I'm sure we could get a strong list of Simulgear strategy (and rule) questions going. I found these in threads I didn't wish to hijack:

Thingol: What could be done is to offer basic general gameplay hints for simulgear (ie. - ordering, when and when not to pretransfer, soaking up attacks, etc) and for general play (ganging up against a clearly dominating player to keep a game alive, what types of diplomatic messages to send thru PM, etc) without giving away trade secrets.

M57: I don't understand why it is advantageous to attack when the standard attack vs. looks disadvantageous.

I am not the most qualified individual to answer such questions, but I shall clumsily do so in the next two posts, hoping that others will find things to add.

Thingol asks for general advice for the uninitiated. The quick answer to most of his questions is that you do a lot of the same things you normally would. When someone is dominating, you and others need to gang up on that person just as you would in a turn-based game. You do have to adjust to the pace of one territory advances, and there aren't big card capturing bonanzas.

The simple answer to pretransfers is that you pretransfer from a territory that has a lesser need for armies to a territory that has a greater need.

In a non-fogged game, the threat of a pretransfer can be a deterrent. Suppose you have a two territory front each with 4 armies on them. Your opponent can attack with 12. Bordering your front is a territory with 9 units, 8 transferrable. If you split your pretransfers, your opponent could do damage against one of them. But you might load it all onto one or the other, which would damage your opponent. So your opponent has to decide if the risk is worth it. (The exact nuances will depend on the situation, the border modifiers, etc.)

In a fogged game, what is nice is you can have a weak front and do damage by pretransferring large armies where your opponent is not expecting them. This is a major tactic in fogged games.

The order of the attacks is where much of the subtlety lies. You have to guess whether it will be advantageous to attack early or attack late for any given attack, and the importance of that attack has to be weighed against the others. By and large, it is a guessing game, pokerish - you make educated guesses as to what your opponent will do, and make your own moves accordingly.

I can't wait for my answer.

M57: I don't understand why it is advantageous to attack when the standard attack vs. looks disadvantageous.

In a turn-based game, it is good to have lots of armies because it increases the probability of the success of an attack. However, the rate at which you and your opponent loses armies is exactly the same. With 6v6 dice, attacker has about a 6 to 5 edge probability-wise; with 7 v 6, it is 2 to 1.

This is not the case with Simulgear. The general rule of thumb is that, not only does having more armies increase the chance of success, it increases the relative expected value.

Suppose the dice are 60% for an attacking army to kill, and 75% for a defending army to kill. If you are attacking with 100 units and your opponent is defending with 12 units, you are going to get all 12 units (barring some miracle). Your opponent kills 9 on average. 12-9 attacker edge arising out of a 60%-75% disadvantage. In turn-based, if you had 100 versus 12 and a defensive modifier such as 6v7, you'd still succeed, but the defender kills more than 12 units.

Where is 60% versus 75% advantaging the defender? Even battles (and battles close to even where the attacker doesn't quite have enough to get an edge). If two players have 10 units ready to face off and attack each other, it is better to be the person who goes last. The first attacker loses more, and then the second attacker gets an increased expected value due to having more units than the defender when he attacks. If the die were reversed, 75% to 60%, you'd want to be the first to attack in an even battle.

To summarize: don't be fooled by the probabilities. Though they mean exactly what they mean, large forces influence the relative expected value in a way that doesn't happen in turn-based games.

Perhaps a quicker way to explain it is this: In turn-based, regardless of army sizes, the expected loss ratio of the two sides is fixed. In simulgear, in an attack, the expected losses on both sides depend on the number of armies both sides bring to the table. A 60%-75% attacker disadvantage is exactly that when the forces are equal, but can change to as high as a 100%-75% (expected) attacker edge when the attacker brings enough force.

I have a question.  Let's say I control territories A & B which can both attack C.   Is there any advantage or disadvantage to attacking with 40 units from A vs. 20 from A & 20 from B.

It seems like attacking from two different territories is worse, because C gets two chances to defend.   Is this right, and if is how much worse is it?

Well, let's do some math. Say there's a 60/75 border, and you are attacking a stack of 100 with 168 units. If you were attacking simply B -> A, then you'd expect to kill 100, and lose 75.

However, if you attack from two territories (B -> A and then C -> A) each with 84 units, you would still expect to kill 100, but now you would expect to lose 75 of B's stack on the first attack and another 38 or so from C's stack on the second attack. You still take the territory, but you've lost 113 units instead of 75. So by splitting your stack evenly, you lose 150% of what you would have lost with a single stack -- clearly better to go with one big stack than two middling stacks.

Now suppose there was a unit max of 100 on each territory -- how much difference does it make to max out B instead of splitting evenly between B and C? Attack first with 100 from B and next with 68 from C. This time, you expect to lose about 75 from B, but since you kill about 60, there are only 40 left to defend, meaning you lose about 30 from C. That comes out to 105 as compared to 113 lost with even stacks, or about a 7% reduction in losses. Not huge, but could be significant.

