4.6 The static crossover table. Anchor: L217

How long lying stays ahead (q0 = 0.002, ... After it add the scroll hint and open the wrapper; close the wrapper after the table's at L230 (before the "More contacts and more shared memory" paragraph at L231). 4.7 Section 5, the −26.42 correction. L248 is exactly

  • An unknown end. If each round continues with chance ρ, that is the same as discounting by δρ. At ρ = 0.99, deception is a net loss (about −26.5 in our run).
    • Keep the first two sentences and replace "(about −26.5 in our run)" with "−26.42 by the closed form D∞(ρ) (the Python Monte Carlo run gave −26.52, which is where an earlier "about −26.5" came from)" plus a .params span with the C2 settings and the ρ = 0.8 / 0.95 values (+3.53, +5.10). The history of the old figure is said out loud, so nothing is silently rewritten. 4.8 The 68/32 grid. Anchor: L264, the paragraph starting

    Plainly: across our 243-cell grid, cooperation holds in 68% of scarce cases After it add

    listing the five axes, 3^5 = 243 cells, 63 scarce and 180 abundant, the sustainability rule, δ = 0.95, and the plain caveat that these are test values and another grid gives other percentages. 4.9 References. Anchor: L303

    References, then
      at L304 through
    at L317. Replace the single list with a
    holding

    Cited on this page

    and the six cited, each with where it is cited and its link;

    Further reading (background, not cited in the text above)

    and the other six, each with its link. A comment records that every reference is kept. Twelve
  • before, twelve after. 4.10 Result box glue. Anchor: the end of the page, L400 then L401 . Add a second inline - + @@ -128,20 +164,27 @@
  • When there is not enough to go around, fighting can beat sharing. In our test grid, cooperation held in 68% of scarce cases and conflict won in 32%. We say that plainly.
  • Still unsolved: anonymous, one-time meetings with nothing at stake. This page shows a strong tendency under stated conditions. It is not a law of nature.
    • +

    The settings behind these numbers (LAYER v-love-logic-v3-phone-heal-v0). All three use the same model from §4: each contact's reality check catches the lie with chance q0 = 0.002, each comparison of shared records catches it with chance q = 0.01, lying gains G = 1 per round, the penalty when caught is Lp = 2, the reputation loss afterwards is Δ = 0.5 per round, and there is no discounting and no complexity cost. About 702 (701.79) is n = 2 contacts with m = 0 shared memory. About 10.5 (10.48) is n = 5 with m = 1 (full shared memory). Never ahead is n = 10 with m = 1: the catch chance is 0.6033 per round, so even the first round loses (G ≤ h·Lp). 68% / 32% comes from the 243-cell scarcity grid in §6, judged at patience δ = 0.95.

    You can check the math yourself: press the button below. Your device runs the same 656 checks as our Python simulation. Nothing is sent anywhere.

    Seed 20260924 · runs on this device only +

    +
    +
    +
    +
    +

    What the 656 checks mean: they confirm that the formulas on this page agree with their own brute-force sums and simulations. They do not show that the world behaves like the model.

    1. What changed from v2 (honest corrections first)

    @@ -215,6 +258,8 @@