The only time I could see this being useful is if you really really need to Disarm or Smash Shield.
The point of this was not necessarily to give the acting player advantage. Remember that other players can enjoy advantage generated by different owners. This is a combo-setup ability for when it'd be REALLY nice to have another, perhaps more situationally effective fighter get their attack in.
I'd have to do the math on this (or run a simulation) but my hunch is that a +1 or maybe a +2 (30% chance at skill), assuming it succeeds at all, is not worth giving up an attack. Also, although the dictionary definition of Insight might seem applicable here, it doesn't really match the RAW description of the skill. Or Search, for that matter.
Thanks for the math. I was most concerned about this one. I wasn't sure if I should do 1, 2, 3 on all enemies in the fight or 2,3,4/Valour on a single enemy.
I do think the game could use more ways for teams to work together to take down those hard-to-kill adversaries with Great Size, and reducing their Protection is one way to do that. I don't quite get the logic of somebody in defensive stance persuading a troll or ogre to do that, nor how song would achieve it, but in general killing trolls is a good thing. The risk/reward has a high beta here, I think: a player with high skill and high heart burning the group's battle dice in the first round, followed by archers and spearmen making prepared/called shots, might risk trivializing trolls, while a player with normal skill and heart and no battle dice is probably going to waste their turn, and possibly even give a troll a free point of hate, which is a Really Bad Thing (unless its endurance is already zero) because it makes it harder to make them weary. I actually sort of like the near-impossibility of doing the math on this one.
This was an ability taken straight out of the Hobbit books wherein Bilbo taunts the Spiders of Mirkwood with song. We didn't know how to translate that in game any better than exposing some kind of vulnerability. This seemed like the most interesting, balanced, flavorful way to do it. While the context of the book would suggest an actual hit to Endurance (as the spiders get hopping mad and literally begin to froth with frustration), I was loath to add some kind of splash damage attack that face characters could use. While it wouldn't have been able to kill enemies, I saw the potential for a lot of face characters simply whittling down enemies from defensive stance. As for how much Heart it's worth, I....will probably change that.
Incidentally, I also see the way Gandalf was screwing with the trolls as an epic Persuade check.
Also, forgive the weapon pickup rule, I should have edited that out. It's an interaction with other houserules in my campaign.
The game rather explicitly left out sneak attacks, or anything else representing D&Dish archetypes of rogues or assassins. Some people don't like that, of course, but it's worth noting. Whenever this topic comes up the usual response is, "Use your bonus dice from rolling Battle and just narrate your sneak attack." (Which suggests that maybe players should be allowed to roll Stealth rather than Battle...)
In any event, if you want sneak attacks in your game, the problem I see with this rule is that for it to be useful requires decent skill with two weapons (although it could be argued that in Forward stance you can get by with lower skill). And the number of situations where giving up one attack in order to have +2 damage on a subsequent attack is limited. But your close combat fighters can't rotate through Rearward in order to use this, and your archers will most likely be better off staying Rearward. So another one in the category of "If your players think it's fun regardless of whether it makes tactical sense, then go for it."
I'm not sure I see why they'd have to have two decent weapon skills...this skill was meant to function similarly to Prepared Shot, but with a common skill involved. How we saw it working is, at the end of the round, a player would call rearward, fully intending to flank the enemy. Even if they don't have a ranged weapon, they can then call it in preparation for Surprise Attack.
What this ability was meant to illustrate was a character fading out of the battle and finding a route around the main part of the skirmish in order to race out of the woods, from behind rocks, etc. For lack of a better description, it is not your typical "physically vanish and then teleport behind an enemy" attack. It's supposed to be more of an aggressive outflanking maneuver wherein the flanker is helping bring down an enemy already engaged (whether mechanically or narratively) with another player character. As one who has been brutally blindsided in a fight before (awareness exercises in a decade's worth of martial arts) it is a severely unpleasant experience.
Further, the task was supposed to help some archers who are having difficulty hitting their high parry targets. Since archers typically have low melee skill, this task was designed to allow their common skills to influence how hard they hit. After all, it's not hard to absolutely murder an enemy that never saw you coming...
I do agree that I should probably up the value of success...it was just that our initial projection for it was STUPIDLY powerful, so we brought it back. The pendulum will have to swing back towards balance, I think.
EDIT: Also, as a sidenote, the most a player could do to a troll is -7. On one or two of the trolls, yeah. That's pretty powerful. On a couple more, the trolls still have 4d6 to roll. On the last couple, that negates their Favored AL bonus...they still get their 3d6.
EDITED EDIT: After doing test rolls, that does actually severely trivialize at least two of the troll basetypes. I will probably change it to be just -2, -4, -6 armor for a single round instead.