The handbook is unclear on how to handle when a Pokemon has an evasion increase due to Sand Veil and is being targeted by a Pokemon that has an accuracy decrease using a move with some number accuracy. Because Sand Veil isn't a +1 stage boost, but rather a straight percent number unlike other stat boosts, the accuracy drop and evasion decrease can't be added together on the table for accuracy and evasion. The handbook also doesn't specify where to add the evasion boost/accuracy decrease, thereby leading to several different ways to calculate the move's chance to hit. i.e.
1.apply the accuracy decrease first and then the flat +20%, so with a Gible in Sandstorm and no Mold Breaker in effect would lead a move with 100% accuracy like Tackle used by a Pokemon with a -3 accuracy stage boost to have a hit rate of 30%.
2.However, hit rate can also be calculated by applying the evasion increase first making Tackle have a hit rate of 80%, and then applying the accuracy drop of 50% would lead to Tackle having a hit rate of 40%.
3.Or you could multiply the two percents together to get a hit rate of 40%.
4.Another way would be to multiply the hit rate after applying the accuracy drop of 50% with the evasion rate, this case 20% of .5 being .1, and subtract that from .5 would lead to a hit rate of 40%.
Here are the formulas for each scenario I've just listed.
Where ae = the coefficient of the accuracy as determined by adding the flat stage boosts of accuracy and evasion that the Pokemon impacted by the move being have as decimal, E representing a flat percent change in evasion as a decimal(e.g. Sand Veil's 20% evasion boost), H being the hit rate of the move after calculating evasion and accuracy drops, and C representing the base accuracy of the move being used.
1. H = (C * ae) + (C * E) when ae ≥ 1 or (C * E) - (C * ae) when ae ≤ 1
2. H = (C * E) * (C * ae)
3. H = |(E - C)| * (C * ae)
4. H = (C * ae) - ((C * ae) * E) when ae ≤ 1 (I wasn't sure what the correct hit rate would be if ae ≥ 1)
I suggest further clarification on how to calculate hit rate in such scenarios.
Here's the battle which led to such a discovery, which you can analyze if you wish.
https://www.smogon.com/forums/threads/rextreff-vs-montegom-3v3-1-singles.3695217/page-2#post-9074380