I think I'm going to steal Yuttt at some point and put together a long post about what is required for stall in this meta. But for now, I'm just writing out observations after running and building two stall teams (both with hundreds of adjustments). I know smogon likes their articles, but right now we need the AED for stall in this meta.
I think one of the adjustments stall takes this year is to focus only on rocks because all others are a waste of time. To stop defog, you would technically need your prankster taunter out at all time. However, stealth rocks have more PP than defog so it is a much more versatile option, requiring less commitment, less focused pokemon and most importantly, less time to set. The old phaze stall is probably dead, but adapting gives us benefits.
Offense makes good defense, too! Mainly, stall is now allowed to look at attack stats with a little envy. Bulky offense has a good deal of good pokemon that, with some EV investment, can be transferred into stall pokes. Moves like Roost on hydregion, rarely used in OU previously, are now completely viable to abuse in stall. Stat investment into the somewhat average defenses allows him the bulk to play as a slightly more offensive stall poke. Latias, Slowbro, mega tyranitar, mega venusaur, heatran, gyarados, aegislash and other pokes with usable attack stats define the future of stall.
Without any investment into the attack side, these pokemon dish respectable damage and take damage very well.
For example, I run a custom spread venusaur-mega with 248/68/188 (all defensive stats) and relaxed nature. Running two coverage moves, I can expect to hit about 33% on neutral pokes anywhere and take much less damage. The ability to simply dish damage while tanking and phazing is so incredibly important this generation.
Scald over Willowisp! Willowisp is good, don't get me wrong, but status stall has little success this generation. I heard talking of trapping status stall, but the issue is heal bells. They're a bit more common this year with the accuracy of willowisp. While statuses are still viable and integral to stall, we can't focus on it, either. For example, Most teams carry willowisp on a pokemon or two, and toxic on another. I reduced the amount of willowisp and toxic by adding high percentage status moves like discharge zapdos and scald slowbro (also nice 100 SpA that doesn't need investment). The added chance of burn/paralysis alleviates move slots, deals damage and effortlessly gives me more damage on pokes I generally would just be status stalling. Also, moves like scald/discharge make switching more difficult on the opponent. They don't want key pokes to take these statuses and may make less-than ideal switches.
Physical walls float the boat. This meta is simply packed with physical threats. From mega lucario's deadly SD sets to Mega pinsir's similar style and the ever-present Khan, the meta simply refuses to go special. This is beneficial as much as it is harmful. Most of us have status moves to lock up this style in willowisp, but it means that "balanced stall" which look to create equal amounts of special walls and physical walls, get wrecked by one sided attacks. The most success I have had, and I have had a good deal of it, comes when I focus four pokes as dedicated physical walls and two blended walls. But these walls can handle special attacks, too. Mega venusaur, mega tyranitar, mega aggron even (when specially invested), aegislash, heatran, registeel, gyarados and a tirade of other pokemon can hang their hat on inherent special bulk (or in aggron's case, invested) and then run a perfectly fine wall to both sides simply because of the lack of OP special attacks right now. With one true special wall, generally a cleric (which is a SPECTACULAR idea, fyi, to have a SDef cleric as you can stall recovery without switching and mitigate issues with special walls), you can really take care of the full range of special attackers on the opponent's team (which is generally just two max).
Fairies stop everything, including OU wall breakers! Another adjustment is probably going to be the requirement of fairies. The wall breakers of OU are generally dragon/fight types, and fairy is a magical stop to wall breakers, although you generally have to specific which wall breakers you are going after. Togekiss can generally switch on garchomp-mega, but hates stone edge. But really, the best wall breaker counter is Clefable, hands down. Clefable cold stops garchomp (33% from EQ), Hydreigon (perfect counter), and haxorus (less from EQ) while also stopping conckledurr and some other fights. Where Clefable hurts is vs Terrakion who has that strong stone edge stab.
Florges (and even Floette)/Slyveon/Gardivoir-mega also stop Hydreigon cold, and can be a decent stop to any special wall breaker (namely latios, kyurem and even chandelure). The key resists to fighting and dragon make fairies incredibly useful to stop the monster attack all purpose wall breakers.
Reduce ground weaknesses! I haven't done a check, but my guess is earthquake was the most used move this month, and ground attacks were the most common. Perhaps flying from BB TF was up there and uturn/volt swtiches, but the mighty EQ is very seen. Though rocks are annoying, there are so many ways to handle them that just going rock neutral is fine. This means reducing steel typings, looking for levitate users and just taking a few good safe switches for ground in general. I try to keep mine with one immunity/resist for every weakness.
But there's a second reason for this. Ground's only physical competitive move is 100 base power and 100 accuracy, and the users vary from just about every physical dragon to most megas (including Khan, Mega venusaur, Tyranitar, Aggron, pinsir, charizard) and every offensive team has at least two. You never know which will have it, but it is really hard to be out trying to stop something and get blasted by an EQ. This is why mega pinsir is so difficult to stop, and the reason charizard X can't be stopped by general rock types, why mega venu can't be walled by steel types without fully knowing the set.
Dark weaknesses are also good to avoid. Pursuit, knock off, thief, foul play and some other moves are all great utility moves to hurt stall badly. Fairies of course can handle this, but watch out for those pursuit traps. Right now, knock off is a staple on most well-built bulky/tanky teams, so be aware that any dark-weak pokemon with an item will be taking massive damage when hit. Mainly, Conckledurr, Tyranitar, Scizor and mandibuzz are running these support moves. Dark moves don't always have good power, but they have excellent utility to the moves.
