A printed rate needs a target
First ask what the number describes: one named regular style, any secret, or one particular secret. These are different events. A combined secret rate cannot be reused as the chance of one secret when several exist. Record the exact series and packaging source beside the rate.
A full case is a different purchase
A stated non-repeating assortment links the boxes together. The HIRONO Mime listing describes such a whole set; SMISKI warns a secret may be absent from an assort box. Neither statement supports treating all case contents as independent. Follow the exact package terms.
Equal styles are an assumption
Our regular-style model allocates the non-secret probability equally. Tier labels such as those on MOLLY Series4 do not justify that assumption. Use a verified single-style percentage when regular styles differ, and leave an unknown value blank.
What a result means
With a hypothetical 5% chance per independent box, six boxes give 26.49% for at least one hit. The opposite outcome still has a 73.51% chance. Earlier misses do not change the next independent box. A 90% threshold remains uncertain and is not a spending recommendation.