Every valuation is a guess wearing a suit. The DCF spreadsheet with its twelve decimal places, the comp table sorted to the third significant figure — these are acts of theatre that perform precision while delivering something far more fragile.
The precision fallacy
When an analyst produces a target price of $43.27, the number implies a knowledge the model does not possess. The inputs — discount rates, growth assumptions, terminal values — are themselves estimates with wide error bars. Feed uncertain inputs into a deterministic machine and the machine does not remove the uncertainty. It hides it.
The honest output of any valuation is a range, and usually a distressingly wide one. For a mature utility, the fair range might be ±20%. For a biotech with one Phase III readout pending, ±60% is optimistic. The difference between a good valuation and a bad one is not the tightness of the point estimate. It is the honesty of the range and the clarity about what would move it.
Why precision feels good
Precision is a social signal. It tells the client, the committee, the counterparty: I have mastered this. The human brain rewards the feeling of closure that a single number provides. This is why analysts round to two decimals and why forecasters who give ranges are perceived as evasive, even when they are the only ones telling the truth.
The market itself is the great counterexample. Prices are precise — $43.27 is exactly what you pay — yet the value behind them is a fog. The market reconciles precision of price with uncertainty of value through one mechanism: liquidity. Liquidity lets you change your mind. A valuation that ignores liquidity is a valuation that ignores the only exit door.
Decision-relevant valuation
The purpose of valuation is not to know the number. It is to know what to do: buy, sell, hold, hedge, or decline. Every decision has a threshold. A buyer needs to know whether the expected value clears the price by enough to compensate for error. A seller needs the inverse. Both need the falsification conditions — the specific events that would make the thesis wrong.
This reframes the entire exercise. Instead of asking "what is this worth?", ask:
- What is the range of plausible values?
- What would have to be true for the high end?
- What would break the thesis entirely?
- How long do I have before the information decays?
The Forreast approach
We build valuations as decision instruments, not ornaments. Every report carries a confidence band, the assumptions that matter most, and the falsification conditions that would invalidate the thesis. We would rather be approximately right and know what would prove us wrong than precisely wrong with nowhere to hide.
The valuation trap is the belief that more decimals equal more truth. They do not. The escape is to value for decisions, not for display.