What Blade is talking about is a style of what is called technical analysis, which predicts future stock valuations based on historical pricing and volume, rather than on the fundamental underpinnings of stock valuations (profits, earnings, credit ratings, and other macro and micro economic factors). In application, he is blending an understanding of both when making a decision where/when to place his money.
The technical analysis theory has soundness to it, and it has historically been a very successful way to trade and invest for both institution and retail. However, there are no absolutes in the market, black and white swan events do happen, and there are some reasons that technical analysis fails.
Without getting too far into technical woo land, or breaking any NDAs, I'll try to outline why technical analysis works and briefly touch on why it doesn't.
There are two primary reasons that technical analysis works.
First, a somewhat fundamental reason, and the reason technical analysis was discovered/invented in the first place. There truth is, stocks do have "support" at certain levels, and some of these levels have stronger support than others. This "support" is due to the reality of volume, and it is pretty easy to understand why.
Let's say that a theoretical company has developed a novel drug, and has a million shares outstanding. Then, let's follow the stock price from $1,000 to $10,000, and back down. At first a few insiders get wind and buy a total of 50,000 shares at $1,000. Then, institutional investors see the news, believe in the company, and buy 290,000 shares at $2,500. Retail traders notice the action, and get the news, and want to get in on the action. They drive the price up rapidly from $2,500 to $5,000, but only buy 10,000 shares total. Seeing the quick doubling of stock price, some of the big money players move in, and buy up 600,000 shares at $5,000. Other traders take note, and the remaining 100,000 shares are bought at a price between $5,000 and $10,000. Then, disappointing phase one results are announced, and the stock starts to drop. The drop from 10k back to 5k is precipitous. It doesn't take long for those 100k shares to trade hands. However, there is huge support at the 10k mark because 600,000 shares need to trade hands before the stock can fall below that price point. Once the price drops below 5k, however, the rapid drop resumes because there are so few shares to trade between $2,500 and $5,000. However, the drop is a bit slower as people who bought above $5,000 start to take their losses. When we hit $5,000 we see another big "support" level.
The market has similar "support" levels where massive buys have happened in the past. The stock market doesn't pause/ turn around at these levels because they are somehow magical, it does so because there are simply so many shares that need to change hands for the market to drop below them. When the market drops below these "floors", the "floor" can become a "ceiling" where trading will slow down on the way back up.
Side note, the market was never going to sustain the recent highs. The buy and hold volume was simply too low. Most of the volume in the market was just speculative trading.
A second reason that technical analysis works is that it has been around for a long time, it is well understood, and a lot of people have been successful trading based on the underlying assumptions. When it comes right down to it, the market moves by herd mentality. If you can accurately predict where the herd is going to buy, and where they are going to sell, then you can trade successfully. So, technical analysis works because so many others are/were doing it.
Now, why is it woo, and why does it fail.
One reason is that technical analysis doesn't just use those "support" levels. It also uses prior support levels to chart the trend of the stock price, and buy in before the price hits the support. The reality is, those big money investors aren't going to sit idly by and wait for the stock to get back to the $5,000 mark before they start selling. They may set their stops at $5,500 or $6,000. Technical analysis tries to predict where the stops are going to be, and where the volume is going to come back into the market. This is fraught with risk because you are starting to build assumption upon assumption. If your assumptions match everyone else's, then you are successful. If your assumptions aren't the same, you can lose a lot of money.
Another reason it may fail is that the floor is weak. Let's say that 80% of the shares bought at a floor were bought by true buy and hold investors. They aren't selling, no matter what. The floor is then only 20% as strong as you thought it was.
But the biggest bugaboo in technical analysis today is the evolution of algorithms/quant trading. You have to understand that all of your technical analysis presuppositions are built into the algos. The algos are built by mathematicians that are way smarter than you, and have nothing better to do than study the market, and use that knowledge to build better algos. They are faster than you. They can turn on and buy or sell at an incredibly fast rate. They know where the "support" levels are, way better than you ever could, and they use that knowledge against you.
It is not uncommon to see the market crash just below a technical "support" level, see retail traders, and institutions, capitulate and flood the market with cheap shares, only to see the market suddenly bounce. The algos waited until you capitulated, bought up all your shares, then reversed the market, leaving you holding the bag. They also do it on the upside. They will drive up the market, convincing you to buy their overpriced share, then they will crash the market, leaving you holding the bag. It is both beautiful and disgusting at the same time.
This is one of those weird things in life where what PGG said is largely true, and what Blade is saying is also largely true. There are no absolutes in the market, and the reality is that quant/algo trading has supplanted both technical and fundamental analysis for the big money that truly determines what happens to the market. Perhaps it is more correct to say that technical and fundamental analysis is incorporated into quant/algo models, and they use it against you.
That is why traders almost always lose. The big money is working against you, and your only advantage is the ïfact that you are moving a, relatively, small amount of money around. You can reverse your positions much, much faster, because you don't have to unload, or buy, thousand of shares at a time. But understand, you are playing against the house, and the house (almost) always wins.
arstechnica.com
