First of all, I have to say, all this while, I thought SABR was an acronym for Super Aggressive Baseline Return. Sneak Attack By Roger sounds so much fun.
Back to the question, this is a very good blog on the effectiveness of SABR. Some excerpts follow
servers win about 64% of points and hold about 80% of service games.
For most players, breaks of serve rely more on the server’s
occasional lapses. To put it in numerical terms: A passive returner is
playing the lottery in every return game–a lottery with only a 10% to
20% chance of winning.
The best way to earn more breaks of serve, of course, is to win more
The alternative is to change the rules of the lottery. Instead of
accepting a steady rate of 35% of return points, a hyper-aggressive
strategy is more likely to make the point-by-point results more
streaky, even if the overall rate doesn’t change.
That basically means, even though, a player using SABR tactic occasionally, will win an overall same number of return points, but he is more likely to break an opponent's service game than with a passive return, because of the increase in the unpredictability factor.
A player who wins 35% of return points will, on average, break in 17%
of his return games.
If that same player wins 30% of return points in half of his games and
40% of return points in the other half, he’ll break serve 18% of the
Double the variation and say that the returner wins 25% of return
points half the time and 45% the other half. Now he’ll break serve in
21% of games, or one extra break per 25 return games.
The real magic happens when we expand the variation to an even split
between 20% of return points and 50% of return points. In that
scenario–when, remember, our returner is still winning 35% of
points–the break rate improves to 26%, almost one more break per ten
return games. On average, that’s an extra break per best-of-three
match, and closer to two extra breaks in a typical best-of-five match.
Many tennis writers have written about how SABR is more fruitful a tactic to unsettle an opponent's rhythm rather than to win return points. Take the case of Federer's win over Stan Wawarinka in the semifinals of US Open 2015. Stan was looking to repeat his comprehensive win over Federer at French Open semis 2015, connecting the ball well & making winners in the first 3 games of the set. Federer SABR-ed him 2 times in the 3rd game & took rest of the serves aggressively, inducing a double fault & subsequently a break. Stan was rattled & couldn't recover that touch on his strokes & Federer just ran away with it.
He did the same thing to Djokovic in Cincinati finals, unsettling his serve & putting pressure on the returns & winning 39% of return points
Coming to the shot itself, SABR is tried on an opponents 2nd serve & that too against average 2nd servers. Federer didn't try it against Isner in his QF match much, I think just once since Isner like Karlovic are heavy servers. Most of the rest of the players have average 2nd serves.
Federer moves in just when the opponent tosses the ball in the air so that the server cant change the direction in which he will serve & also to create a disturbance in the peripheral vision of the server to increase the chance of a double fault. This has happened on many occasions against Stan, Kolchreiber(spellcheck), even Djokovic in today's US Open finals.
Which players can try it? Well, those that are excellent movers on the court. Someone that comes to mind ahead of even Federer is Justine Henin who was much better mover on court but she never tried it. Other active players who could try it are Radwanska(excellent hand-eye coordination & silken touch). Nobody else comes to mind because most players on both ATP & WTA play from the baseline & are not good movers towards & away from the net while they can chase balls from one baseline corners to another without breaking a sweat. The perpendicular movement is crucial because the one SABR-ing has less time to cover an open court if the SABR return is short & within reach of the server
EDIT Here is an excellent video compilation of SABR with slo-mo reply