Kjeld wrote:

Now suppose there was a unit max of 100 on each territory -- how much difference does it make to max out B instead of splitting evenly between B and C? Attack first with 100 from B and next with 68 from C. This time, you expect to lose about 75 from B, but since you kill about 60, there are only 40 left to defend, meaning you lose about 30 from C. That comes out to 105 as compared to 113 lost with even stacks, or about a 7% reduction in losses. Not huge, but could be significant

I'm wondering if you should flip the order of the attack (i.e. attack with the 68 first then the 100). On the first attack, you would expect to kill about 40 and lose all 68 (since the 100 would expect to kill 75). That leaves about 60 for attack two. On your second attack, you kill about 60 and lose 45. So, in total you would lose 113 (68 + 45); however, you would capture the territory with about 55 units instead of 38. The extra units could be useful to defend against a counterattack in the same turn or to attack with the next turn.

Another thought - I think it's safe to say that either way, the battle will not be won on the first attack (your 100 would all have to hit which is unlikely). Given the numbers, the actual results will likely be more or less than the average. If I attack with the 68 first, I go in expecting to lose all 68. If my opponent only hits on 60% instead of the expected 75% - great! - then I have 8 units left. If my opponent hits on 90%, though, I still only lose the 68. On the other hand, if I attack with the 100 first, then I stand to lose anywhere from 0 units to all 100. If I lose less than the expected 75, then that's good, but those extra units may end up behind the lines instead of on the captured territory ready to attack or defend.

Thanks good answers.   The more of these scenarios we go through the better I understand this.  And good point CK66 - when you have uneven stacks,  you can attack with the big stack last to keep the largest army together.

CK66 wrote:

Kjeld wrote:

Now suppose there was a unit max of 100 on each territory -- how much difference does it make to max out B instead of splitting evenly between B and C? Attack first with 100 from B and next with 68 from C. This time, you expect to lose about 75 from B, but since you kill about 60, there are only 40 left to defend, meaning you lose about 30 from C. That comes out to 105 as compared to 113 lost with even stacks, or about a 7% reduction in losses. Not huge, but could be significant

I'm wondering if you should flip the order of the attack (i.e. attack with the 68 first then the 100). On the first attack, you would expect to kill about 40 and lose all 68 (since the 100 would expect to kill 75). That leaves about 60 for attack two. On your second attack, you kill about 60 and lose 45. So, in total you would lose 113 (68 + 45); however, you would capture the territory with about 55 units instead of 38. The extra units could be useful to defend against a counterattack in the same turn or to attack with the next turn.

Another thought - I think it's safe to say that either way, the battle will not be won on the first attack (your 100 would all have to hit which is unlikely). Given the numbers, the actual results will likely be more or less than the average. If I attack with the 68 first, I go in expecting to lose all 68. If my opponent only hits on 60% instead of the expected 75% - great! - then I have 8 units left. If my opponent hits on 90%, though, I still only lose the 68. On the other hand, if I attack with the 100 first, then I stand to lose anywhere from 0 units to all 100. If I lose less than the expected 75, then that's good, but those extra units may end up behind the lines instead of on the captured territory ready to attack or defend.

How many units advance is one thing, but there are other things to consider, For instance, let's say that in the turn between attacks, defender is unexpectedly able to send 40 reinforcements. Of course, you probably won't take the territory either way, but if you attack with the larger stack first, you should expect to lose 75 + 60 = 135.  Attack with smaller stack first and lose 68 + 75 = 143.

• Attack with the larger stack first and defender loses 60 + 40 = 100.
• Attack with the smaller stack first and defender loses 40 + 60 = 100.
• Attack with the larger stack first and defender will defend with 40 + 40 = 80 on the second round.
• Attack with the smaller stack first and defender will defend with 60 + 40 = 100 on the second round.

Disclaimer: This all assumes that I did the math right AND actually understand how #Dice work.

A good point by CK66, and a good caveat (though much less common to occur) by M57.

M57 wrote:

How many units advance is one thing, but there are other things to consider, For instance, let's say that in the turn between attacks, defender is unexpectedly able to send 40 reinforcements. Of course, you probably won't take the territory either way, but if you attack with the larger stack first, you should expect to lose 75 + 60 = 135.  Attack with smaller stack first and lose 68 + 75 = 143.

• Attack with the larger stack first and defender loses 60 + 40 = 100.
• Attack with the smaller stack first and defender loses 40 + 60 = 100.
• Attack with the larger stack first and defender will defend with 40 + 40 = 80 on the second round.
• Attack with the smaller stack first and defender will defend with 60 + 40 = 100 on the second round.

Disclaimer: This all assumes that I did the math right AND actually understand how #Dice work.

Yep, you would have to consider the possibility that the defender could reinforce between your attacks.

I like Hugh's explanation...very straightforward.

Hugh wrote:

Though they mean exactly what they mean, large forces influence the relative expected value in a way that doesn't happen in turn-based games.

