founding member of this board you know that these MI screens are higher *risk* - more susceptible to market drawdowns - than general investments. When the market goes down, these screens go down more.

Let your winners run and cut your losers short, as the saying goes.
Print the post Back To Top
No. of Recommendations: 6
Some years the market goes down. Sometimes a lot. With this scheme you take all the losses but chop off your gains at 4%....Screens always look great if your backtest omits downturns like 2000 and 2008.

Ray, you might spend some time reading the thread.

A) The gains are not capped at 4%. Screens are checked at the end of each month, so returns can exceed 4%. The best year was 1999 with a 23% return plus interest; 2000 and 2011 clocked in with 18% returns plus interest as the screens had a great start and then sat out the rest of the year in cash.

B) The approach outperforms in down years. Here are the years that SPY went down:

Outperformance
SPY percentage points
2000 18% -10% + 28
2001 6% -12% + 18
2002 10% -22% + 32
2008 8% -37% + 45
2018 0% - 4% + 4

DB2
Print the post Back To Top
No. of Recommendations: 10
I was part of the original discussion about profit latch used on the decision moose picks. I noticed that the moose picks would often gain 30-40%, only to revert back -20% before the sell signal triggered. My original intent was to take a % of your current holding off the table after it had gained x%. And continue to take x% of current holdings off the table as it continue grow x%.

I personally use a 20% profit latch, if I hold 100 shares, I sell 20 at 20% profit. I next sell 16 shares at the 40% profit latch. I next sell 13 shares at 60% profit. And continue the process till the screen sells the stock. It allows me to comfortable take profits off the table, but still let my winners run. I've just had to many stocks make an amazing run for a couple months, only to give half my profits away when it falls out of favor. May not the "best" or highest CAGR possible, but if I'm able to make 20% in 6 weeks and lock in a profit with a portion of my holding, it lets me sleep better.

Zee and others did a great study and backtest of the idea here: https://boards.fool.com/profit-latch-for-dm-26633655.aspx?so... His backtest was similar to DrBob2 and a complete sell and hold cash after x% and wait till next signal. My approach is to just take some off the table and let the winner run.

Peace,
Opihi
Print the post Back To Top
No. of Recommendations: 4
This post uses the profit latch concept with screens....

I decided that I would be content if my screens went up a 4% a year (not counting dividends). Therefore, beginning in January, at the end of each month if a screen has a gain of at least 4% for the year the proceeds are not reinvested but put into CDs/money market until the following January.

But what screens to use? You want one with a good CAGR in case you end up in the screen for more months than anticipated. And, since the profit latch is driven by the volatility pump, a screen with high GSD is also wanted.


People seem to have trouble wrapping their minds around this application of the profit latch.

Any any rate, we've finished the first two months of the year, and it is time to check on the performance of the six volatile screens.

Jan Feb End
78RPM 1.42 42%
GS 1.18 18%
PIH...Naked 1.13 13%
POG 1.10 10%
Turnarounds 1.20 20%
Up5 1.02 x 1.22 24%
.
Average return 21.1%
Ave time in market 1.2 months

All six screens have now exceeded the 4% bogey so it's all cash and time play some golf (or go skiing). We'll revisit in late December to see how the S&P has done relative to this low volatility approach.

DB2
Print the post Back To Top