5 stock strategy w/1.3X better CAGR, better beta, same sharpe as SPY

Investing Strategies Mechanical Investing
1

sp500.a !=null
ratio(tr(1,5),tr(1,200)) bottom 10
ratio(Mcap,float.s) bottom 5
where Mcap=product(sprc(1),csoq1d.s)

Create [Mcap]: [[daily SI-adjusted Price; share_lag=1 days; quote_lag=1 days]*[SI (d) Common Shares Outstanding; lag=1 days]]
step0: [S&P 500 Member; lag=1 days] != null
step1: [[Total Return Multiplier over 5 days; lag=1 days]/[Total Return Multiplier over 200 days; lag=1 days]] Bottom 10
step2: [[Mcap]/[SI Float; lag=1 days]] Bottom param0; Cash When Fewer
Holding period = 20 mkt days
20081231 forward to end of gtr current tester time limit

CAGR            19.82  v.    14.77
GSD(20)        29.2           21
Sharpe(20)        .8              .8
LDD(20; 0%) 16.5            6.76
UI(20)            12.5             6.2
Beta(20)            .93           1.1
max dd         -44.3         -39.8

https://gtr1.net/2013/?s20081231::sp500.a:nenull:ratio%28tr%281,5%29,tr%281,200%29%29bn10:ratio%28Mcap,float.s%29bn5:Mcap:product%28sprc%281%29,csoq1d.s%29
10 Likes
2

Translating to English from GTR1-ese (for the 3 people in the world left who might read this board and donā€™t know GTR1)

  • out of all S&P500 stocks
  • total return over last 5 days as a percentage of total return last 200 days is in the bottom 10 (worst 10 5 day mean reversions)
  • of those 10, highest 5 float %s using most recent quarterā€™s market cap
  • hold for 1 trading month (20 market days)
  • backtest from end of 2008 to current daily backtester end date (as of writing is March 10 '23 due to hard drive failure of the backtester server.)
16 Likes
3

It also works with the Nasdaq100:

CAGR      20.56      11.0
GSD        16.2         28.0
Sharpe       .697         .42
LDD        20.39       16.3
UI            22.4         28.1
Beta          1.17         1.3
TI            18.6            8
AT             5.65           .69
mdd       -69.26    -80.8

19970902 to 20230310
nas100.a !=null
ratio(tr(1,5),tr(1,200)) bottom 10
ratio(Mcap,float.s) bottom 5
where Mcap=product(sprc(1),csoq1d.s)

Create [Mcap]: [[daily SI-adjusted Price; share_lag=1 days; quote_lag=1 days]*[SI (d) Common Shares Outstanding; lag=1 days]]
step0: [NASDAQ 100 Member; lag=1 days] != null
step1: [[Total Return Multiplier over 5 days; lag=1 days]/[Total Return Multiplier over 200 days; lag=1 days]] Bottom 10
step2: [[Mcap]/[SI Float; lag=1 days]] Bottom param0; Cash When Fewer
Holding period = 20 mkt days

4 Likes
4

Would there be any chance of getting the GTR1 link for the sp500 version? (The link shown was not working.) It would be nice to see, now that GTR1 is up to date.
Thanks
John

1 Like
5

https://gtr1.net/2013/?s20081231::sp500.a:nenull:ratio%28tr%281,5%29,tr%281,200%29%29bn10:ratio%28Mcap,float.s%29bn5:Mcap:product%28sprc%281%29,csoq1d.s%29

4 Likes
6

Many thanks for that! Sorry I wrote that the GTR1 link was missingā€“Iā€™ve since realised that the link you sent was the SP500. This in fact Iā€™m going to find quite useful to compare with my results.

Iā€™m getting to understand better what your screen is about. In the first pass it picks the currently worst-performing stocks in the sp500 in relation to their 200-day performance. Second it looks for the proportion of available capital. This has been shown to pick the more stable stocks in the set as these stocks have proportionally more equity in the market. I like its contrarian nature and simplicity, making use of ā€˜reversion to the meanā€™.

