Eh, sounds complicated given how streamlined next seems to want to be. I'd rather not add anything anyone would need to track for this.
I really think disengaging from one enemy to go after another should be punished by the rules in exactly the same manner as attempting to fire a range weapon while engaged. Just slap disadvantage on it and you're good. Nothing to track, nothing to remember, don't need to slow down combat with turn interrupting opportunity attacks, if you use your movement to leave a creature's threatened space, then you have disadvantage on any attacks you make with that action.
Maybe other effects could be added as special abilities for tank-oriented classes or defender-oriented themes, but as a core default mechanic that seems sufficient to me.
I would like to see an action similar to the Dodge action in the "How to play" doc, where you give up your move and action for the round but instead of an AC bonus you would essentially gain the benefits of the 3.5-style Combat Reflexes feat. Anyone would be able to do this and it is basically readying an action to "attack anyone who tries to get past me".
Possibly feats or a theme would be available that would expand on this to allow movement with this action, and other features that typical Opportunity Attack specialsts would want.
My biggest problem, with relation to DDN's design goals is that the purpose of removing OA is to speed combat up. There were two major issues with regards to combat speed with OA in 4E.
1) Interuptions to the action - Everytime you leave a threatened square, you trigger an action. Each of these requires a die roll, possibly two. If there's a hit, now there's time spent recording damage done. In a crowded room, there could be several OAs. From this alone, where a move might only take a few seconds to accomplish, this single move action, with OAs could now take a minute or two.
2) Time spent deciding how to move without triggering OA - Maybe your experiences were different, but in my groups, my players spent a lot of time carefully considering their moves. Without OA, if you know where you want to go, there should be no time thinking about how to move. With OA, it feels like players are trying to act out Family Circus Cartoons just to get from point A to point B, and it takes time to plot those paths.
Now, assuming OA adds a measly 15 second, on average, to the length of an average players turn, when you do the math, this adds up very quickly. 5 players, means you add 1.25 minutes per round. Let's say your average combat (ours was right around here) lasts 8 rounds. You've now added 10 minutes to this combat encounter, and that's only considering the players. With monsters/NPCs this would be even worst.
I understand my math is strictly conjecture, but I think it shows the point that just a few seconds of added time per turn can really add up quickly. But, having said all of that, I think that there needs to be something preventing a monster from rushing past your front line to get to your softies in the back.
I think something like a disengage mechanic (which was suggested earlier could work. It might be something like, "If a character or monster are adjacent to an enemy, and they move out of combat range (or out of a threatened area, or whatever) to engage another target, they get disadvantage on their first attack against the new target." This would help prevent monsters/players from disengaging the front line and rushing past to engage the softies in back but not completely negating the tactic. There would also be no interuptions to the movenment, and you wouldn't have to plot a course to prevent those interuptions. The only decision that would be required would be to decide whether the disadvantage is worth changing targets. I'm sure there's something mechanically wrong with this suggestion, but I just want to suggest something that alleviates as many problems with OA and no-OA as possible.
Eh, sounds complicated given how streamlined next seems to want to be. I'd rather not add anything anyone would need to track for this.
I really think disengaging from one enemy to go after another should be punished by the rules in exactly the same manner as attempting to fire a range weapon while engaged. Just slap disadvantage on it and you're good. Nothing to track, nothing to remember, don't need to slow down combat with turn interrupting opportunity attacks, if you use your movement to leave a creature's threatened space, then you have disadvantage on any attacks you make with that action.
Maybe other effects could be added as special abilities for tank-oriented classes or defender-oriented themes, but as a core default mechanic that seems sufficient to me.
I would agree with this as a simple and elegant solution.
I also think that moving slightly away from the "square-by-square" tactical combat as a base and into "declare an action and follow it" (i.e. "I approach the Wizard by passing around the Fighter's left side, then attack"), you can have an effective means of blocking.
Let's say the goblin tries this, moving past a fighter who has declared that he'll be shield-blocking (knocking them back/over as possible) anybody that tries to get by.
