If a meteorite hits earth, aliens are discovered, quantum computing breaks encryption, someone launches a nuclear bomb, etc, any of those will affect the stock market by a lot. Technical analysis can't predict a meteorite or any of those other things, therefore it cannot work.
It "might work if nothing significant happens" but that is just a weasel way of saying it doesn't work.
Now if you're saying you decided to invest in something and you're trying to get an entrypoint and you spot a moment of large deviation from the mean and you use that to influence your operational buying mechanics like how you space out your buys or time them, that I can definitely see. But online you always see these two very different things mixed together.
It's a lot more simple than that. You don't need to discover aliens to move the market in unexpected ways. A crop failure halfway across the world, a border skirmish between two countries, discovery of a new oil reserve, a ship blocking a canal, a tech company not selling as many phones as anticipated in a quarter...all have ripple effects through the economy and can screw all market projections. Stuff like this isn't a "black swan"...it happens every month.
If you want to use math and statistics to get an edge in the market then become a quant trader. People who aren't smart enough for that but want to feel smart get into "technical analysis".
Cicada populations in two different countries emerge the same year, decimating sesame crops while Indonesian crops remain unscathed and the futures shoot through the roof.
Yet, similarly to brownian motion, the thousands of events that cooccur more or less cancel each other out in the grand scheme of things.
>Technical analysis can't predict a meteorite or any of those other things, therefore it cannot work.
This feels like you're attacking a strawman version of technical analysis. The claim isn't that you can predict future price movements with perfect accuracy, it's that you can predict it well enough that you can make some money from it. None of our economic models can predict a meteor hitting the earth either. Does that mean we should conclude that all of them "cannot work"?
The fundamental problem is that everyone else—particularly people with unfathomably deeper pockets and much better access to information to you—can do the same TA as you. To whatever extent it does work, that opportunity is rapidly exhausted.
So all that’s left is the unpredictable bits, and those are what you’re at the mercy of. And the unpredictable bits are happening constantly.
In this case profit exhaustion doesn't really work. The assumption of technical analysis is that those with the deeper pockets move the market in certain ways and patterns and it's the job of the short-term speculator to anticipate those ways.
But if you assume enough rich people do technical analysis and pattern day trading you can earn money by doing what's essentially technical analysis, but with different patterns - ones designed to capitalize on the old patterns used by the rich.
Keep in mind, I'm not saying technical analysis works. I don't think it does, but that's without seriously examining the evidence. I'm just saying that the diminishing profit argument doesn't work. At least not as presented.
> The assumption of technical analysis is that those with the deeper pockets move the market in certain ways and patterns and it's the job of the short-term speculator to anticipate those ways.
if this were the case, it seems likely that others with deeper pockets than you, perhaps other high frequency traders, would rapidly exhaust those profit opportunities
Which they do. Many kinds of HFT trading strategies actually look a lot like a more mathematically rigorous version of technical analysis. It is very likely that they take all of the alpha that TA traders used to get.
HFTs generally consider shorter timescales (both in the feature set and the forward horizon) than human TA would lean on, and make tremendous use of microstructure information rather than outright price movements.
Having said that, “let’s take difference to EMA” is pretty the one-size-fits-all solution, and there are no shortage of firms looking at minute to hour long trades to gobble up anything human TA practitioners would ever try to trade.
> HFTs generally consider shorter timescales than human TA would lean on
This just further emphasizes the point that retail investors relying on TA are having their lunch eaten. HFTs are picking the minute opportunities because that's all that exists. If there were longer-term strategies that were reliably profitable, do you think they simply ignore them?
sub-minute opportunities are absolutely not all that exist, in many ways it’s the opposite. The hft opportunity space is extremely competitive, and much more zero sum than the mid/low frequency space. A relatively small set of players compete for the lions share of anything one might consider high frequency. The set of trades you compete for and the relevant information to generate forecasts is limited / similar across firms, and table-stakes on engineering side are very high.
