What are the rules for the LargeCapCash screen?
(I guess this from around March 2020)

boards.fool.com - Jim what is this the quotlargecapcashdivquot

http://www.datahelper.com/mi/search.phtml?nofool=y...

This screen is intended as a SPY replacement, just with somewhat better long run returns.
So, by design, it leans very strongly to large caps.
It just gives an edge by holding fewer losers: it sticks to those with high ROE and strong balance sheets. That's about it.

There are two versions, one with dividends and one without.
Hence the "Div" on the end of the name of the other one, "LargeCapCash".

Summary:
Original version used the Value Line set of stocks as its starting point. Others created an equivalent using the Stock Investor Pro set of stocks.

The steps:
(1) Keep only those stocks that have a Timeliness ranking. (good or bad, just not missing)
This isn't strictly necessary, but it rules out a few firms that have very little history or just
went through transformative M&A, or in the process of being bought out or going bust.

(2) Find those stocks with a dividend.
This step is entirely optional, depending on your preference for stocks with or without dividends.
The dividend version leans to *slightly* smaller large-caps, but the average annual return in backtest is actually a bit higher.
As you wish.

(3) Of those still in the running, keep the 30% of them that have the highest reported ROE.
Taking 32% might be better if you're obsessive, makes little difference.
At this point you have an average of 475 candidates remaining.

(4) For those remaining, calculate their cash minus long term debt.
Not cash as a percentage of market cap, just absolute size of cash pile beyond debt.
Buy equal dollar amounts of the 40 with the most cash.
Hold for a month (or two or three), and repeat.

Recommended version:
Buy equal dollar amounts of the 40 "best" as above.
Hold for two months.
Calculate the rankings again.
For each stock you own no longer ranked in the top 45, replace with the highest ranked one you don't already own.
Repeat.
In the jargon, that's "top 40 hold-till-drop 45", or 40 HTD 45.
40 HTD 60 is also very nice, less trading.

Hear are year by year results for the 40 HTD 60 backtest, the version without dividends.
Includes reinvested dividends, and allows 0.4% per round trip for trading costs.

Period     Screen    S&P      Diff
   1997      32.2     27.9     4.3
   1998      42.8     34.5     8.3
   1999      16.3     18.4    -2.1
   2000      23.9    -10.8    34.7
   2001      -6.3     -8.0     1.7
   2002     -16.5    -18.9     2.4
   2003      23.0     23.0     0.0
   2004      15.3      8.9     6.3
   2005      10.4      7.5     2.9
   2006      16.7     13.8     2.9
   2007       6.1      1.9     4.3
   2008     -27.8    -32.9     5.1
   2009      39.5     25.5    14.0
   2010      20.5     14.2     6.3
   2011       4.2      2.5     1.7
   2012      14.7     17.1    -2.5
   2013      28.7     27.6     1.0
   2014      18.7     12.9     5.8
   2015       1.5      1.9    -0.4
   2016      18.5     14.4     4.1
   2017      31.5     21.8     9.7
   2018      -2.7     -3.5     0.9
   2019      34.5     29.9     4.6
   2020      31.0     16.9    14.1
Jan-Mar      13.7      9.8     3.8 (not annualized)

Overall      14.8      9.3     5.5

So, from the backtest anyway, when it underperforms it usually isn't by much.
The long run advantage is heavily flattered by 2000. Best to ignore that.
In reality I wouldn't expect an advantage of more than 3 points as a wild guess.

It was proposed about a year ago.
First 11 months since then, screen +67.9%, S&P +45.7%.
Doesn't prove anything, but off to a nice start. A lack of prompt failure, anyway.

Since by design it is a lot like the S&P, it works not so badly with a bit of hedging.
Own this long, then go short the S&P 500 with futures a little bit.
Same long run returns or a little better, but with smaller net stock exposure and a steadier rate of return.

Here's the post that introduced it, a year ago this week.

boards.fool.com - A spy alternative screen

I have a couple of extra tweaks I like, but that's my nature.
For example, insert a step (2.5) that has you keep considering only the 50% of stocks closest to their 52 week highs.
Sort of like using crowdsourcing to cull some losers and some that the market *thinks* are losers for now.
Rate of return not that different (backtest is 1.6%/year better), but it's a much smoother ride.

Jim

------------------------------------------
boards.fool.com - Please let US know if you have more ETF
10/30/2021

GOOG MSFT FB AMZN BABA SONY TSM CSCO HUM ACN
VRTX COST NVDA TROW ATVI ZM BBY PYPL NKE INFY
MOH EQH ERIC ANET TTWO SGEN AB AMP SAGE MNST
CTSH NTAP AMD GRMN FTNT INTU LOGI CHKP VEEV REGN