As the goblin moves past, the Fighter interrupts his turn to initiate a contest (the shield-block). He wins, throwing the goblin aside and sending him sprawling (possibly dealing damage, etc.), and the goblin's remaining turn (if it has any) is spent recovering from that.
Some very interesting ideas; having read the play-test docs, I would say a lot of what was suggested will end up being considered too complicated and involved. Here's an idea for D&D Next Attack of Opportunity/OA (to maintain natural language, call it a Parting Shot):
As ability scores have a greater role in D&D Next than in any other past interation of the game, involve them to adjudicate this. If a foe leaves a square that you threaten, consult your respective Dex Scores. If you exceed their score, you get a parting Shot. If not, no dice. And you can only get as many Parting Shots per turn as your Dex bonus allows.... So say maybe 3 for a Dextrous Rogue, maybe 1 or 2 for other characters maximum. This can be further simplified as a standard amount of damage, oh, say 5 Hit Points plus Strength bonus, so only a To Hit roll is required. As we all have those d20's close at hand, it would be quick, simple and in-line with the flavor of the rules. Feats could later expand upon this, either allowing more swings, a +3 (considered) Dex bonus to qualify for a Parting Shots, a Feat to gain more Parting Shots, etc.
This simple mechanic would cut down on willy-nilly running through a party of orcs or having them do the same, at least somewhat. Deciding when to take Parting Shots could become part of your strategy...
Some very interesting ideas; having read the play-test docs, I would say a lot of what was suggested will end up being considered too complicated and involved. Here's an idea for D&D Next Attack of Opportunity/OA (to maintain natural language, call it a Parting Shot):
As ability scores have a greater role in D&D Next than in any other past interation of the game, involve them to adjudicate this. If a foe leaves a square that you threaten, consult your respective Dex Scores. If you exceed their score, you get a parting Shot. If not, no dice. And you can only get as many Parting Shots per turn as your Dex bonus allows.... So say maybe 3 for a Dextrous Rogue, maybe 1 or 2 for other characters maximum. This can be further simplified as a standard amount of damage, oh, say 5 Hit Points plus Strength bonus, so only a To Hit roll is required. As we all have those d20's close at hand, it would be quick, simple and in-line with the flavor of the rules. Feats could later expand upon this, either allowing more swings, a +3 (considered) Dex bonus to qualify for a Parting Shots, a Feat to gain more Parting Shots, etc.
This simple mechanic would cut down on willy-nilly running through a party of orcs or having them do the same, at least somewhat. Deciding when to take Parting Shots could become part of your strategy...
Seems it would bog things down to compare dex scores with every movement that provokes with all the complication of having to figure out if a movement would provoke the comparing of dex scores or not. Doesn't seem as abuseable as 3.5 AoOs though.
Playing with the current combat rules doesn't seem that odd except for the situation where someone wants to defend an area or block people from passing them. A "Defensive Fighting" action, where you basically ready an action to attack those who try to get around you would use the same familiar AoO rules but not be in play most of the time.
Perhaps Fighters, who should be the ones defending, automatically get a Zone of Control (ZoC) mechanic as part of their class. As long as the Fighter can take Immediate Actions (so if Dazed or Stunned he would lose this, though feats or themes might change even this, or higher level improvements), anytime a foe moves into any square in his ZoC (defined as adjacent squares) they must stop, end of movement. This means if you are already in the ZoC you can move 1 square, that is it. If you leave the ZoC you are unhindered (back away from the fighter), but anytime you move into a ZoC you stop. You can use another action to move again but still stop if you move into a ZoC. Simple and gives fighters good battlefield control.
Other classes could do this also, but for them, it is not always on. They must spend a Move Action to do this. So a new Move Action now exists for all classes, lets call it Defensive Stance, which lets you Move 1 square only and then you gain the above ZoC until you move away from that spot. Thus you have to decide to move full speed and not control any area, or move only 1 and set up your ZoC.
