The thing is, these charts are all pattern recognition. Patterns are fine, but you have to be able to to define the patterns precisely enough to code into a computer algorithm. That way you can apply some data science and prove whether a given pattern has a positive expectancy. Otherwise, one could waste a large amount of time and money on subjective analysis trying to collect such proof.
I 100% do not agree with that sentiment at all. You 100% do not need a computer make big profits in the stock market (other than to access your broker and an end of day charting program for visually scan for set ups). all you need to do is study those examples of past big winners enough so that the most wonderful computing machine ever created - your own human bean brain - can recognize the stages from which the large moves advance, and the willingness to pull the trigger when price begins such an advance, the willingness to cut all losses at no more than 8% from your entry, and the willingness to take ,ost profits at 20% gain. The exception to that 20% take profit rule is that any stock that advances from a proper base buy point by 20% in three weeks or less is likely to advance much further quickly, so such purchases are to be held for a minimum of eight weeks.
Imagine manually scanning all the charts in the SP100 every day looking for one of these patterns, or conversely watching a handful of stocks for weeks at a time hoping one of these patterns develops. No thanks, I need to be able to back test patterns on years worth of data in order to validate positive expectancy. Then I need to be able to automate the scanning of data for that pattern over hundreds of issues if need be.
Yep, David Ryan said recently all you need is an iPhone. Myself and Dan Zanger trade/d off their Phone at work.
First off, the S&P 100 is not where you'll typically find these stocks, though during the bull market in megacap tech, you may. And the S&P 500 only has ... 100 stocks to scroll through. That is exactly what I do. I live and die on the weekly chart, for the most part, and it takes five whole days to finish just one bar. Typical bases can be 12 to 30 weeks. I never hope a pattern develops. I wait to see IF a pattern develops, and then, because of my study, I know exactly how to react and when to buy, where to place my stop, where to look to book profits. Have you ever heard of Kristjan Kullamägi? Look him up as he has pretty much done what you say isn't possible with respect to finding these stock set ups.
My main scan is shown below: When I started taking control of my stock trading, I used the "week's biggest gainers" table published in Barons each week. That worked great until the proliferation of ETF's and Barons refused to scan for common stock only. Many weeks the entire list was ETFs that weren't going to be on the list the next week. The drawback to IBD's tables is that it excludes stocks priced under $20 per share. Back when I started circa 2004-5, Barons alerted me to tickers such as BOOM, HANS (now MNST), NSI (then NTRI, and since acquired by Kanos Capital), DRYS (since acquired by SPII Holdings). All of these I purchased for less than $10, some for less than $5, and I believe nutrisystems was bought for around $2. I did not then nor do I now use any screening software.
David Ryan said he scans 500 Charts in 50 minutes on his way home from Church every Sunday. He uses Marketsmith on a Tablet which allows 4 Charts to be viewed on Screen at a time.
He probably shouldn't be doing that while driving. I really wish I could warm up to Market smith, but I have no feel for their charts. Great fundamental information and the fund ownership statistics is useful, but you can see most of that in the charts themselves. If Darvas didn't need it, neither do I.