The thing that jumps out at me most with this iteration of the Rogue is that it doesn't cater to a wide enough array of tastes. The Rogue has traditionally occupied a wide spectrum, with near-uselessness in combat (but the only guy that got skills!) on one end, and an extremely deadly skirmisher (at the top tier of damage output) under a combat-parity paradigm at the other. I had to check all the way back to the August 2012 packet for the last version of the Rogue that I remembered. As unpolished as it was, the Thief vs Thug dynamic is what I'd assumed from the Rogue in Next. The out-of-combat utility monkey AND the dirty tricks version of the Fighter, along with everything in between.
The schemes still have the potential to bring back that spectrum into one class in the same game, but not if they're written as rigidly as they are in this packet. Which is a tough line to walk, because this rigid structure is more foolproof in terms of game balance. Every scheme gets two free skills, and then three bonus feats. Unfortunately, like my problem with Fighters and maneuvers, the mess of a feats system in this current packet invariably affects the Rogue. Many of these bonus feats are just more skills with a new name, and from a different resource pool. It's unnecessary.
Even more mind-boggling is the fact that skirmisher types have feat support in Next, but the Rogue doesn't get much of it for free. Tumbling Movement is a start (2 schemes get that), but no Spring Attack? I suppose one could make the argument that Spring Attack is a more powerful choice than the feats in the Expert category, but if that's the case then it just shows that the feat balance game has already failed. Choices taken from the same pool of resources should be roughly comparable. Otherwise you end up with the dreaded trap feats. Those usually become more apparent after feat bloat has set in, but it's already a problem even with a relatively short list of feats in this playtest!
Ultimately, this Rogue fails at providing me with what I want from a combat Rogue out of the box - a highly mobile skirmisher that can reliably pull off hit-and-run tactics. Yes the Rogue needs this functionality if they plan on going into melee. Otherwise their leather-armor-wearing, low HP butts are going to be toast.
Finally, every scheme gives you a way to generate Advantage for yourself in combat. There's Backstab, which gives it to you when you gang up on an enemy (when an ally is adjacent to the target), Isolated Strike does the opposite (no creatures hostile to the target adjacent), and finally Tumbling Strike (easily the worst of the bunch) gives you Advantage if you started your turn 20+ feet from the target.
Backstab is the best option for a skirmisher playstyle (the enemy is deterred from chasing you down if your friend is threatening them with an opportunity attack), but the two schemes that best represent that playstyle mechanically (i.e. those with Tumbling Movement) don't get it. Awesome, Spring Attack has just become a feat tax if that's how you want to play. Isolated Strike will be the go-to option for your sniper Rogue; melee Rogues that go this route would do well to focus on enemies weaker than themselves. Mopping up the riffraff that everyone else is content to save for later, after the real threats have been eliminated. And Tumbling Strike just forces you into NOT focus firing. And opens you up to the most opportunity attacks, to boot (sure, they're made with Disadvantage, but if you're provoking them every single turn they'll catch up to you pretty quickly). It could be made much more viable by reducing the movement requirement down to 10 feet. That way at least a Spring Attack Rogue stands a chance of avoiding OAs and possibly even attacking the same guy two turns in a row.
This just might be the most unintuitive, convoluted version of sneak attack I've ever read. If you don't have Disadvantage already, you can apply your sneak attack damage by making the attack with Disadvantage. Oooookay. From a game balance perspective it actually works out somewhat well. You'll only use Sneak Attack if you manage to snag Advantage, which will cancel out the Disadvantage. So by using Backstab, Isolated Strike, and Tumbling Strike you're really applying your Sneak Attack damage (in a really roundabout way) as a normal attack, but the door's still open for using Sneak Attack if you can get Advantage in some other way.
Cool, easy to attain Sneak Attack damage, yeah? Yeah!? Sure, but the price you pay is that you're still pretty bad at combat. If an ally has some way of giving you Advantage then great, you don't have to jump through your hoops to get Sneak Attack damage, but you're still less accurate than pretty much everyone else because they'll actually be USING that Advantage. And since Advantage doesn't stack, even if you have it from multiple sources your Sneak Attack dice will always cancel it out. In other words, YOU CAN NEVER USE ADVANTAGE AS A ROGUE. Well, at least not if you're "spending" it to trigger sneak attack. Doesn't quite fit the image of a guy who doesn't fight fair.
And no, dealing a little more damage than the Fighter (maybe; if he's got a big weapon you might be about even with him) isn't worth giving up that accuracy. Because the tradeoff SHOULD be that you're dealing more damage because you're much, much squishier. If I'm playing a squishy character I better be getting something awfully cool in return. Some people might be satisfied with "more skills!" (and admittedly I'll occasionally play such a character), but more often than not what they want is more offense. The assassin archetype of the guy who can damn near one-hit kill guys he takes by surprise.
Oh come on! Giving up your next action AND a reaction to halve the damage of an enemy's attack? This COULD have been that Feature-That-Compensates-For-Your-Squishiness, but alas, it's just a Bad Idea. Especially if you're trying to fill that striker role your action should be among the best in the party for killing things faster, and killing things faster prevents those things from getting a turn in the first place.
You're basically telling the monster that for the next turn, you might as well be a common housecat (unless they're still capable of one-shotting first level Wizards; then maybe the housecat is more dangerous). You won't be attacking, and you can't even make an opportunity attack if they move away.
Now it does have its uses, situational though they may be. A really big opponent trying to land a solid hit on your Wizard who wants to use any flavor of the Make Him Go Away spell is probably worth a use of Distract. You weren't going to kill him anyways, but if you can keep the Wizard in the fight long enough for him to take the enemy out of the fight then the tide can be turned. Assuming you happen to be standing next to the Wizard and in melee with a big, dangerous opponent, that is. Situational is right.
More likely you're just delaying the inevitable, with the number far from being in your favor. Lets look at two rounds, assuming you're fighting an opponent of roughly equal strength (a duel, perhaps), where you both have the same average damage per round (DPR). DPR is a calculation that accounts for average damage on a miss, hit, and crit. It's not something that you'll ever see in-play; rather it's a value that stands in for all of the probabilities involved with attacking. The following example can certainly be tipped in the Rogue's favor if his attack crits but the Enemy's second attack misses, but over the course of a large sample size those variances get smoothed out into something that looks a lot more like the DPR.
Round 1: Rogue attacks, deals DPR.
Round 1: Enemy attacks, Rogue used Distract, Enemy deals half DPR.
Round 2: Rogue sits there, because Distract takes away his action.
Round 2: Enemy attacks, deals DPR.
Totals: Rogue has dealt damage equal to DPR, Enemy has dealt damage equal to 1.5(DPR)
It gets even less favorable when you consider the limitation in tactical options. If Rogue wants to run away he'll risk taking even more damage from an opportunity attack (even if he has Tumbling Movement there's still a chance that it might hit him). But if Enemy wants to run away, he can do so freely.
This does NOT compensate for the Rogue's squishiness. This is something that you'll use when an Enemy crits you, and you hope that the dice gods impart less luck onto him next round because taking that risk is better than being unconscious on the ground. But you'll probably end up that way anyways.