Where PULSE lost

Complexity can cost more than it contributes

In the weak-institutions world, PULSE ranks fourth under all six scoring systems. The simpler liberal constitutional design ranks first. Only 2.225% of PULSE runs meet the model’s combined free-and-integrated outcome. These are conditional model outputs, not predictions.

• Inspect world 19
• Inspect assigned effectiveness

A winning rank can conceal failure of the goal

In the deeply separated world, PULSE ranks first but reaches the model’s free-and-integrated outcome in only 0.9125% of runs. Being better than the alternatives in that world does not make the outcome good.

The liberal comparator slightly leads on avoiding the model’s catastrophic failures

Across the baseline worlds, 62.710625% of PULSE runs and 63.091250% of liberal-comparator runs avoid all five failures counted by this measure: collapse, unrecovered species loss, domination, authoritarian takeover and failure to preserve individuality. Avoiding those five does not mean no disaster happened.

The assumed benefits can determine the conclusion

The average across six scoring systems falls below the fixed liberal comparator between PULSE benefit settings of 0.85 and 0.80. This supports a rough 15–20% discount warning for the aggregate. It does not establish a precise real-world performance threshold or the same threshold for every scoring system. Here the percentage describes moving selected institutional-effect coefficients toward the model’s neutral setting, not reducing all assumptions. It is not a measured advantage, safety margin or empirical confidence interval.

What should follow from these failures?

Test simpler transitions where institutions are weak. Challenge whether the promised cooperation and safeguards actually work. Compare costs and alternatives under less favorable assumptions. These findings do not, by themselves, require a change to the five doctrines.

• Evidence that would require revision
• Unresolved implementation questions
• Download code and results