Perhaps the fighter could choose this option also, but for him, if he wields a reach weapon, his ZoC now extends to 2 squares around him. He does not get this extended ZoC always, only if he uses the Defensive Stance option, other classes cannot get this extra reach option.
Some themes or classes may then have other benefits that trigger from this. So a rogue might get to apply Sneak Attack damage to any foe who enters his ZoC if he sets it up assuming he took the feat or theme. A fighter might be able to Prone (trip) the creature not merely stop it if they have the right feat or theme, perhaps it is auto or maybe it takes a skill check.
Maybe creatures larger or smaller than you are harder to control in this way. Each size difference means they can ignore that many of your ZoC squares before being subject to it (and thus stop). So an ogre does not stop when it enters the first square of the ZoC, but if it enters another ZoC square (any, not just the same PC) it stops then, only able to ignore 1. This is an extra complication but not too much to slow things down. It lets feats or themes address this, allowing a fighter to act as if 1 size larger or smaller for this purpose, as a possible improvement.
I vote for: Threatened squares count as difficult terrain for enemies. (This means it would cost 30' of movement to sneak past a fighter guarding a 15' wide corridor.
My second vote: If you move into a square threatened by an enemy your movement for that turn ends. (This means that it would take three turns to sneak past a fighter guarding a 15' wide corridor.
Both of these rules only apply if your enemy is aware of you.
In my opinion they both also accomplish the intended goal, allowing enemy movement control by frontline fighters. And since this control is passive on the player's part these options do nothing to slow down speed of play. In some instances they might actually speed it up.
Shane
Help make Combat Mastery happen: If you like the idea of Combat Mastery, as outlined below, for fighters copy it onto your signature and add interesting combat maneuvers to the list. Two new examples could be throat punch or spit in eye.
Combat Mastery: When a Fighter performs combat maneuvers such as bull rush, disarm, sunder, trip, hip toss, eye poke, ball kick, hair drag, blind with sand, slide down banister, swing on chandalier, walk on barrel, use enemy as shield, interpose self in front of arrow trying to kill wizard, intimidate, pick up kobold by the neck, etc, the minimum die result is 10.
Fighter Combat Maneuvers: On a given round the fighter can bull rush, disarm, sunder, trip, hip toss, eye poke, ball kick, hair drag, blind with sand, slide down banister, swing on chandalier, walk on barrel, use enemy as shield, interpose self in front of arrow trying to kill wizard, intimidate, etc, in place of his/her move action. This is a nonattack action that might cause the fighter's opponent to be rendered prone, unarmed, blind for a round, etc, or otherwise grant the fighter advantage or his/her opponent disadvantage as the Fighter sees fit.
Re Proposed new Rule for AO/AoO: If a foe leaves a square that you threaten, consult your respective Dex Scores. If you (EDIT: Meet or) exceed their score, you get a parting Shot. If not, no dice. And you can only get as many Parting Shots per turn as your Dex bonus allows.
Seems it would bog things down to compare dex scores with every movement that provokes with all the complication of having to figure out if a movement would provoke the comparing of dex scores or not. Doesn't seem as abuseable as 3.5 AoOs though.
Strongly disagree & not just because it's my idea . Actually it wouldn't bog down combat whatsoever, in fact, it would speed it up (provided AoO/OA end up in the final rules) because it only requires a glance at a sheet or references a number the player will probably know by heart. A DM will just need to know or have on hand the Dex score of the bad guys in combat in case it comes up. Once known, most players and DMs wouldn't have to look again that encounter. Sometimes comparing Dex scores would bog down combat NO MORE THAN comparing Attack Roll to Armor Class every single attack, right?
. Actually it wouldn't bog down combat whatsoever, in fact, it would speed it up (provided AoO/OA end up in the final rules) because it only requires a glance at a sheet or references a number the player will probably know by heart. A DM will just need to know or have on hand the Dex score of the bad guys in combat in case it comes up. Once known, most players and DMs wouldn't have to look again that encounter. Sometimes comparing Dex scores would bog down combat 
