Open any financial terminal and pull up the rating list for Hong Kong large caps, and you will see a spectacle: one company, Tencent, tracked by more than forty investment banks — forty-two of them rating it Buy or Strong Buy, two Hold, one Sell.
Forty-five analysts study the same company. Why can almost none of them bring themselves to say the two words: don’t buy?
Some say Hold is the neutral stance, refusing to pick a side. I find the idea of Hold strange — if you hold, aren’t you buying? A stock you bought yesterday now carries a Hold rating, meaning keep holding it; keeping holding it is staying long. The people who truly refuse to take sides sold long ago. So this list contains no neutral party: forty-four longs, and one person too frightened to speak.
Analysts’ track records are actually public. Dedicated platforms tie every analyst’s historical calls to the subsequent share price and score them; anyone can check online. The result: even the star analysts at the biggest banks hold a hit rate stuck between fifty and fifty-five percent — barely better than a coin flip — with anyone above sixty percent a rare exception. The statistics sit right there, and nobody is ever held accountable for being wrong. That is exactly the point: even when the record is one search away, why do they still have a market?
Put yourself in the analyst’s chair and it becomes clear. Your price target was 550, and the market has now made a fool of you while the whole internet cheers that technology is changing the world. If you don’t raise your target now, your boss thinks you lack vision and your client thinks you’re a dinosaur. Run the thought the other way: if you had issued a Sell and the stock rose twenty percent, you’d be called blind. Bullish in rallies, bullish in drawdowns dressed as caution — that isn’t a judgment problem, it’s a livelihood problem. That is the real chase-highs-sell-lows, except it happens inside research reports.
So a rating should not be read as research; it should be read as a product. Its customers are not retail investors but listed companies and big money: Buy maintains the relationship, and Hold is an escape pod delivered to large clients — the people who learn it’s time to leave before the crowd need an article to keep everyone else seated. The lone Sell is the one that takes real courage, which is why it is nearly extinct.
For me the conclusion is simple. I don’t listen to what analysts say; I watch what companies do: are they paying real dividends, buying back and cancelling shares, is management putting its own money in or telling a story out. And one plainest question: if this stock goes nowhere for three years, how much cash will it hand me each year?
Ask that question and the forty-two Buy ratings all fall silent. Ratings answer how others see it; cash answers what I do. Investing can only live on the second one.
