Experiments 9 through 17

Experiment 9:
 In this first, crude strategy, mutant doubles, beavers and raccoons at GWC > 50% or takes
at GWC > 0% but drops at GWC <= 0%

Experiment 10:
 This is a variant of the above Experiment 9. Since winning the opening roll gives players
a GWC > 50%, mutant also doubles if it wins the opening roll. Everything else is the same.

Experiment 11:
 In this strategy based on 5 game stages: "beginning, early, middle, late, ending", mutant
doubles at GWC's > 50%, 55%, 60%, 65%, 70% and takes at GWC's > 0%, 5%, 10%, 15%, 20%

Experiment 12:
 This is a slightly more conservative variant of Experiment 11, with the mutant doubling at
GWC's > 55%, 60%, 65%, 70%, 75% and taking at GWC's > 5%, 10%, 15%, 20%, 25%

Experiment 13:
 Mutant randomly doubles, beavers, raccoons and randomly takes but never drops. This
was started out of plain curiosity but not completed because it would require a huge number of trials.

Experiment 14:
 These are four related, improved variants of Experiment 11, with different double and take
points, as well as newly added beaver and too good points, as explained in the experiment abstracts.

Experiment 15:
 This bot-vs-bot test was done to see the cube skill difference between 2-ply and 3-ply, in
order to check whether mutant's performance difference against 2-ply and 3-ply was the same or similar.

Experiment 16:
 This is similar to Experiment 14, with further simplified sets of double, take and beaver
points, all to be within 20% of each other and each then incremented by 5% for the next 4 game stages.

Experiment 17:
 This is essentially the same as Experiment 16, with all double, take and beaver points
increased by 5%, to see if more conservative cube play would do better. Results were very interesting.