Credibility of CANSLIM and William O'Neil

[murray t turtle]

For what it's worth, AAII's CANSLIM model has outperformed the S&P pretty impressively. I believe their models are 100% mechanical and I think they've tracked CANLSIM for a long time.
AAII

The only thing I don't know is how much of this performance is real-time vs. backtesting. Some services "tweak" their rules and post a new backtest whenever their models start underperfoming. I think AAII is pretty reputable but I've never subscribed to them.

%%
I never did make the kind money off single stocks that I wanted to, thru 2023 anyway;
even though its most likely= seldom happens to expect even best stocks have REALTY like returns [RE returns =steady, almost never a loss,LOL].
BUT using all his books + ETFs , helped a lot.
Seldom found any serious errors in his principles;
except his anti obvious bias can be wrong a lot[what is obvious in stock market seldom works, he loved to repeat ]
Actually an obvious bull market uptrend works a lot;
but even there, like my banker dad loved to say ''son dont tell everything you know''LOL
And i found his daily IBD newspaper [weekly now]about 777 times more useful than WSJ.

And i wrote the WSJ once ''why dont you give me 10 free copies like IBD?? NEVER heard back from WSJ on that LOL .
FFTY seldom beats SPY ; but strangely , in 2024 it has , by a bunch %

 

[ms33]

Given the size of the community IBD serves, its investment "theorems" are dramatically under-researched. They make a big fuss over counting distribution days to anticipate selloffs while a quantitative research site I respect finds that distributions days signal exactly the opposite behavior. Their fund performance is troubling. I recently dropped Marketsmith, finding it a huge time suck that didn't lead me to a single winning trade in two years. 0 for 40 is not what was promised. I've done way better with my patented dumbfuck strategy.

My tweet to a meticulous quant promoter of Can-Slim picks: 20 days in: your names up an average of 7.78%, not bad. Especially with MORN up 12%, USLM up 22% and APP up 27%. Ported those three names to my watchlist. My "dumbfuck" strategy - IBIT - up 29%.

 

[deaddog]

Given the size of the community IBD serves, its investment "theorems" are dramatically under-researched. They make a big fuss over counting distribution days to anticipate selloffs while a quantitative research site I respect finds that distributions days signal exactly the opposite behavior. Their fund performance is troubling. I recently dropped Marketsmith, finding it a huge time suck that didn't lead me to a single winning trade in two years. 0 for 40 is not what was promised. I've done way better with my patented dumbfuck strategy.

My tweet to a meticulous quant promoter of Can-Slim picks: 20 days in: your names up an average of 7.78%, not bad. Especially with MORN up 12%, USLM up 22% and APP up 27%. Ported those three names to my watchlist. My "dumbfuck" strategy - IBIT - up 29%.

What was your critetia for entry and exits?

 

[ms33]

I look for stocks with great tits and if it's starting to move but still under a 50 bps gain in the morning, I buy it. With crypto though that could be 400 bps. Exits: sucker starts to turn on me. I day trade so in quick and out quick. Spend a lot of time comparing stocks prior to the open. I can be in and out of appreciating stocks 5-10 times a day. Can-Slim doesn't specifically address a stock's real-time behavior - do all this shit and the stock should go up - I just focus on strong stocks (Can-Slim or otherwise) that are going up each day.

 

[murray t turtle]

I've got one or more CANSLIM books in my library, and only read them out of interest of humor. Nothing else.

Let's think about this for a minute. Do you honestly think that a simple solution such as CANSLIM really can out-beat the market? Considering, that even just one supercomputer owned by Jim Simons and his hedge fund carries out algorithms of magnitudes of orders higher than this simplistic system?

Any possible repeatable gains would have been arbitraged away long ago.

%%
Good points;
but part of that sounds like the random walker theory that never would pick up a $20 bill , because someone would have already picked it up ,if it was really there LOL
And while his books have been helpful,[repeating patterns.....] FFTY has mostly underperformed SPY.
But taking note of your '' any possible repeatable gains would have been arbed away long ago'';
YTD/
1mo/
6mo /, FFTY beat SPY.[3 years , not so good]

 

[deaddog]

I look for stocks with great tits and if it's starting to move but still under a 50 bps gain in the morning, I buy it. With crypto though that could be 400 bps. Exits: sucker starts to turn on me. I day trade so in quick and out quick. Spend a lot of time comparing stocks prior to the open. I can be in and out of appreciating stocks 5-10 times a day. Can-Slim doesn't specifically address a stock's real-time behavior - do all this shit and the stock should go up - I just focus on strong stocks (Can-Slim or otherwise) that are going up each day.

