Filtered dossier

Paradox & Proof

Statements that stay true while contradicting themselves.

  • 12 / 12
  • No visual on file
    Paradox & Proof Entry #0162

    The reason a queue can move faster with fewer servers

    Adding a server lets customers choose a line. Choosing a line means choosing wrong, and the time spent switching exceeds the time saved by the extra server.

    Adept
    The hidden part #0162

    Three queues offer choice. One queue offers coordination. The coordination saves the time the choice was wasting.

    Open file
  • No visual on file
    Paradox & Proof Entry #0161

    Why unanimous agreement lowers confidence in a verdict

    Perfect consensus in a large group implies a shared error. The very agreement that looks like certainty is evidence against it.

    Novice
    The hidden part #0161

    Perfect agreement among many is not strong evidence. It is evidence of a shared input, which may be the truth or may be the error.

    Open file
  • No visual on file
    Paradox & Proof Entry #0160

    The reason a prediction can cause its own failure

    The forecast changed the system it was forecasting. The prediction was accurate about the world before it was announced, and inaccurate about the world after.

    Master
    The hidden part #0160

    A prediction that enters the system it describes is no longer a prediction. It is a participant, and participants change outcomes.

    Open file
  • No visual on file
    Paradox & Proof Entry #0159

    Why the safest route is the one nobody takes

    Safety attracts traffic. Traffic erodes safety. The route that was safest when empty is the most dangerous when full, and the dangerous one is now empty.

    Adept
    The hidden part #0159

    Safety attracts use. Use erodes safety. The safest route is the one whose conditions deterred you from choosing it.

    Open file
  • No visual on file
    Paradox & Proof Entry #0157

    Why a growing company can have shrinking margins on every sale

    Total profit and profit per unit are separate quantities that can move in opposite directions. Growth is reported in one of them and health is assumed in both.

    Novice
    The hidden part #0157

    A total is a rate times a volume. Growth in the product says nothing about the direction of either factor.

    Open file
  • No visual on file
    Paradox & Proof Entry #0153

    Why adding a better player can weaken a team

    Individual talent and collective performance answer different questions, and the gap between them is exactly where most recruitment strategies fail.

    Novice
    The hidden part #0153

    Roles matter more than raw quality. The team that absorbs a stronger part may lose the coordination that was its real advantage.

    Open file
  • No visual on file
    Paradox & Proof Entry #0155

    Why the average person has fewer friends than their friends

    In almost every network, most members are less connected than their own contacts are on average. The comparison is rigged by how contacts get sampled.

    Master
    The hidden part #0155

    Your contacts are a sample weighted by connection count. The comparison is against the well-connected by construction.

    Open file
  • No visual on file
    Paradox & Proof Entry #0154

    The reason a fair coin produces unfair-looking runs

    In two hundred fair tosses a run of seven identical results is likely. People asked to fake the sequence almost never allow one, which is how fakes are detected.

    Adept
    The hidden part #0154

    Randomness clusters. The expected longest run in a fair sequence is far longer than intuition permits, so real data looks rigged.

    Open file
  • No visual on file
    Paradox & Proof Entry #0156

    The reason improving every hospital can worsen the statistics

    Every department improved. The overall figure got worse. Both statements are true at once, and the reason is one of the most useful traps in data.

    Master
    The hidden part #0156

    A total is not a neutral summary of its parts. It is the result of a grouping decision, and that decision can point the conclusion in either direction.

    Open file
  • No visual on file
    Paradox & Proof Entry #0158

    The reason more information can lower accuracy

    Adding weakly relevant detail to a judgement raises confidence and often leaves accuracy flat or worse. The extra material dilutes the signal it was meant to sharpen.

    Adept
    The hidden part #0158

    Non-diagnostic detail is averaged in rather than ignored. Confidence tracks volume of information; accuracy tracks its quality.

    Open file
  • No visual on file
    Paradox & Proof Entry #0155

    Why the average person has fewer friends than their friends

    In almost every network, most members are less connected than their own contacts are on average. The comparison is rigged by how contacts get sampled.

    Master
    The hidden part #0155

    Your contacts are a sample weighted by connection count. The comparison is against the well-connected by construction.

    Open file
  • No visual on file
    Paradox & Proof Entry #0152

    The reason more roads can mean slower traffic

    A new road is opened and the journey takes longer. Nobody made a mistake. Every driver chose correctly, and that is precisely the problem.

    Adept
    The hidden part #0152

    Braess’s paradox is not a story about bad drivers. It is a demonstration that a network of perfectly rational choices can settle on an outcome worse than the one it started from.

    Open file