User blog:Homuhomu123/Epilogue for Exp't F ~ K

During the last 5 exp'ts, even larger sample sizes were used (170 for p > 80%, 256 for p < 70%).

For DA exp'ts, factors such as HP state were not very firmly controlled. Approximately 3 out 5 sorties were done in 小破 (51% ~ 75% HP), and the other 2 were beyond minor damage. If HP state modifier does exist, this combination might slightly lower the actual DA chances measured.

For CI exp'ts, HP states of the test ships were firmly controlled (always beyond minor damage).

For both experiments, 2 factors were not controlled and rolled randomly:

According to limited time and labor, such factors may not be possibly evaluated until more ppl agree to participate in this project and share their exp't data. Before that I will assume negligible effects from these two and they could zero out each other as sample size gets larger (> 190).
 * Engagement form (+//, -//, gT, rT)
 * Enemy formation (LA, DL)

Hypothesis:

According to the 10 exp'ts, here are my hypothesis regarding trigger rates of artillery spotting:
 * Flagship modifier is actually +15% rather than +10%, as cited on most websites. (fairly confident)
 * HP state will affect DA chance. Minor damage (小破) has an effect of -5%, and moderate damage (中破) has an effect of -10%. (To be tested)
 * Between FLoS = 300 ~ 400, there is no bonus on DA chance from extra FLoS. In another word, at FLoS = 350, standard DA chance is 50%, not 54%. When FLoS goes beyond 400, the LoS calibration formula is valid again. (eg. FLoS = 450, standard DA chance is 62%.).
 * Basic DA chance for CAs is ~50%, not 47%.
 * Plane counting on the slots does affect DA chance.
 * Level does not affect DA or CI chance of a ship. (To be tested)