I can see where marketsmith wouldn't be much help.
Canslim holding period is much larger time frame.

 

[murray t turtle]

%%
CAN SLim [m = market direction ]is such an unpopular nickname;
may work well in the future, maybe LOL

 

[ms33]

I can see where marketsmith wouldn't be much help.
Canslim holding period is much larger time frame.

My dumbfuck strategy has tripled my daily take from a clueless $60 a day on $100,000 over five years to more like $180 a day, enough to eliminate the drawdown on my retirement capital. (That point is actually $134 a day which is 34% a year and no mutual fund or ETF consistently earns that; they appear to top out around 26% a year long-term.) But I think my approach teaches the wrong lessons; essentially I'm behaving like the mad hatter in the first thirty minutes of the market looking for a stock where I can put the petal to the metal. I've resubscribed to MarketSmith, now MarketSurge, for its ability to assess a stock in less than 10 seconds. Along the way testing TC2000 and DeepVue for what they can add to the process. I'm impressed with what I've learned from Richard Moglen at DeepVue about how he's applied 10/20 moving averages to asses a name post pivot. The most significant takeaways for me in the last three years of chasing the White Whale come from Mark Minervini and his exquisite trade craft. Don't know if Can-Slim actually works. If it does, it's coming from pattern recognition and not from fundamentals. A long-term MPA participant of Minervini's pricey trade partnering say Mark rarely pays attention to fundamentals; he trades almost entirely from the tightening dynamics of the Volatility Compression Pattern. So back to square one, much better informed, and ready with a fresh heart to listen to my elders, even though I'm 77.

 

[almx]

My dumbfuck strategy has tripled my daily take from a clueless $60 a day on $100,000 over five years to more like $180 a day, enough to eliminate the drawdown on my retirement capital. (That point is actually $134 a day which is 34% a year and no mutual fund or ETF consistently earns that; they appear to top out around 26% a year long-term.) But I think my approach teaches the wrong lessons; essentially I'm behaving like the mad hatter in the first thirty minutes of the market looking for a stock where I can put the petal to the metal. I've resubscribed to MarketSmith, now MarketSurge, for its ability to assess a stock in less than 10 seconds. Along the way testing TC2000 and DeepVue for what they can add to the process. I'm impressed with what I've learned from Richard Moglen at DeepVue about how he's applied 10/20 moving averages to asses a name post pivot. The most significant takeaways for me in the last three years of chasing the White Whale come from Mark Minervini and his exquisite trade knowledge. Don't know if Can-Slim actually works. If it does, it's coming from pattern recognition and not fundamentals. A long-term MPA participant of Minervini's pricey trade partnering say Mark rarely pays attention to fundamentals; he trades almost entirely from the tightening dynamics of the Volatility Compression Pattern. So back to square one, much better informed, and with a fresh heart to listen to my elders, even though I'm 77.

I find it counter-intuitive to subscribe to Smith and Surge and then take as most significant takeaway the Minervini setups that don't care about fundamentals. I'm a subscriber of IBD digital only and trying to trade consolidation into CANSLIM stocks, aren't Smith and Surge trading only strong fundamental stocks ?

 

[ms33]

I find it counter-intuitive to subscribe to Smith and Surge and then take as most significant takeaway the Minervini setups that don't care about fundamentals. I'm a subscriber of IBD digital only and trying to trade consolidation into CANSLIM stocks, aren't Smith and Surge trading only strong fundamental stocks ?

MarketSmith has been renamed MarketSurge. Mark Minervini is one of the most successful Can-Slim investors I'm aware of and he only uses a few of its core factors: RS > 89, Composite score > 80. A lot of 50/150/200 day filters to make sure a stock is pointing north. I'm sure he glances at the quarterly sales and earning exhibits. But my contact tells me he never talks about fundamentals; they're looked at as benefitting longer-term returns and Mark is usually out of a name after a 10% gain.

With everyone and their mother running fundamental screens or quant models, it's hard to imagine that fundamental factors aren't heavily arbitraged. I can't think of a single Can-Slim oriented whale who spends much time on fundamentals. The genius of MarketSurge is how well they've implemented pattern recognition.

I'm personally influenced by the fact that I can find almost no cross-sectional correlation (10,000 stocks on a given day) between O'Neil 's fundamental factors and 1 day, 5 day and 1 month returns. Even the flagship RS fails to show any correlation. The only factor that did was quarterly earnings during earnings season. That jumped off the charts with a 47% correlation; everything else was below 10%.