Quote Originally Posted by Telok View Post
There have been repeated issues with dev failures of understanding basic statistics and dice probabilities through the last 3 editions, especially with regards to the outcomes of things with multiple rolls. Since mid-3e it's been my habit to check the math for anything they do that involves more than one roll.

That's actually semi-off thread, but there have been enough face-palms over the years that I don't assume the devs really know basic math any more.
Tabletop gaming developers are generally writers by training, not statisticians, mathematicians, or people in any other field with a high degree of numeric literacy. Heck, Gary Gygax's training as an insurance underwriter probably puts him well ahead of average in terms of mathematics training among TTRPG devs. Also, TTRPG design teams are small. These aren't big corporate projects overseen by huge groups of people with multiple departments. Whole game systems are put together by 6-8 people on a regular basis, and even large gaming books can be composed entirely by one person and only edited by one other person. This is definitely true in 5e, which has a tiny design group (3e's was larger, probably the biggest its ever been for any TTRPG).

Consequently math errors and continuity errors with regard to fluff are extremely common, doubly so among books that are produced by the method of writers contributing largely independent sections that are then cobbled together by a single design tasked with oversight. To get a continuity error like the Wall of the Faithless in 5e all you need is one person to misremember something, and one other person to not question the factoid and bother to double check. This is especially likely since the actual writers may not be especially familiar with the setting, certainly not on the same level as many of the biggest fans.