Assuming a standard type board and no dice mods. 

..or if you prefer - the same question inverted.  When don't you attack with 2?

I only attack with 2 when 1 is defending and taking that territory is critical to either elimination, bonus, or position.

AttilaTheHun wrote:

I only attack with 2 when 1 is defending and taking that territory is critical to either elimination, bonus, or position.

When you say "and", do you mean that you would never attack when 2 or 3 are defending?

Agree with Atilla.  I've probably attacked 2v2+ only a handful of times on this site.

If a 2v2+ is needed in my strategy -unless I am really desperate - it's better wait another turn -

But, I did successfully pull off a 2v2 with a d6vd7 recently to win a game - that was a lucky shot

yep yep yep

If I remember right the odds to win 2 v 1 (1 rollling against 1) is about 44%.

 I will only attack 2 v 2 or 2 v 3 when desperate. 

I will attack 2 v 1 if desperate or:

* The territory I am attacking from is so far from the front lines that they will never reach the hot spots anyway so might as well get them involved.

* I have surrounded a territory having 1 unit with 2 or 3 territories each having two units, so the reward is worth the risk (or sacrifice), plus the odds of winning 1 of 3 battles each at 44% is very high.

Desperation: I'll do it if I am definitely going for an Elimination.  

Strategy: I'll normally try it once or twice do it if I don't want to "waste units" at a dead end.  Depending on the map layout, I'll do it to remove a player from an area that they wouldn't then have easy access to (ie Battlestar).

Mix of the two: I'll occasionally do it for a Card if needed or I can spare the territories.

Mad Bomber wrote:

i challenge you take note the next 10 times you roll 2 v 1....it is way less than 44%.....way

I just checked MB, 41.67% for d6 Risk style

When you're desperate, even 40% is pretty good odds.

If I can think of just about any reasonable excuse, most if not all of the above included, I'll toss 'em at 2v1.  I'm hard pressed to add to the list, but one comes to mind.

If it will take away 1 army from an opponent's standard in-hand bonus, it's like getting a 2-fer.  40% seems like a good deal to me.

Statistically, in a 1v1 situation it is ALWAYS worth attacking 2v1 to deny your opponent a bonus, even just one army.

Breaking a +1 bonus is like killing 2 armies (the produced and the army itself), so naively you only need 33% to justify it. However, making the 2 armies into a 1-1 defending situation offsets that somewhat. In spite of that, it's still gotta be worth it, in a 1v1.

Statistically...but not always Strategically...

2v1 means attacker rolls 2 dice and defender rolls 1, right?

Attacker wins almost 58% of the time.  Why shouldn't attacker attack all the time (all things strategic being equal?

M57 wrote:

2v1 means attacker rolls 2 dice and defender rolls 1, right?

Attacker wins almost 58% of the time.  Why shouldn't attacker attack all the time (all things strategic being equal?

Ahh, others were going with the near 40% of 1v1 dice attacks (2v1 on the board), so I did too. Anyway, in a two-player game you'd need a reason involving defending critical spots or some sort of special consideration to not roll. ie it's a good idea to roll! 

In multiplayer games, "all things being equal" takes on a new meaning. Attacks better have meaning, at the very least a card grab. If all your 58% attacks are focused on one player, the players not involved in these battles benefit quite a bit. If you spread it out, 58% no longer looks so good. With an attack on two different players, you're killing 0.58 armies of each player and losing 0.84 of your own. All things being equal, you should lay off the dice in a multiplayer game.

A win is not just a territory lost for your opponent, it's also a territory gained for you.  Assuming a 1army/3territories bonus structure, and assuming that other players are spreading the love, that's a +.33 bonus for you and a -.33 bonus for your opponent (on average).  E.g., in a multi-player game, you don't know who else is going to attack the opponent you just attacked.  At a net .66, that makes the .68 look pretty much like a wash, so there needs to be other advantages.

In other words, It's looking like the instincts of the player's posting here are pretty on the mark.

I think for many multi-player games, at least in the beginning, most players are so far under the bonus limit that accounting for a territory bonus +.33 might not make sense.

Also, keep in mind that when you lose your 2 v 1, you are leaving a weak 1 army territory giving your enemy all the bonues that others have attributed to attaking 2 v 1.  So, I am saying that whatever you might think is good about attacking 2 v 1, you are giving to your opponent  58% of the time.

In general, I dont do it.

Good example:  

http://www.wargear.net/games/view/140112

anywhere I can take a territory, it's worth it, especially if he gets one less bonus.  Concept being: every turn we take each others continents away and get some of our own, but someone is left with more territories in the final analysis, and therefore more long term power to gain territory. It's worth it to try out some of those "attack with 2 situations", in order gain some kind of advantage, get rid of their territories in the middle of mine, or 2 kill a bonus of some sort. 

If there were another player in this game, things would be different, but for 58% of the time I'm losing my second guy, it's worth try to eek out any things I can. 

since reading this thread and going through the stats, i have been doing this more often than before.

I will do it in an absolute emergency. Almost never any other time.