Weighting

Weighting is the practice of multiplying each factor's score by a numeric "importance" before adding. CF rejects arbitrary weighting because:

1. Weights are made up. Where does "price has weight 0.3" come from?
2. Weights are not in any dimension. They're abstract numbers.
3. Multiplying across dimensions is undefined. Weight × dollars = ?
4. The result is a meaningless number. "Total score 73.4" doesn't correspond to any goal.

What people actually do

CF's analysis of "people use weighted factors":

"People Use Weighted Factors" — most decision-makers (and most decision-making software) use weighted sums without realising they fail for cross-dimensional factors.

CF doesn't say people are stupid for using weighted factors. It says the approach is broken; use breakpoint-based pass-fail instead.

What to do instead

• Pass/fail at breakpoints. Most decisions reduce to this.
• Convert to same dimension if possible. "Is this fast enough?" rather than "how fast?"
• Don't combine across dimensions. Treat different-dimension factors separately.

When weights do work

• Within a single dimension. "$X per pound" makes sense.
• In clearly-defined composite metrics. "GDP" combines sub-factors deliberately.
• When the conversion is real, not made up. "How many yen equals one dollar?" has an answer; "how many dollars equals one unit of cuteness" does not.

"CF's solution is stop adding factors together. combine with multiplying. multiplying is problematic in general, e.g. if you multiply -20d with 5c you get -100dc which is not useful b/c the number is in terms of a multi-dimensional unit of cuteness-dollars." — criticalfallibilism.com