Zhihu Taught Me a Lesson

I recently came across an answer on Zhihu, and after reading it I felt a bit struck: the quality of discussion in that community seems a lot higher than I remembered—or maybe it was always high, and I just used to keep running into the low-quality part of it.

The author of the answer is Simon, and the original is here. Using the question “Why can’t most people achieve a major breakthrough?” as a starting point, he explained how he discovered an investment blogger worth following: this blogger had been heavily invested in NVIDIA since 2017, began positioning in Micron in 2024, loves skiing, did 50 push-ups in 45 seconds on his 54th birthday, and has recently been arguing online every day with people shorting Micron.

How did he find him? Simon summarized a method: first, find people whose values are close to yours—people who are also bullish on U.S. stocks, AI, and NASDAQ; then look back at whether their analysis at major market turning points was objective, and whether their later judgments were validated; if they were proven right and the returns were decent, keep following them and use their views as cross-checks when making decisions.

Reading this answer, my thoughts went through three rounds, and each round was different.

First Pass: Sounds Pretty Right

On the first read, I thought: that makes sense.

The investment world is full of noisy and fragmented information. Instead of fumbling around blindly by yourself, it’s better to follow people with a verified track record; first check whether their past calls were accurate, then decide whether to trust them. The logic is pretty smooth. Cross-checking and reviewing history also sound rigorous enough. And that blogger was heavily invested in NVIDIA back in 2017 and started positioning in Micron in 2024—he nailed both moments. That alone is enough to make you want to follow him.

Second Pass: Actually, Maybe Not

On the second read, I started to feel something was off. When you break Simon’s method apart, it’s full of holes:

First, “similar values” is already a pre-filter. If you’re looking for people who are “also bullish on U.S. stocks and AI,” then you’ll only see bullish people. In the eyes of bulls, the world is always full of bulls—this isn’t cross-checking; it’s building yourself an information cocoon. One comment put it perfectly: “If you’re bullish, you should read the bears; if you’re bearish, you should read the bulls. Analyze the other side’s logic to test your own view.”

Second, “major market turning points” and “whether they were validated” are both defined after the fact. Which point counts as “major,” which judgment counts as “objective,” and which return counts as “decent”—these standards can only be written from the far side of the outcome. During the process, you have no idea whether what you’re experiencing is a “major turning point” at all.

Third, “eventually earned decent returns”—survivorship bias. The people you can look back at are all the winners who survived. Where did the losers go? Someone in the comments asked the question for me: “How do you know they didn’t delete the articles where their predictions were wrong?”

Third Pass: So It Was All in the Rearview Mirror

On the third read, I finally saw the fundamental problem: this method only works in the rearview mirror.

By the time you discovered that blogger, NVIDIA had already risen dozens of times over; every “historical judgment” you verified was hindsight. When you’re sitting in the present making decisions, there are no “already validated people” in front of you—only a crowd of people saying different things, none of whom knows the future. So-called “retrospective validation” is essentially taking outcomes that have already happened, using them to score past judgments, and then pretending that score can predict the future.

It’s like driving by looking in the rearview mirror: the road behind you is perfectly clear, but that’s the road you’ve already driven. Whether there’s a pothole ahead, the mirror won’t tell you.

The comment section summed it up in four words: “driving by the rearview mirror.” Someone else called it “hindsight 10/10”—perfect score in postmortem analysis, zero score in prior prediction. Every year, someone beats Warren Buffett, but in the end it’s still Warren Buffett who wins; someone who calls ten coin flips in a row doesn’t mean they’ll be right on the eleventh.

The Comment Section Finished the Critique for Me

Honestly, the most valuable part of that answer wasn’t the main text; it was the comment section. The things I quoted above all came from the comments: driving by the rearview mirror, circular reasoning, information cocoons, coin flips, mistaking luck for skill, deleting articles with wrong predictions—the commenters laid out all these concepts one by one, and they did it more brilliantly than the answer itself.

The most painful one was this: “What’s the difference between the second paragraph and circular reasoning? You should be looking for someone whose values are not that similar to yours but whose strategy has a high hit rate. The people you choose are actually the same as you; even if you didn’t read their views, you would still make the same choices.” In one sentence, it exposed the essence of Simon’s method: you were never looking for “the right person”; you were looking for your own echo.

A few other comments suddenly shifted tone and felt especially real: “If I buy now, is there still a chance?” “Just tell us what to buy now.” “The real question is: what should we buy now?” One moment people were discussing methodology, and the next they were already anxiously asking for tickers—probably a live demonstration of “reading it and still learning nothing.”

Reflections

This answer stayed with me for a long time. Its sharpest point isn’t teaching you how to pick bloggers; it’s forcing you to admit one thing: the hardest part of discernment is recognizing the hindsight perspective.

We too easily mistake “someone who has been validated” for “someone who can predict,” mistake “luck” for “skill,” and mistake “what looks inevitable in hindsight” for “what should have been known in advance.” These mistakes aren’t stupidity; they’re the brain’s default setting—because the world in the rearview mirror is always much clearer than the world beyond the windshield.

So the next time you see someone showing off “I said this long ago,” flaunting a perfect record, or displaying a history of flawless judgments, don’t rush to believe them, and don’t rush to attack them either. Ask one question first: were these judgments written at the beginning, or reverse-engineered from the ending?

And as for when you should listen to others, and when you should get out of the car and decide for yourself—the rearview mirror won’t give you that answer.