I do understand that what constitutes a turn in the regular structure of play is divvied up in SG, but there is nonetheless an order to play.  I.e., each player attacks with one territory and play cycles through until all orders are depleted. The only thing simultaneous about SG is that all orders are placed at the same time.

Anyways, here's another question regarding how to accomplish delayed attacks.

Let's say I would like to delay an attack from A -> B.  I have 25 armies on A.  Can I place orders to:

1. attack B with 1
2. attack B with 1
3. attack B with 1
4. attack B with 1
5. attack B with 21

I seem to think that I have tried to do this and wasn't allowed to attack the same territory twice.

M57 wrote:

Hugh wrote:

Though they mean exactly what they mean, large forces influence the relative expected value in a way that doesn't happen in turn-based games.

I do understand that what constitutes a turn in the regular structure of play is divvied up in SG, but there is nonetheless an order to play.  I.e., each player attacks with one territory and play cycles through until all orders are depleted. The only thing simultaneous about SG is that all orders are placed at the same time.

Anyways, here's another question regarding how to accomplish delayed attacks.

Let's say I would like to delay an attack from A -> B.  I have 25 armies on A.  Can I place orders to:

1. attack B with 1
2. attack B with 1
3. attack B with 1
4. attack B with 1
5. attack B with 21

I seem to think that I have tried to do this and wasn't allowed to attack the same territory twice.

This would be disallowed.  The rules will only allow you one A -> B attack.  You could, however, attack from B to any other territories...with each unique territory constituting a separate order.

I wonder why this is disallowed. Is there a game-play related reason, or is it perhaps a GUI related thing?

M57 wrote:

I wonder why this is disallowed. Is there a game-play related reason, or is it perhaps a GUI related thing?

Gameplay related.  Essentially this would drastically increase the ability to stack orders.  Which has its Pros in added "strategory", but can be a Con if overly needed/used.

CK66 wrote:

Kjeld wrote:

Now suppose there was a unit max of 100 on each territory -- how much difference does it make to max out B instead of splitting evenly between B and C? Attack first with 100 from B and next with 68 from C. This time, you expect to lose about 75 from B, but since you kill about 60, there are only 40 left to defend, meaning you lose about 30 from C. That comes out to 105 as compared to 113 lost with even stacks, or about a 7% reduction in losses. Not huge, but could be significant

I'm wondering if you should flip the order of the attack (i.e. attack with the 68 first then the 100). On the first attack, you would expect to kill about 40 and lose all 68 (since the 100 would expect to kill 75). That leaves about 60 for attack two. On your second attack, you kill about 60 and lose 45. So, in total you would lose 113 (68 + 45); however, you would capture the territory with about 55 units instead of 38. The extra units could be useful to defend against a counterattack in the same turn or to attack with the next turn.

Another thought - I think it's safe to say that either way, the battle will not be won on the first attack (your 100 would all have to hit which is unlikely). Given the numbers, the actual results will likely be more or less than the average. If I attack with the 68 first, I go in expecting to lose all 68. If my opponent only hits on 60% instead of the expected 75% - great! - then I have 8 units left. If my opponent hits on 90%, though, I still only lose the 68. On the other hand, if I attack with the 100 first, then I stand to lose anywhere from 0 units to all 100. If I lose less than the expected 75, then that's good, but those extra units may end up behind the lines instead of on the captured territory ready to attack or defend.

This is an interesting point.  Does this only hold true for relatively even stacks of armies?

AttilaTheHun wrote:

This is an interesting point.  Does this only hold true for relatively even stacks of armies?

I assume you are talking about even stacks for the attack and the defense (so, 168 vs 100 in our example)?

I would say yes when two attacks will be required to take the territory. Assuming your goal is to take the territory and have the most units possible on that territory when you are done, I think you are best off making your stacks as lopsided as possible and attacking with the weaker stack first. Of course, there may be other factors that might change that (i.e. how likely is the defender to reinforce their units? Is there any possibility that the two territories you are attacking from can be attacked? etc.).

On the other hand, if victory can likely be achieved in a single attack, then I think you are best off attacking with the larger stack first. That way, the second attack becomes a transfer/fortify. For example, if you are attacking with two stacks of 100 and 68 versus 50 with 60/75 attack/defense odds, then you are likely to capture the territory by attacking with the 100. You would lose about 38 units leaving 62. Then the second "attack" of 68 would transfer at least 38 units (remember a max of 100 units on the territory) - possibly more if your 62 were attacked in between your two attacks.

Can someone let me know how 'invalid' orders are handled? 

For example, I know normally Order 1 is processes for each seat in sequence, then Order 2, etc as shown below for a game where 3 seats are occupied:

Order, Seat

1,1

1,2

1,3

2,1

2,2

2,3

What happens when one of my orders is invalid, like if I attack only a country I own.  For example, if my first order is invalid is my second order executed immediately, or am I skipped and then my second order is executed the next iteration?  See below (I would be seat one in this example):

1, -skip invalid order

1,2

1,3

2,1

2,2

2,3

or

2,1 skip first order and execute second order

1,2

1,2

3,1

2,2

2,3

Thanks,

Squint