I get a lot of flack for this, but Al Brooks and other writers of chart analysis don’t really know what they’re talking about. Stock prices and chart patterns you see are the result of lit crossed orders and do not tell you the intention of the trades (remember for every seller there is a buyer) on each candle. Brooks has never worked or studied with market makers or institutional traders, also he rehashes the “common” thinking of retail traders without doing any study. Some key concepts that serves as the basis of his thinking that’s wrong are: There is intention behind every candle/tick — this is patently false, I’ll explain below Trading is not zero sum — he’s confusing trading with investing, trading is zero sum (actually slightly negative skew due to paying for bid/ask), investing isn’t The guys on CNBC know what they’re talking about — 90% false, most of CNBC is garbage (same goes for Bloomberg unfortunately) In the real world, trading has become very fragmented. Most larger investors cannot facilitate their trades across just one venue, and absolutely are not trading by tick or candle. This is because of the disintermediation trend (banks are no longer the major market makers in equities, regulation also drove this), the rise of alternative trading networks (dark pools and other exchanges, lit or dark), and the role of hft firms (as market makers and speculators). When you think about trading, you should first seek to understand it as being an exchange that facilitates the logistics of investing. ABC fund wants to buy XYZ stock, and they submit an order to the exchange (or source of whoever may have the shares). That order can take time to fill — from seconds to hours to weeks. The intention to buy is driven typically by an analysis of the growth or value prospects of the stock (most fund investors rely primarily on equity research, internal or otherwise, to develop a view on a stock). From there, the fund will usually start a position (if it’s a billion dollar fund then even 1% is still $10MM in value) and will usually set a max weight of 5-10% of the portfolio in any one name. To start a position the fund’s trader will first look at displayed liquidity to gauge what the transaction cost might look like. Transaction cost is primarily the slippage — the amount the shares move as you are trying to fill the order. If you look at the bid ask of any stock you’ll see price x size. So when you, as a larger trader, try to buy, you are lifting the ask x size. If the displayed size is 1k and your order is 10k, you will easily take the nbbo’s offer, and start to get worse prices. Imagine you’re on Amazon and are trying to buy a book, and sort it by price. The size at best price might 1-5 copies— so if you buy 100 copies the price will get worse for you. Because of this challenge, traders mainly fill their orders using algos that seek liquidity or that manage the size of the order as a % of volume in order to hide what’s going on. If you overlay all these different approaches, you begin to realize that at a very short time scale, the trends will look random but that as you “zoom out”, trends in price and volumes are correlated. What this also means is that you have no edge in shorter time frames (unless you are the guy dumping or buying stock in size). On the chart, the price will oscillate wildly. Once you begin to add on the normal volumes of the stock, you get a noticeable, incremental trend. Because large investors can take days/weeks to fill orders (and because there are lots of them moving in and out of stocks), you get momentum in both the short and medium term. For example, 5-day momentum typically reverses in the subsequent week as does 20-day momentum (at the margin more funds were buying/selling and it takes them that long to trade). With that being said there are possible trades you can make using charts: Momentum is a good one, simply if price has risen over the last 6-12 months (in absolute and relative to local index), it tends to persist In the short run momentum reverses (5 and 20 day), there is possible intraday momentum but there is actually a negative drift during trading days for most stocks Market anomalies such as price spreads between companies doing a merger/acquisition when one stock moves, small cap anomaly due to funds seeking liquidity at the end of day, slow diffusion of information anomaly where large cap stocks are more efficient than small cap so you can trade the small cap on large cap news, etc. but these all require more work to figure out the anomaly and how to harvest the returns from it Tl;dr Al Brooks never worked on a trading floor, nor has he apparently ever spoken to someone who has, so his understanding of how trading works is factually incorrect. I would avoid reading his books or spending time trying to figure out “price action” at this level. I’d recommend that you pick up a basic book on corporate finance first to understand how managers of companies buy/sell/issue stock and debt, read a book on market microstructure to understand the logistical considerations (different exchanges and ATNs and how it all consolidates on the chart making the chart super noisy), and read about the investment process of institutional investors (can read books written by them or training material for their jr staff). Once you have the foundation you should then start reading leading papers on major anomalies you’ve heard of such as the momentum, value, and size factors. From there you’ll be able to learn about various anomalies or discrepancies and can then do your own research to see if there is an opportunity. This is the path of successful traders.
Most accurate takedown of Brook's general BS. However, he does one thing well: he categorizes days and journals conditional probabilities. This is his doctor training at work. There are gems to be found here but his "algos do this, candles do that because of X" is total bullshit, as you have said.
You describe hft's as well as make the above claim. Given the preponderance of sub-penny decimalization, front-running, complex order types and dark pools - the operational mechanics of the hft domain are not congruent with your claim. My understanding might need to be updated with details but the overall sentiment has not changed. I know of the above only through Flash Boys which is almost 7yrs old now.
My sympathies. I read Bar by Bar and I own Brooks's video course. Bar by Bar is like reading language you don't even no alphabet. Video course is clear and basic. Tell Brooks to credit you the Bar by Bar book toward your purchase of video course.
It's a step in the right direction but IMO it's a little outdated and unsophisticated. Better than Al Brooks by lightyears, though. Have you read Margin of Safety? Or anything by Joel Greenblatt?
HFT's don't have an intention per trade. HFT's primarily act as market makers and internalizers. I'll explain how it works. Market making: Top of book quote is the NBBO -- say 10.15 x 500 and 10.17 x 200. I'm a random trader and I lift the offer (10.17) but my order is for 500 shares. If you were to look at level 2 offers you'd see: 10.17 x 100 10.18 x 200 10.25 x 500 10.28 x 200 So what happens is that I first lift the NBBO (10.17 at 200), then take the next 100 at 10.17 and after that I experience "slippage" and fill the rest at 10.18. In this scenario, you are an informed trader -- you know that you have an order that is larger than the top quote. Market makers do not yet know, but they will once you get your first fill at 10.17 x 200. As soon as that happens, the rest of them will attempt to adjust their prices higher. Larger HFT firms will use sub-penny decs to jump in front of queues to cross orders. The HFT does not have a view on the stock and does not care about a candle or pattern forming lol. All they care about is adjusting their price in the face of changing levels of volume. Internalizing: Citadal is a wholesaler that can fill orders between "the spread". In the above example, you see the NBBO as 10.15 bid / 10.17 offer. Citadel tries to match market orders from customers before it reaches an exchange. This, also, has nothing to do with charts and patterns and "bulls on the 5 second candle" lol. It has everything to do with the flow of orders and adjusting prices quickly to improve fills and collect a fraction of the spread. Citadel, Virtu, and other large market makers need a high amount of uninformed order flow. So they typically back your discount brokers (TD Ameritrade, ETrade, TradeStation, etc.) to make their commission costs cheap or $0 in order to increase the amount of flow they receive. Higher flow means higher odds of crossing trades within the network vs. having to "give up" the trade and send it to an exchange. So yes, Citadel would love it if you kept buying and selling on 1 second charts lol.