User blog comment:Flonnaru/Results of Non-Drum Expeditions sparkle effect/@comment-24430017-20150609005754

I have been trying expedition 5 for about 3 weeks with 5 sparkled ships. I have failed GS at expedition 5 with 5 sparkled ships twice today already. It definitely isn't 100% GS rate. 95% seems possible. I usually run around 10 times a day and having it fail once every 2 days seems like a reasonable estimate. Saying 20% GS rate per sparkled ship definitely doesn't make sense (at least to me), and we will just have a flood of users saying that their 5 sparkled ships failed and the wiki rate is wrong.

Calculating the statistics I get 79.7% +- 3.0% at a 95% CI.

I am not sure if i can agree with rationale for eliminating outliers. I was under the assumption that we are assuming that everyone should have equal chances and that there are no other factors besides number of sparkled ships should influence the results (like ship/flagship/HQ levels etc). Thus the justification for eliminating data would be that the people that submitted their data either did it incorrectly or that they are submitting false data? And over half the people submitting results did it incorrectly? That makes me question how you can be sure others actually performed the experiment correctly.

I am not sure I understand exactly what Ebara is doing on expeditions, but I think it would be worthwhile to look into other numbers of sparkled ships if we want to try to get a better understanding of sparkle effects on GS rates.