As a Brit Iā€™m unable to invest in $-based USA ETFs [a stupid old pre-Brexit rule], so I keep hunting for a good SP500 proxy. In my tax-free account [ISA] I have an SP500 ETF but itā€™s in British pounds.

I see GTR1 has temporarily haltedā€“letā€™s hope it continues very soon. Iā€™ve missed it.

All the best
John

7

https://www.datawallet.com/stocks/buy-sp-500-uk

8

Thanks for that. As Iā€™ve mentioned, Iā€™m invested in SP500 as a proportion of my ISA account in British pounds and thatā€™s pretty cheap to run. I also have a taxable $ account and thatā€™s the one I canā€™t invest in an SP500 ETF so have to use a proxy. Unfortunately thereā€™s capital gains to pay this way, compared with a buy-it-and-forgetā€“it purchase.

All the best
John

1 Like
10

Excellent screen, @musselmant . It only has 2 steps, focused just on price behavior and shares outstanding. No value or growth factors whatsoever, yet it beat every single SI Pro screen in the regular rankings.

The first step is based on the type of stage analysis used by trend followers and US investing champions like Mark Minervini. They look for stocks in Stage 2, which have already had a significant upward move within the past 6 months, but which have leveled off (consolidated), and are trading closer to their 50dma over the past few weeks. These are ideal situations, they have found, for another thrust upward. (They also set stop losses to limit their losses.) The final step is to find stocks with high liquidity within this group.

Letā€™s call this screen Liquid Consolidators. Here are the results over the past 15 years.

2009-2023 Liquid consolidators S&P 500
CAGR 20.8% 14.0%
LDDD3 9.9 6.1
MDD -44% -34%

I also like that the universe of stocks starts with the S&P 500, so these are all large cap, liquid companies. Iā€™ve subscribed to SI Pro to track the picks. (I now have subscriptions to Seeking Alpha. Portfolio123, Whale Wisdom and SI Pro, lol.)

5 Likes
11

I just ran this ā€œLiquid consolidatorsā€ screen and got the following picks:

Ticker Symbol Company Name
CCL CARNIVAL CORP
NRG NRG ENERGY INC
UBER UBER TECHNOLOGIES INC
PHM PULTEGROUP INC
RCL ROYAL CARIBBEAN CRUISES LTD
BLDR BUILDERS FIRSTSOURCE INC
META META PLATFORMS INC CL A
NVDA NVIDIA CORP
LLY ELI LILLY AND COMPANY
AVGO BROADCOM INC

Interesting bunch of companies. Note, the original backtest by musselmant only includes the top 5 picks listed above. Screen link is below:

https://gtr1.net/2013/?!!QlpoMTFBWSZTWap!2B!2F!2FwAAEyfgAIFenAABgAAvyX!2BASAAdQ1R6nig0ANqAfqg1MU09NQD1BoMhp4Oqksr8ahAdCiszGUiNdIUJ6vja4EGsyZIXFAB5cO3t0zkmcQJwFgJyOtzsRbDNp2PEdT8NVMguc9iw7MiZpRCDsWTCM9ycneBqnPq9fi7kinChIVT9!2F!2Fg

Results
Statistics are calculated from 20081231 to 20240111 over daily closing portfolio values for all 20 trading cycles of Variant 0 .

CAGR: 20.780493
TR: 1670.473022
SAWR(20; 0.95): 13.309692
GSD(20): 29.065104
DIGSD(20; 0%): 31.105606
LDD(20; 0%): 16.288181
LDDD3: 9.947602
MDD: -44.373623
UI(20): 12.487172
Sharpe(20): 0.831882
Beta(20): 0.954895
TI(20): 22.373478
AT: 6.006714
9 Likes
12

up 27% in 2023 for S&P version

3 Likes
13

I imagine that for most stocks the ratio (Mkt Cap / Float) is pretty close to (Mkt Cap / Shares Outstanding) = Price. So why make things complicated? Replace the last ratio by Price and you get about the same thing.

John

1 Like
14

Yes, no complaints; others also suggest simplifying and it all makes sense to me.
People donā€™t post new screens very often so Iā€™m glad we are back to examining
possibilities.

3 Likes