> If there were long term strategies that were reliably profitable, do you think they would simply ignore them?
Yes? Mid frequency trading (say holding period of 5 minutes to ~1 hour) is a totally different class of quantitative trading, from the features you use to how you build portfolios (if at all) to how you execute on forecasts. At least one top-of-the-game hft firm (that already relied on forecasts instead of speed) got burned on their first attempt to enter mid frequency.
> At least one top-of-the-game hft firm (that already relied on forecasts instead of speed) got burned on their first attempt to enter mid frequency.
Put differently: "A bunch of smart people who have proven to be able to reliably extract profits from microsecond trades tried their hand at exploiting patterns on longer timescales and were unable to. Even though I have less money, information, access to markets, and am just one person, I believe I can succeed where they failed."
I don’t know why you think I’m saying TA works. The only thing I’ve said is that HFTs are by and large not responsible for trading away those inefficiencies
I’ll add that most hft firms are trying to move to longer time horizons. Some with great success, some mixed, some not at all. There’s not much unclaimed pie left in the hft world
The thing is there's no reason why new patterns can't exist, created by the patterns of the HFTs, who could create new profit opportunities for other HFTs or day-traders.
Now, perhaps each new generation of patterns would have diminishing returns over the previous generation. My intuition is that this almost certainly true. But either an argument or evidence has to exist for it and I'm not aware of any.
> The thing is there's no reason why new patterns can't exist, created by the patterns of the HFTs, who could create new profit opportunities for other HFTs
correct, exactly what I'm saying, any opportunities, including those derived from other opportunities being seized, would themselves be seized by such deep pockets and HFT firms
> The assumption of technical analysis is that those with the deeper pockets move the market in certain ways and patterns
If those with deeper pockets had reliably exploitable trading strategies, other teams with equally deep pockets would be exploiting them. You need to simultaneously believe that widely known and understood patterns exist which are profitable to exploit but also that nobody with large sums of money and a financial incentive to profit is interested in taking that easy money.
You're wrong in your conclusion, even though the argument looks good.
The opportunity is there for the average person to take, but it needs sufficient skill / training, like any other skill. I can't "prove it" to you, but I already proved it to myself and many people in the trading community did.
You don't need any sophisticated tools or huge amounts of money. It's bloody difficult though, not just for the technical reasons (there are many strategies but most go beyond a MA crossover, even though I'm not denying even that can produce alpha -- I don't know or care), but mostly for psychological reasons. And psychology is what's behind many of the patterns you see in the markets.
Every year there are people who beat the indexes and there are people who perform worse than the indexes. Every successive year some proportion of the winners stay winners and some become losers. The interesting thing to note is that—if you look at the statistics—winning one year has almost no bearing on whether or not you win the next year.
Still, there are participants who flipped that weighted coin year after year and caught heads five years in a row. It doesn't mean they have a successful strategy. It just means that if you flip a 45/55 coin five years in a row, two out of one hundred people will get a string of heads.
I've personally witnessed all of my day-trading friends go through this (I almost said "learn this lesson" but I'm not sure they have). For a while they beat the index and they're convinced their strategy is sound. One day something unexpected happens and they're suddenly in the hole. Often this is compounded by having an enormous tax bill on "gains" that no longer exist, necessitating selling off holdings that are deep in the red, making it even harder to get back to positive.
Do some individuals exist who can reliably beat the markets? Sure. They just are exceedingly likely to not be you, and none of them are offering their services to you.
Agreed about the statistic of failure rate. For consistent profitablity it's probably at or below 2%, even if it's higher in a given year. The reason is not only lack of a strategy that doesn't depend on specific market conditions, but often (unless you're talking about algos) is psychological. Trading is psychologically very difficult.
You're right in that most of the time, the directional probability of the market is close to 50% for a 1:1 trade.
