Discovering trading success by accident

  • Post-IT Notes
  • Microwave Ovens
  • Coca-Cola
  • Penicilin
  • The ink-jet printer
  • X-ray images
  • Potato Chips
  • Teflon

What do all of these things have in common? 

They were all discovered by accident, usually as a result of a lot of experimentation.  Some were created with a general goal in mind and in other cases a completely opposite use-case resulted.

Take the case of the Post-IT note.  A scientist for the 3M Company came up with a very weak adhesive that could be peeled off when it was applied to any type of surface.  It also didn’t leave any residue.

What is amazing, is instead of trying to make a very weak adhesive, he was actually trying to create a super-strong adhesive.

It was a complete failure… exactly the opposite result of his intent.

It certainly was a failure as it relates being a strong adhesive, but another scientist at 3M noticed that when applied to strips of paper, it made great book marks for his church songs, without leaving the adhesive on the pages of his book.

And the Post-IT note was born.

What is inspiring about these inventions is the fact that they were mistakes.  Most of the time, they were created by tinkerer’s, people who kept trying different variations of this…or that…or this…or that…

With trading…accidents happen with amazing results

Let me give you a few examples of how some of my experimentation lead to successful trading ideas.

My “Dead Cat Bounce” screen…

I have described a screen that I use on Finviz that looks for stocks that have the largest single-day percentage declines, but have also been behaving well in the more recent past by being above the 200 Day Moving Average (DMA), and an even tighter screen with the 50 DMA > 200 DMA.

From that screen, I pick out stocks that look to fit the patterns I typically trade… a rebound back up after trending further down after that large one-day move to a point of support.  I track these stocks in the following days/weeks, waiting and watching for the ones that reach logical areas of support.

Some of them launch back up immediately after the big one day decline, but others meander down to exactly where I wanted, and then, if they start to show signs of slowing down and potentially reversing off of that support that I had previously drawn, I enter the trade with the hopes of a move back up.

That particular method works amazingly well!

But I didn’t just come up with using this screen in this way. 

It all started with using this exact same screen, but I called it the Dead Cat Bounce.  My idea was to make note of the 5 biggest losers that fit that screen and pick one of them to buy before the closing bell.  I would then sell the next day at a higher price.

Here is a sampling of the results…


Notice Column J, “Day 1 High Percentage“, which is the high the day following the big 1 day decline…nothing but a sea of green.

Column C is the closing price the previous day and Column F is the high on the following day.

You can see that at some point during the next day (after that big one day decline), it was very, very likely that there would be a high that is greater than the price of the previous close.  In this screen-shot sample of my spreadsheet,  you there was exactly zero stocks that did not have a price that was above the close of the previous bid down day!

What is the overall percentage of the time that the stock, at some point during the following day, higher than the previous big down day’s close?

Look at the totals from the bottom of my sheet summarizing the results…

There you have it….88%.

That means if you bought at the closing price on the previous day, at some point (88% of the time), the stock was higher than on the following day!  Look at that average percentage amount of how much higher the price got when compared to the previous close: 7.59%!

Seems pretty amazing right?  It is, and I still find it intriguing.  By the way, I tracked the biggest one-day losers WITHOUT regard to whether they had been performing well or not (not above the 200 DMA), and the performance was nowhere close to this.

Here is what the problem was though…at one point will that high occur? 

  • At the opening bell during the first few trades?  Yes.
  • At around 9:45 after an initial dip lower? Yes.
  • At around 10:00 AM? Yes
  • 11:00 AM? Yes
  • Right near the close of the day as it surged back higher after dipping lower? Yes.

All of the above.

On paper, I looked to have a solid idea, backed up by actual evidence, it turned out to be hard to implement.

Sometimes I ended up getting stopped out at a loss, only to watch the stock reverse and surge higher.  Other times I would get a small gain and sell as the stock showed signs of weakness, only to watch it continue higher.

I also had to be tied to the screen, watching the trade unfold and try to decide where to get out.  That doesn’t fit my personality, or what I want to do all day.

At the end of the day, it wasn’t for me.

Maybe I’m not sophisticated enough.

Maybe this would work for someone  who is good at developing some type of a robot or script that could automate some of this, add some parameters to it, maybe add a trailing stop, etc..

If that’s you…awesome, feel free to take this and run with it.  🙂

So I abandoned trading the “dead cat bounce” as I had intended. 

But something did come out of it.

When I went back and looked at the charts of these stocks days and weeks later, I saw that many of them headed further down to logical support areas and predictably bounced higher.  They also had the typical trading pattern setups that I look for in a chart.

