EMPIRES & PUZZLES - Hero Utility

Hero LEADERBOARD Rankings: A Definition
Released with Data Extract #4 and Applied on 8/23/2018

Modification with Data Extract #12 and Applied on 10/15/18.
Also Applied Retro to all Historical Data.

RANKING CHANGES 10/15/18: Instead of leaving off where duplication ends for the next Rank# (i.e. 1,2,2,3,4), it will be pushed to the current counter (i.e. 1,2,2,4,5). This ensures there isn't an overstatement in rank. The top 5 have averaged about a 50% use increase above those below them based on data collected to date.

This method I have come to term: ROLLING 'FORWARD' THE RANK.

An actual Rank (Rnk) column has been added to report 1 and 8. Using report 1 below, just count from 1 to 15 starting from 1 and it should make complete sense if it doesn't already!

The Ranking Examples below are not being updated to reflect the ROLLING 'FORWARD' THE RANK. Just apply what has been discussed here if you review them.

The method used below I have come to term: ROLLING 'WITH' THE RANK.


Thanks - RaZoR



A process was developed to provide LEADERBOARD Rankings to each Hero who made at least 1 appearance in the data extraction (ok - code!).

It results in a larger scale of diversity in the Output as opposed to limiting it to the Top 5 and Top 10 (Current / Cumulative Data; respectively).

Note however, the further you move away from the original 5/10 rule, the more likely duplication occurs with the ranking number(s).

I'm including 2 views of the data side by side (Current on the Left and Cumulative on the Right in both images).

The 1st image providing the summarized data shows that the Overall ranking range is the most diverse as its combining all 3 positions of the Ranking values. You can click the image to enlarge it.



The 2nd image providing the summarized data shows the Center position has a smaller range of values simply because its a single hero role position. You can click the image to enlarge it.



Also see 'Grading Scale'