There are instances where the directional probability is significantly above 50% to make a trade have a positive EV (if it's not 1:1, the baseline is not 50%, and there are similarly moments where the directional probability for a different risk-reward is above its baseline for some direction). Different strategies tailor to different RRs, but you can absolutely find those entries.
However, as you said it's easy to get caught up in the illusion of profitability for a given market condition. For example, in a strong bull market like the post COVID stimulus, you could trade to the long side in a large timeframe and just win money because of the anomaly of such a protracted spike in the daily charts.
That's not a sustainable strategy though which explains what happened to your friends. A price action system that takes into account markets structure works in every environment, be it ranging, surging or a bear market.
Learning price action TA takes time to learn though, it's a skill like when we learned to code till the time we got paid for it. And trading, in general, goes against how our brain works. Psychology is the great barrier to profitability once you have a strategy. It explains the high failure rate. it being a hard skill is why there's so few people doing it, But it's not because it needs a very high IQ or Rafa Nadal's strength of mind. Most people could get there, but it takes a lot of effort, and theres so much deception and scamming around it, it's easier to pretend it's impossible based on the amount of people that fail.
Something to be said about playing in pockets of the market those fat cats stay away from due to lack of the liquidity needed, market cap constraints, etc. There it can be a more level playing field for us minnows.
I would be willing to bet significant sums of money that virtually 100% of your trades, in any any of these "pockets of the market", have one of those fat cats as your counterparty.
I think TA is bad because it's priced in but your argument seems like nonsense.
It boils down to "the market is probabilistic and you can't know for sure it won't get impacted by things" but if TA gives you even a couple % edge (with black swans averaging out in both directions) it would have been worth it.
Depends on how "real". A microbe discovered on some rock 60 trillion miles away isn't really going to move the markets. An alien attack fleet entering our solar system...maybe go long Raytheon.
This is where my thoughts went as well. Harry Turtledove is prolific at coming up with scenarios where aliens are capable of interstellar travel but not good at war. His World War series sets up a slightly more plausible scenario. If you haven’t read it, it’s worthwhile and I don’t want to spoil it here. An interesting aspect of it though was the aliens having finite resources with very long delays in resupply. They underestimate humanity who takes advantage of this to slowly turn the tide.
Paxys thanks for the short story, I truly enjoyed that and it was also rather funny.
Regarding the case for lowtech interstellar travellers: I can accept the idea that there is some symple physics trick to make a jump drive that we haven't discovered, fair enough. Just as in the story, there would also had to be a similar undiscovered trick for antigravity, since otherwise they have no way of getting to orbit.
Can you however build a ship that can resist vacuum and provide life support with 16th century tech? I guess perhaps your species might be able to live in vacuum, needs only a bit of heat and doesn't need to breathe gasses?
- - -
Anyway, I’d like to recommend another book. High Crusade by Poul Anderson[0]. In the first chapter pacifist aliens land in 13th century England and get immediately massacred by the locals. Now the king has a starship.
You're conflating micro and macro price movements. Sure, you can't predict black swan events with technical analysis, but it wouldn't be out of the question to predict short term fluctuations based on a mixture of market psychology and herd thinking.
I don't employ any technical analysis in trading, nor am I a strong advocate, but technical analysis is more about reactionary psychology than about predicting the future. In the most micro sense, the market is dictated by single individuals buying and selling stock, and in the broadest sense, it's a statistical result of millions, or billions of unknowns. Those are two totally different games.
To add, it provides a general measure of sentiment. It's a measure of actions taken rather than words. It's not a crystal ball, but does give a slight edge if you figure out which indicators to use.
I've used it for years and had very good success when I was actively trading (traded longer timeframes. Usually made weekly/monthly trades) to the point I was able to live off of it. The last couple of years I've taken off from trading to focus on developing my business, but will likely jump back once that launches and is in a good enough place, as it's both a time and emotional drain.