Define what a counter, check and wall is on and to your team. To me, I use the following definitions:
Counter: A pokemon that can come in and take little damage, less than 33% mainly, and win any fight should the opponent stay in. An example of this is what Slowbro is to Excadrill, or heatran to talonflame (esp when Heatran goes ancient power).
Check: A pokemon that can come in on any safe play and win a fight 1v1. It cannot take damage so easily, but can win after rocks damage and some other residual. preferably losing only 60-70% health max. This is what Slowbro is to mega-lucario at +2, or what talonflame is to a lot of different pokemon. Stall really should avoid this as much as possible.
Wall: A pokemon that can take no damage, or almost none and never die to another pokemon, but really can't do much back. This is Blissey to Gengar, special kyurem or any special threat. This is also skarmory/air balloon excadrill/heatran to gliscor.
Yuttt thinks of walls as overall counters (Blissey to special attacks, skarm to ground/physical). I mean, I see that as more of the idea of a tank, glass canon, bulky attacker category of pokemon, but it does work. My idea is simply a poke that prevents another from achieving any purpose but can't achieve much, either (at least in damage).
Some changes, however, exist in these definitions and that's mainly preference on what damage marks each tier. In checks, a lot of time a check will outspeed what it is checking for faster teams.
Lastly, and most importantly, spend time in team building.
If I was to tell you how I have learned stall or how I've gotten to be good with stall, I'd say it's because I build good stall teams. I invest more time building and adjusting sets and theorymonning than I do battling, and I make quick observations about the way pokes work in battle. The Dreadnought team I recently made has had revisions for now just under 30 days, and there was a good bit of teambuilding before that. Haunting Synergy, my semi-stall 5th gen stall team, also took upwards of 50 hours to put together. And I have side projects and stall teams that never take off, but I find concepts and cores that worked from them. I have a sigi burn team that I don't use anymore, but pieces and parts of it are currently in my newest creation that I'll hopefully have out by Tuesday.
But team building is way more than pick 6 bulky pokemon, or selecting a type core. It's finding utility, objectives, ideas and employing calculations to make sure the team runs well. As mentioned, I have four good cores I came up with before starting this most recent core. The dreadnoughts even just run a simple fire/water/grass core, and haunting synergy did the same. These aren't original, right? But when you employ objectives you want to reach:
Keep the team healthy at all times
Have _ amount of phazers
Check __ pokemon, counter ___.
Burn all offensive threats, stop heal bell users
Then you have basis and objective. Stall literally relies on the house you build, and the house you build must able to take on severe weather. One way of learning is to just read RMTs on stall (I did this to Meru's sand stall team, as well as a few early teams). Others is just through trial and error. Most stall teams don't run well out of the gate. I had a stall team that got outsped by base 80 speed taunters and could not win afterwards. Had to adjust and learn to counter that.
A good stall team has:
Core(s) that cover itself and expand a good amount of resistances (F/w/g, dark/psychic/steel, dragon/fairy/steel, dark/ghost/fairy)
Many, many resistances, even at the sacrifice of an extra weakness (as long as that weakness is covered somewhere). Remember, normal typing still takes neutral from a ton of attacks and a whole team getting hit for neutral would probably lose.
Objectives. What do you plan to achieve as your endgame. Phazing is no longer an option.
Utility: If it was all about walling, all stall teams would have dusclops, the world's greatest wall (short of ubers Giratina and lugia). Have knock off, wish, heal bell, willowisp, taunt, roar, stealth rocks, foul play, defog, rapid spin, haze or any move that gives you the ability to keep yourself in advantageous positions. Not all are needed, but you definitively need some of these.
Good Bulk/Recovery: This, of course, is a given. A stall team still needs to squeeze all of the above into bulky pokes and pokemon with good recovery.
Synergy: This is part of cores, but this goes deeper. For example, a core you would never expect, Latias/Lanturn/heatran existed on my HS team because lanturn heal belled heatran when he used rest, latias wish passed, and heatran tanked. The weaknesses were mostly covered in this core as well, but the combination of moves helped these three become effective. Lanturn also used discharge to slow pokemon down, heatran burned physical attacks, and latias could pass off statuses using psycho shift (alleviated switching and heal bell usage).
Testing: Unfortunately, even the most perfect-looking teams on paper have issues. Sometimes, it's just one or two very fixable flaws, other times you are rearranging pokemon left and right to steady the team out. Sigilphy, for example, has passable bulk and really great bulk after a few cosmic powers. However, it is walled by dark types. Sometimes over psycho shift, it is a good idea to run HP fighting to cover this. A simple fix, really.
Patience: Stall teams are the hardest teams to design, hence the belief that they are cheap. A great stall team probably had 100's of times more work put into it than a great HO team. The time is strenuous, too. That's just the nature of stall, you hang your hat on your ability to make good teams. You also have to scour the whole metagame, including lower tiers and even pre-evolutions (did you know that misdrevious, gligar, sligoo, bayleaf, magneton, roselia all have better defenses than their evolutions when coupled with eviolite? Most do, but these are just the better ones to look into). When you think about all the pokemon a stall team takes into consideration for what is viable, you start to see why it is such a long and hard process.
Lures: We're stall... we do have to have some underhanded tricks to get certain pokemon out of the way. Fire blast slowbro for ferrothorn? Yes please. Meru did the same thing with Latias and Roserade (I think there was some testing to note roserade had scizor switching into it a lot...) for scizor. EQ mega-venusaur for heatran? Happens. Just make sure it has more viability than the pokemon loses using that slot.
And that's it for now. Come on, let's get some more stall in OU pokebank, stall is fairly important to our meta. Keeps the world in harmony.