It was 2:47 in the afternoon and I was staring at a single candle that hadn't closed yet.
SPY had drifted up to 512.40 for the third time that month. I'd marked that level in June after it turned price twice, drew the line, and mostly forgot about it — the way you forget about a fence until something runs into it. Price came up slow, almost lazy, the way a stock moves when nobody's in a hurry. Then it touched 512.40 and the candle started building a long upper wick. Sellers showing up right where I expected them to. I didn't do anything. I just watched the wick grow, watched the body get pushed back down toward the open, and waited for the candle to close so I'd know if the rejection was real or if it was going to fake me out and close near the high instead.
It closed with a wick twice the size of the body. I took the short. Small size, tight stop above the wick. It worked, not because the candle was magic, but because the candle told me something true about what had just happened at a spot I already cared about.
That's the whole article, honestly. Everything below is just me explaining why.
The lookup-table problem
Most candlestick education is a glossary. Doji means indecision. Hammer means reversal. Bullish engulfing means buyers are back. Memorize the shapes, memorize the labels, scan the chart for matches. It reads like a field guide to birds — this wing pattern means this species — except price isn't a bird and it doesn't care what you named it.
I used to trade this way. Long before I deleted every indicator off my charts — that's a separate story, a Sunday afternoon and fourteen tools I finally admitted weren't helping — I had a candlestick cheat sheet taped next to my monitor. I'd scan five, six tickers a day looking for shapes that matched the pictures. A doji here, an engulfing candle there. I traded the shape, not the situation the shape happened in.
Here's what that approach misses: a candlestick isn't a signal. It's a record. Every candle is a compressed history of one session's argument between buyers and sellers — who opened stronger, who pushed harder, who ran out of ammunition before the close. Read correctly, a candle tells you who won a specific fight at a specific price. Read as a shape to pattern-match, it tells you nothing, because you've stripped out the only part that mattered: where the fight happened.
Three patterns, read as behavior
Let me walk through the three I actually use. Not as a signal list — as what was physically happening between the two sides.
The hammer, or pin bar. Price opens, sellers take control and shove it down hard — sometimes a full percent below the open. For a while it looks like sellers own the session. Then buyers step in, absorb the selling, and drive price back up so the candle closes near where it opened, sometimes above it. What you're looking at is a rejection. Sellers tried to win and got overwhelmed before the bell. The long lower wick is the receipt — proof of how far price traveled before buyers said no. It's not "the market is bullish now." It's "one specific group of sellers lost a specific fight, at a specific price, today."
The engulfing candle. One session's range completely swallows the prior session's range — the whole body, open to close. This is one side winning decisively, in a single sitting, with enough force to erase the entire previous session's territory. A bullish engulfing candle means buyers didn't just show up, they ran the table. That's a different kind of information than a hammer. A hammer is a rejection at the edge. An engulfing candle is a rout in the middle.
The doji. Open and close land almost on top of each other, wicks stretching out both directions or barely at all. Neither side won. Both sides showed up, pushed, and canceled each other out. A doji is a shrug. It's the market taking a breath — not bullish, not bearish, just paused. On its own it tells you almost nothing except that conviction dried up for one session. That absence of conviction is itself information, but only in certain spots.
Notice none of these three descriptions contain a prediction. They describe what already happened. That's the whole point.
The part that actually matters
Here is the piece the glossary approach leaves out completely: a hammer in the middle of nowhere is noise. A hammer at a level is information.
Go back to that SPY candle at 512.40. If that exact same wick had formed at, say, 507.10 — a price with no history, nothing I'd marked, nowhere price had reacted before — I wouldn't have looked at it twice. Same shape. Same long wick, same small body, same technical definition of "hammer." Completely different meaning, because the context is different. At a random price, a rejection candle just means somebody blinked for one session. At a level — a price where supply or demand has already shown up more than once, a price I was already watching before the candle formed — a rejection candle means the level held again. That's not the same claim at all.
This is the entire reason I trade off a blank chart with horizontal lines and nothing else. The lines are where I've already decided the fight is worth watching. Support and resistance aren't magic either — they're just prices where enough buyers or sellers have shown up before that I expect them to show up again. Candlestick patterns are how that expectation gets confirmed or denied in real time. The level tells me where to look. The candle tells me who won when price got there.
Learning how to read charts, for me, stopped being about memorizing forty pattern names and became about answering two questions in order: is price at a level I already marked, and if so, what does the candle there tell me about who's in control. Drop either question and you're just watching shapes float past.
Where I lost money learning this
My first real losing stretch wasn't from picking wrong. It was from pattern-hunting everywhere. I'd scan ten charts a night looking for anything that resembled a hammer or an engulfing candle, and I'd find one on almost every chart if I looked long enough — because patterns happen constantly, at random prices, for no reason at all. I took a bullish engulfing candle on a stock with zero support underneath it, no level, no prior reaction, nothing. Just a shape I recognized. It failed within two days and kept failing, and over about six weeks that habit cost me close to four thousand dollars in an account that wasn't large to begin with. The pattern wasn't wrong. The location was meaningless, and I hadn't learned yet that meaning comes from location.
The fix wasn't a better pattern list. It was fewer places I was allowed to look. Mark the levels first, in advance, before the session even opens. Then only pay attention to candlestick behavior when price actually arrives at one of those marked prices. Everything in between — the shapes forming in open air, on no level, for no reason — gets ignored completely, no matter how textbook it looks.
What this isn't
Patterns fail. Often. A hammer at a level can get run straight through the next candle — sellers who looked exhausted show up again with more size, and the "rejection" turns into a pause before continuation. An engulfing candle at resistance can engulf again the other direction the very next day. I've had setups that looked identical to winners on my own chart, at levels I'd marked with real conviction, fail anyway. Nothing here predicts. A candlestick pattern at a level raises the odds that something meaningful is happening — it doesn't guarantee an outcome, and anyone telling you otherwise is selling something.
That's why the stop matters more than the pattern. I size small, I put the stop where the pattern's story breaks — below the wick, below the engulfing candle's low — and I accept that a meaningful number of these will fail even when everything about the setup was reasonable. Risk management isn't the boring part you skip to get to the good part. It's the part that lets you be wrong repeatedly and still be fine.
Close
Right now there's a candle building somewhere, at some level somebody marked days ago, and neither side has won yet. I won't know which way it closes until it closes. I'll just be watching the wick, the way I was that Tuesday afternoon, waiting to see who gives up first.
I trade with a blank chart and a few lines in Static, the free daily chart room run by Draw Lines Make Money. If this way of reading charts makes sense to you, you can sit in and watch how it's done live:
Join the free Static chart room →
Disclosure: that's an affiliate link — I may earn a commission if you join a paid tier later, at no extra cost to you. The free room is free.