While I agree that TA is an invalid method for investing, I disagree with your take on it. What you describe are, charitably, outlier events. Statistical models generally aren't useful for predicting singular events, even if they aren't outliers. So you aren't really invalidating the use of TA with that argument.
You may think so, until it’s revealed that one of the bidders makes all bidding decisions by roll of a die / flip of a coin (you can invent some scheme for doing so).
The goal of technical analysis is to find pools of supply and demand at different price points, much like a clustering algorithm. It is reductive by its very nature but it's still a useful approximation because big interpretable clusters can absolutely drive price action and it's helpful to have a model for that. On a good day, you might get the biggest 2 or 3 clusters correct to within 10% and miss several others totaling 1000000x the size and importance of a hypothetical bozo trading based on a coin flip, and this highly imperfect outcome would still be enormously valuable and worth pursuing.
It's like politicians assembling a platform by considering the sizes and interests of voting blocs. This strategy was always going to be a reductive, imperfect, noisy mess but that doesn't make it irrational or useless.
It depend on whether they are also trying to use each other's bids to determine each other's psychology. At that point you may be observing a recursive system (I know that he knows that I know that he knows...) and can't be sure how many layers of reasoning each actor is employing.
Not with the prices alone. You need a context to use the prices for insight. They could literally be using the bids as a channel to communicate with each other because they are colluding, or as a performance for observers.
Behavior is different at different scales but I see your point. I typically walk to my car after an event, but if I see a group take off running with urgency I will probably do the same thing. Know what I mean?
Does non-technical analysis work for those events? Nope. TA can incorporate all currently known information about the likelihood of those events though. As they say on WSB, it's priced in.
Doing that is not technical analysis. Technical analysis is predicting stock prices strictly from the previous price history.
The steel-man position for technical analysis is that humans are herd animals, who tend to do things in similar, predictable ways. Such as, people who pick stocks like to sell and take profits after the stock reaches some nice, round number. When enough people do this, it can result in repeatable, predictable patterns in prices.
The massive caveat there is of course that while that might be true in the absence of outside stimuli, any news always trumps that. It doesn't matter how nice your patterns are, if it comes out that the company somehow did worse than expected, it's price is going to fall, and if it did better than expected, it's going to rise. Trading on technicals might be something that works if you are really fucking careful, but it seems to me like you are picking quarters in front of a steamroller, in a way. The actual reliable profits from technical analysis are always going to be small, and the risks of investing with no understanding of the fundamentals seems fraught.
> Technical analysis is predicting stock prices strictly from the previous price history.
Is this true? For example, is using AI models that take into price history, current and historical events, insider and/or institutional trading, analyst opinions -- would that not be considered technical analysis? I know this is just a matter of semantics, but curious as to what is considered TA nowadays.
Not really. When you add that much information that gets into 'analysis' rather than 'technical analysis'. A bit like medicine vs traditional medicine.
Technical analysis is specifically the stupid line drawings you see on stock charts that try to predict trend lines from the motion of the previous lines
Burgers' equation models traffic flow. Traffic comes from humans and humans behave in patterns. What difference does it make if it's a PDE or a teacup?
Because you are not drawing a teacup, you are looking at price patterns. IF you believe in efficient markets, price should be an indicator of "all currently known information"
If you believe in efficient market and things being priced in then there is no past price information that can affect future price as future price is entirely martingale. You can’t both believe in technical analysis and in market efficiency.
You can predict situations like crop failure - if you're an expert in agriculture and have access to data that impacts crop yields.
Some things, like weather and major political issues are easily obtained bits of information and are high-impact events. Data like satellite imaging of fields are available at a price and will tell an awful lot about what best-case production figures will be like. Lastly, insider information is around too - asking companies what they planted, sales data on seeds/fertilizer, etc. If a disease is hitting one company, it's probably hitting a lot of them, especially true with live stock.
