What Makes You Think You Can Get Out Before the Big Players?

Start with one question: when you decide whether to buy or sell a stock or a coin, what are you actually going on?

Most likely you open a trading app and look at the candlestick, a few indicators, the order book. You look at these things every day, but where they come from, and whose hands they passed through on the way, almost nobody stops to think about.

This piece is about exactly that: how many hands has the data you are looking at already been through.

A single daily candle in your app takes thousands of trades from one day and compresses them into four prices — open, high, low, close — plus a volume number. The indicators you rely on, moving averages, MACD, KDJ, are just another fixed formula layered on top of those four prices.

What you are staring at is a thumbnail: the market after it has been sampled and squeezed down. The raw tick-by-tick trade stream is far bigger, and it is not handed to retail traders in its original form.

Investor-education material from the brokers puts it plainly: a technical indicator is the result of processing raw price and volume data “by a fixed method set in advance.” Anything that has been processed has necessarily thrown information away.

I built a small quant system of my own once and ran backtests. On the backtest curve the signals looked clean — the golden crosses and death crosses lined up nicely. The moment it hit live trading, it stopped matching. One Saturday afternoon I pulled the backtest log and the live log side by side and went through them line by line, and found that on the day the indicator showed a golden cross, the live price had already been grinding down for days.

The signal was not miscalculated. The problem was that the thumbnail it stood on lagged the real market by a stretch. That is not a copy-paste error in my formula; it is that I had only a thumbnail in hand and was trying to reconstruct the whole room from it.

What makes it worse: some of the lines on that thumbnail were drawn there for you on purpose.

China's A-share market has a term, “marking the close”: in the window before the bell, a concentrated burst of buying or selling pushes the closing price up or holds it down. In the manipulation cases the CSRC has published, “manipulating the closing price at the close” is a tactic that gets named again and again — years ago people were fined hundreds of thousands of yuan for exactly this move.

The closing price is that candle's face. The next day, plenty of people make decisions off that face: closed above the moving average, go long; left a long lower wick, must be support. But if that face was hammered out in the last ten-odd minutes before the bell, the judgment you build on it is standing on a pose someone else struck.

You think you are reading the market's intent; you are actually reading the intent the other side of the trade wants you to read.

This happens more easily on small-cap stocks and small coins. The float is small, few people are trading, and not much money is needed to shove the price a long way and shape the candle into whatever is wanted. On large caps and major coins, drawing a single candle costs far more, so they stay relatively clean. But what ordinary people actually pile into, size on, tends to be exactly those “cheap, lots of upside” small names — a float small enough for someone to push around at will.

Which brings the real question to the surface: if the data is secondhand and may have been staged, why do so many people still believe they can make money by watching the tape, chasing tips, “following the right person”?

Because everyone assumes they sit near the front of the information chain. The line in the tip-follower's head is: “I know the big player is pushing it, I ride along, and when they're about to run, I run first.”

That line is the flaw itself. If you were really near the front of the chain, you would be the one drawing the candle, not the one reading it. The fact that you can see the “big player is pushing” signal is precisely what tells you the signal has already been let out for you to see.

Back in my dev days I had a realization: an engineer who only ever calls someone else's packaged interface is basically stuck when a system-level bug shows up, because they have never looked at how the underlying layer is implemented. They have the docs, not the source. Watching the tape to try to beat the people staging it is the same thing — holding the docs, racing the people who write the code.

Someone reads the Odyssey as a story about following a big player and running: a boatload of people follow the leader, hoping to reach shore before the storm, and find out the one left on the boat is themselves. Once you see that layer, you know where to place yourself: not the “insider who can get out early,” but the “outsider who might be left holding it at any moment.”

I am in no position to lecture from the shore. Back when I was building that system, I too genuinely believed in the indicator signals for a while, sure that if the formula were a bit finer and the parameters tuned a bit harder, I could pull something out of that thumbnail nobody else had seen. It took me a good while to admit the direction was bent from the start: however fine the math, what goes in is still secondhand data.

These days when I see “just watch the tape and you'll make money” or “follow so-and-so,” I don't really argue. I have just gotten clear that the real dividing line is whose data is firsthand, and it has little to do with who studied technical analysis better. Everyone else, me included, is tracing lines on a chart someone else already drew.