Friday, June 7, 2013

Box-Mapping 2!

I got some extra box-mapping data from the Pojo and Reddit threads, so I've added them to my data.

The new spreadsheet:

Does this new info change any of my predictions from the last Box-Mapping post? Let's see:
  • The total number of RR and higher still matches expectations
    • Total Rares Theory: Still good.
  • The bands of RRs is still mostly visible, but box 5 is shifted pretty far. 
    • RR Bands Theory: Weakened, but still good.
  • The SRs aren't positioned like mine at all, so my prediction about them being next to RRs is invalid.
    • SR Location Theory: Dead.
  • The RRRs all seem to be 2 packs from an RRR, so my theory that they push the nearest RR could still be true.
    • RRR Location Theory: Still good.
  • The SP was on the opposite side of the box, so we can't tell much there.
    • SP Location Theory: (see next theory)
  • Only 2 of the dead zones (rows that only ever had Rs) still remain, BUT the SPs are both the only things keeping their row from being a dead zone. 
    • I'm going to go out on a limb and say that these dead zones (rows 1, 5, 16, and 20) can only have Rs or SPs.
    • Deadzone Theory: Revised, now the "SP/R Zone Theory"
  • The RRRs and SRs are much more equally distributed now.
    • Back Half of the Box Theory: Destroyed.
As before, if you have any new box mapping data, let me know so we can keep narrowing this down.

1 comment:

  1. Threw together a spreadsheet for you with 4 more boxes I opened today. I hope it helps!