How did you come to this conclusion? Did you RE the game and pull out the numbers?
What's the determination criteria in the best practices? Pure experience and statistical analysis?
How does the game "spawn" elites, is it on map load? Or player proximity to a tile?. Why do I consistently find an elite pack remaining north of trist when the entire thing is done then? Wouldn't superunique and the other guaranteed spawns have taken that slot?
On Evil urns, should you just pop 'em or avoid them early in the clear of a zone till saturation to avoid eating a slot of a specific area?
I'm a proponent of straight line farming, but in pairs imo. A pair of ppl traversing A3 river banks to Durance 2 as well as Frigid onwards (downwards to Dac since it's narrow there anyway, then skip the rest imo) in A5 are great examples - way more coverage. Lots of similar places without A1 style paths where elites spawn along the route anyway. I consistently get anywhere from 4 to 8 more heralds myself clearing obvious missed spots or half arsed straight lining. Might not seem worth it but at least I don't waste time with T1-T4 and I'm almost surprised when 2 or more acts are cleared and I spawn a T3 herald, obviously it wasn't cleared or the players had a specific goal in mind such as density for xp or act boss for statue.
Initial numbers come from level data pulled from game files, but I wasn't satisfied with the data presented and tested every single zone over a period of ~500 hours.
Examples of discrepancies include zones like "
den of evil", being coded to have 2-4 elites in hell difficulty, and the
Monastery gate is set to 0. But obviously in the trialing those numbers were inaccurate. with Monastery gate rolling 0-2, and the den of evil being exclusive to corpsefire only (likely a quest linked area bug).
Fatty180 (reddit) was instrumental in our approach to data collection in the City of the damned and
Roneye has been awesome in providing insights to routes, and spawning logic.
Pavke was a huge help as well in understanding herald spawning logic and in our 9000+ elite pack data collection for herald spawn rates between games.
In order to determine the best routes we compared traditional routes, and full clears vs. an elite kill per minute focus, usually skipping the last elites in an area if an elite isn't observed within 2 screens.
D2R forward loads monsters and rolls them into elite packs usually within 2.2 - 3 full screens of the player character, if there are additional spots in the native cap then an elite will roll. If you see large packs with no elites present, you've likely exceeded the native cap for the area, and that large group is a "failed" elite pack. Superuniques and guaranteed champion packs will always spawn in the area on their respective tile when a player is within 2 screens of the tile, only then will they consume an elite number. Ideally we want to stick to saturation pathing (when possible) that allows the native cap to be reached before we target super uniques / guaranteed champions in order to maximize our returns.
Evil urns will consume an elite pack number, so its best to save them after you've cleared a bit of your zone, if nothing spawns its a sign that you've already spawned the maximum elites for the area
(Avoid them in the early clear)Monster shrines These are true bonus elites, they wont count towards the elite counter
The biggest issue I have in running in pairs is
I don't have any friends 
, group farming heralds and area clears is quite literally the best way to maximize your ekpm. The best group I'm aware of reaches 1 herald every ~1.15m, running split p7 farming with a horker
We have experienced some bugs where Tristram can spawn 7 total packs that include the area super uniques, though this was observed twice in about 300+ runs. Similarly we had a couple runs where specific TZ's spawned with no mobs at all (ruined temple, Tower Cellar, Tristram (single occasion)
I'd like to hear more about this north elite you are seeing, is it in the tristram town area? If not, what is your usual route / pathing