Thread by @wmthomson22

1/ @alixpasquet suggested that Soros on Soros had a clearer communication of Soros's thinking than his book Alchemy of Finance, so I have been reading it. On page 93, Soros states that an investment idea he has is a hypothesis, not a prediction.

The difference will save — or cost — you money. 🧵

2/ Good investors do not make predictions; predictions are for talking heads and the sell-side. Investors have hypotheses tested by reality and are subject to rapid change based on events.

2/ A prediction sounds like “Oil will be $120 by year-end.”

It’s definitive. It’s performative. It’s meant for headlines, not portfolios.

3/ A hypothesis sounds like this: “If OPEC+ extends cuts & demand stays above 102 mbpd, Brent likely trades +$90. If cuts break or demand slows, thesis is wrong, this thesis is expected to be valid until X date. It’s conditional. Falsifiable. Updatable.

4/ Predictions lock you in.Your ego gets tethered to a point-in-time call.

When reality shifts, you defend the call instead of defending your capital.

5/ Hypotheses free you. You’re running a test. Inputs change → thesis changes.
No shame. No ego. Just probabilities.

6/ Why investors need hypotheses:
•Faster feedback loop
•Explicit risk triggers
•No sunk cost fallacy
•Decision-making rooted in what is, not what you wish it was

7/ Talking heads sell certainty.
Investors trade flexibility.
The first gets airtime.
The second gets paid.

8/ Takeaway:
•Predictions are static declarations.
•Hypotheses are dynamic frameworks.

Trade the tape and reality not your ego.

9/ If you’re in the markets:
Write down your thesis. Define exactly what disproves it. Review often.
Reality is the ultimate P&L arbiter.