This screen turned out to be an excellent way to find the type of trades that I particularly like, stocks that are performing well, had a previous big move up, then a move back down to logical support…I buy…then ride back up.

Now I use this same screen to find and then monitor only the stocks that fit the chart pattern I look for.  If the price does continue to trend lower to logical support areas, I’ll very likely get in for a swing trade back up.

The “Message Board Screen”

Way, way back in the day…in a .COM bull-market boom far, far away, I used to spend a lot of time on stock message boards like Raging Bull, searching for undiscovered Penny Stocks that had potential to get noticed and then zoom higher.

In these message board sites, I would wade through what was mostly a sea of trash, to find penny stocks that had the potential to launch and pump higher.  I would buy many of these dubious stocks knowing they were likely trash, but hold and sell when they pumped higher.

From reading the message boards, and watching the “buzz” of the rumors and opinions, I knew what the “story” was behind why this stock could potentially pump higher, but I’d have to wait for the rumor mill to spread that story to everyone else.  If it happened as planned, then the activity on the board would start to surge with excitement, new posters would start to show up, the message volume would increase and the stock would start to launch higher.

Many times though, I would be holding these low-priced pigs and nothing would happen, or they would head lower and I would sell for a loss.

That was the big problem.

I over time I got a feel for typical pattern and flow of message board traffic…low, stable activity with discussion about the product or story, a bump here and there related to news developments forming, a rather large surge related to some type of news or press release event and many times, the anticipation of news, then maybe a small decline, then a huge surge as the story or news unfolded and everyone heard about it and rushed to the board to find out and post their thoughts.

In reality though, it was complete luck that I happened to be early to some of these penny/tech stocks, getting in low and then having them pump higher.

Overall, it wasn’t working for me.

Finally it hit mewhat if I tracked the message board activity of ALL of the tickers of these low-priced stocks?

This way, I could monitor the flow, when I saw a stock board with an increase in message posts, I would pop over, read and investigate what was going on and then get ready to trade that stock if it started to launch.

I created a spreadsheet, manually copying the index of stocks and the number of message counts on a weekly basis and look for the ones with the bigger changes.

This worked so amazingly well!

Instead of randomly finding a penny/tech stock that seemed to have a great business story, waiting for everyone to discover it, I would circle around the boards like a hawk and get in on the ones that showed huge surges of message post activity and jump in on the stock, often just before the stock moved WAY higher.

For several years, this is all I did, and it worked like a charm.  And it came about completely by mistake.

These days,  message boards are basically replaced by Twitter/Stocktwits and other social media technologies.  Now, some traders and hedge funds are using sophisticated tools to extract and monitor social media activity…”buzz” for their trading strategies.

I stopped using this method after the .COM bust when everything dried up, but now… apparently this is now a thing.

Do you have any beautiful trading mistakes?

I find it interesting how, as a result of tinkering or thinking outside of the box, you end up somewhere completely unintended.

I’m curious if you might have any stories of how you discovered something related to your trading completely by accident…or as a result of experimentation.

Leave a comment below!

-Glenn

2 comments

  • Hi Glen. It is always good to read you posts. I have a question regarding to your dead cat bounce. It seems a very effective trading strategy.

    First, for the chart you provided, what kind of screens did you use? I am guessing price > SMA200 and Curr Vol. > 300k, and you sorted them by top losers? I tried this screen, it came out over 1k stocks. Not quite sure if my screen is correct. Can you share your screen?

    Second, I can try to do some experiments on paper by using Finviz. For instance, records the data for previous day and next day to see the effectiveness of your methods. But the question is if I am ready for a real trade, how can I find those stock fall into the scenario of dead cat bounce in real-time? Finding the top losers of the day? Could you please also share your idea?

    Your methods are really attractive. And based on your data, it is also very effective. I would like to take a try and see how things are going.
    Please kindly share your thoughts.

    Best,
    Bill

    Like

    • Hi there Bill,

      That is a Finviz screen. I think the reason why you are seeing a huge number…like around a 1000 results or so, is because the key is to select “Top Losers” within Finviz. After the market has closed, select the “Top Losers” pulldown, then a price that is below the SMA200, and then the Current Volume > 300,000. That should give you maybe about 20 stocks or so, sometime more or less depending on how things were that day for the overall market.

      About your 2nd question. With Finviz, you can run that screen during the day, but I’ve never done it. You would probably need to have more of a real-time service. That is also getting more into day trading…buying and selling in the same day, which I don’t really do. I don’t have the time for it, and also found that I like to take a bit of time to analyze a stock pattern and make a decision…on my time.

      Glenn

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.