People do pay for this information, and they use it to trade futures. Which is how it eventually gets "priced in."
Forecasting is a huge factor in a successful agriculture business. Which is part of why a lot of niche industries tend towards vertical integration - it's much easier to forecast demand when you produce a lot of derivative products. Buy your own stuff; if you planted a surplus then create new products / markets, if there's a shortage, cut production on lower profit items.
It can be an input, but is it a significant input?
If I know for a fact that a company is about to collapse and I establish a short position, will anybody notice? It depends on a lot of factors, but I could make an assload of money on puts without nudging the market.
Technical analysis is all about answering the question: are you able to filter who know something vs those that don't merely by looking at a buy signal? The answer: no, not really. Insiders acted well in advance of the SVB issues being made public, and there are lot of other examples of people making guaranteed bets that weren't obvious in the technicals. The noise is just too much.
By the time the signal is clear, the market has "priced" it in, and now you're just trying to predict the random actions of the know-nothing crowd.
Technicals are always after-the-fact. Since enough people with deep knowledge have to have already moved on that information in order for the market to have reacted to it.
It's much like trying to predict a fire by watching a crowd running away and hit the fire alarm. You didn't predict the fire and you can't have predicted the fire. Maybe you could have noticed one person booking it for the exit and follow along, but you still didn't know what was happening and were going on blind faith.
>If I know for a fact that a company is about to collapse and I establish a short position, will anybody notice? It depends on a lot of factors, but I could make an assload of money on puts without nudging the market.
Assuming you are the only person who knows for sure that this company will collapse, technical analysis is the wrong tool to use and all other strategies, experts, big money players will also fail.
Assuming you aren't the only person who knows for sure that this company will collapse the price will reflect this knowledge.
>It's much like trying to predict a fire by watching a crowd running away and hit the fire alarm. You didn't predict the fire and you can't have predicted the fire. Maybe you could have noticed one person booking it for the exit and follow along, but you still didn't know what was happening and were going on blind faith.
It's nothing like that at all, and reasoning with analogy is pointless because the discussion becomes about the analogy not the issue at hand. But I will try. If people running away is people selling stock, it doesn't matter if there is a fire or not. Selling stock impacts price regardless of if the event is real or not. And as you know, having been in a building where the fire alarm was activated, people don't rush to the exists. Some small number leave right away. The vast majority won't enter the building. The remaining people look to others and as more leave the momentum builds until the mass moves out.
And once the fire department arrives and gives the ok they will go back in. But since we are talking about stocks, not all buyers won't be waiting for that. They'll have put their money elsewhere, money may go back into the stock slowly. Obviously this won't work on the top 3 most popular stocks in a raging bull market that everyone is paying attention to. In that case, people will be waiting to jump back in.
By the time you have this information and are reasonably act on it, it has already been priced in by participants with deeper pockets, better access, and rapid processing of relevant information sources.
The prediction of a crop failure isn't a binary event and money isn't deployed in a binary strategy. It becomes more or less likely over time.
Also, the situation is very dynamic with an unknown and very large number of variables, even with something as seemingly simple as a crop failure. Such as such as a countries deciding to ban exports to secure food security, lack of access to fertilizer due to sanctions against major fertilizer producers in wheat producing countries, the politics of national and local water rights.
All these factors have parties with inside information acting on them (see members of congress trading stocks) which no player with big pockets will have access to but is all reflected in the price.
Nobody is saying these things are binary events. But betting that you're going to more accurately guess the odds of these events than dedicated players with enormous sums of money behind them and who are supported by professional research teams is… maybe not a strategy with positive expected value.
It "might work if nothing significant happens" but that is just a weasel way of saying it doesn't work.
Now if you're saying you decided to invest in something and you're trying to get an entrypoint and you spot a moment of large deviation from the mean and you use that to influence your operational buying mechanics like how you space out your buys or time them, that I can definitely see. But online you always see these two very different things mixed together.