Thought I run more tests regarding Elements (via Flashpoint). I was curious about the fat old guy's - a.k.a. Oracle - rolling chances.
My method consisted of rolling the Oracle repeatedly and editing the *_lastoracle.txt file in between rolls to make multiple rolls per day possible.
Initially did the testing manually, then started using a mouse clicker/keyboard press utility.
Things went quite ok until 150 rolls - though I noticed some patterns, some cards got rolled a bit more often than others, but I didn't really care,
thought it is only about obtaining a big enough sample. After 150 rolls something happened: certain cards started to build up
a godawful lead against the rest. Really wanted to figure out what is causing the anomaly, so I started to change some variables.
One might point out that using a script might have affected the outcomes, but that's not true, I started using that only after 250 rolls.
Things I changed:
- the date in *_lastoracle.txt;
- randomized the pixel where the script clicked;
- created new logins.
None of these helped. Always the same set of cards appeared, again and again. I wonder that the culprit might be a setting, some
session file, or who-knows-what in Flashpoint. Maybe someone else has an idea about what is causing this?