Auto-evasion

[the_iron_duke]

In FoG I, whether a mobile unit would evade or stand and fight could be set by the human player, subject to a cohesion test. In FoG II, it appears to be completely automated. My Light Spear/Sword cavalry evaded from another unit of the same type when charged. If two armies made up of these units were in a battle together, would they ever fight, or would they keep evading each other?

[TheGrayMouser]

In FoG I, whether a mobile unit would evade or stand and fight could be set by the human player, subject to a cohesion test. In FoG II, it appears to be completely automated. My Light Spear/Sword cavalry evaded from another unit of the same type when charged. If two armies made up of these units were in a battle together, would they ever fight, or would they keep evading each other?

Were the quality levels the same or did one have a leader? If the combat is roughly equal, they should fight. ( have certainly seen it in my games)
I'm not sure if there is any randomness to it.

[rbodleyscott]

In FoG I, whether a mobile unit would evade or stand and fight could be set by the human player, subject to a cohesion test. In FoG II, it appears to be completely automated. My Light Spear/Sword cavalry evaded from another unit of the same type when charged. If two armies made up of these units were in a battle together, would they ever fight, or would they keep evading each other?

Were the quality levels the same or did one have a leader? If the combat is roughly equal, they should fight. ( have certainly seen it in my games)
I'm not sure if there is any randomness to it.

There is slight randomness, but what you say is largely correct. Also, even if they would like to evade, they won't do so if the chargers start their (entire) move in the adjacent square, because they would have too high a chance of being caught.

[the_iron_duke]

Yeah, they were identical except the other guy was a general, which would explain it.

How would this rule affect a game where two identical non-shock cavalry armies faced off, with the only difference being one side had better armour or quality. How would melee combat occur?

[rbodleyscott]

Yeah, they were identical except the other guy was a general, which would explain it.

How would this rule affect a game where two identical non-shock cavalry armies faced off, with the only difference being one side had better armour or quality. How would melee combat occur?

If the side with better troops moves up to the adjacent square in one turn, and then charges in the next turn, the enemy will not try to evade unless they are hugely outmatched, because of the high chance of being caught.

Also, units can only evade once in a turn, if they are charged again they automatically count as rear attacked.

[the_iron_duke]

Yeah, they were identical except the other guy was a general, which would explain it.

How would this rule affect a game where two identical non-shock cavalry armies faced off, with the only difference being one side had better armour or quality. How would melee combat occur?

If the side with better troops moves up to the adjacent square in one turn, and then charges in the next turn, the enemy will not try to evade unless they are hugely outmatched, because of the high chance of being caught.

Also, units can only evade once in a turn, if they are charged again they automatically count as rear attacked.

Ah, thanks.

What about for the other, weaker side, though? Once they are being chased, can they stop themselves being pursued off the board? After first evading, their backs are turned to the enemy and it takes a turn to turn around, so it wouldn't be possible to then charge, and the next turn if the enemy charges, the unit will be auto-evading again.

In my current game, the cavalry general chasing my cavalry stopped pursuing when my unit was at the square at the edge of the map, allowing my cavalry to turn around and charge it the following turn.

[rbodleyscott]

Yeah, they were identical except the other guy was a general, which would explain it.

How would this rule affect a game where two identical non-shock cavalry armies faced off, with the only difference being one side had better armour or quality. How would melee combat occur?

If the side with better troops moves up to the adjacent square in one turn, and then charges in the next turn, the enemy will not try to evade unless they are hugely outmatched, because of the high chance of being caught.

Also, units can only evade once in a turn, if they are charged again they automatically count as rear attacked.

Ah, thanks.

What about for the other, weaker side, though? Once they are being chased, can they stop themselves being pursued off the board? After first evading, their backs are turned to the enemy and it takes a turn to turn around, so it wouldn't be possible to then charge, and the next turn if the enemy charges, the unit will be auto-evading again.

In my current game, the cavalry general chasing my cavalry stopped pursuing when my unit was at the square at the edge of the map, allowing my cavalry to turn around and charge it the following turn.

Well generally speaking, if your troops are at a disadvantage, evading would seem a good idea. They have earned their pay by keeping a more expensive enemy unit out of the battle. Even if they evade off the battlefield they may return.

However, it is usually possible to avoid fleeing off the battlefield by continuing to retreat but changing direction, and hence leading the superior enemy unit(s) on a wild goose chase.

[the_iron_duke]

If a pursuer chases an evader off the field of battle, do they have equal chance of returning to the battlefield?

[rbodleyscott]

If a pursuer chases an evader off the field of battle, do they have equal chance of returning to the battlefield?

Pursuers do not chase evaders off the battlefield. (Only routers).

[the_iron_duke]

Thanks. Could you also clarify exactly what the conditions are regarding whether a unit will try and evade or not? Take the example below. The Velites skirmishers did not try and evade from the elephants when charged from one square in front of them. In testing, and with no troops behind the javelinmen, the javelinmen did evade from the elephant when charged from one square in front. So that would suggest it was the troops behind the javelinmen in the example below that were interfering with the javelinmen's evasion routes, although I would have thought that at least the ones with a space behind them would be able to evade. What is the reason?

[rbodleyscott]

Thanks. Could you also clarify exactly what the conditions are regarding whether a unit will try and evade or not? Take the example below. The Velites skirmishers did not try and evade from the elephants when charged from one square in front of them. In testing, and with no troops behind the javelinmen, the javelinmen did evade from the elephant when charged from one square in front. So that would suggest it was the troops behind the javelinmen in the example below that were interfering with the javelinmen's evasion routes, although I would have thought that at least the ones with a space behind them would be able to evade. What is the reason?

Mainly that they are less likely to evade if the enemy start in an adjacent square, because they might get caught. If there is greater separation they are more likely to evade because more likely to escape.

Incidentally, they can only pass through one unit of friends, not to units one one behind the other.

[the_iron_duke]

Would the fact that the elephants are in a line adjacent to each other be a factor, i.e. were the skirmishers less likely to evade because there were additional elephants diagonally in front of them, as well as an elephant directly in front? If there were no troops behind the Velites would they have evaded?

I'm finding the auto-evasion rules one of the biggest changes from FoG I, where every individual skirmisher could be individually set to one of three settings: an auto-evasion setting (governed by chance of winning in combat), always evade, or never evade (subject to cohesion test). I almost never used auto-evasion in FoG I, so I always knew what a skirmisher would do if charged. That FoG II has only auto-evasion is perhaps an improvement, as the user-controlled setting of light troop evasion in FoG I meant that skirmishers were especially useful as tactical tools. It does mean a need to gain a greater understanding of what governs their automatic behaviour, though.

[rbodleyscott]

Would the fact that the elephants are in a line adjacent to each other be a factor, i.e. were the skirmishers less likely to evade because there were additional elephants diagonally in front of them, as well as an elephant directly in front? If there were no troops behind the Velites would they have evaded?

No and no.

It does mean a need to gain a greater understanding of what governs their automatic behaviour, though.

You probably won't be able to predict it completely because it is a balance between how badly they are outmatched in combat, and how likely they are to be caught if they evade.

The important thing to remember is that all movement by a charger (even prior to the actual charge) uses up AP, so even if a unit moves right up to the skirmishers before charging (on the same turn) it probably won't catch the skirmishers if it started its turn further away, and hence the evaders are more likely to escape and hence the unit is more likely to evade.

To maximise the chances of catching evaders, the chargers need to start their (whole) turn in an adjacent square.

Hence archers and slingers should always try to extend the distance to 2 squares before shooting, so that they can escape. Javelinmen have to be more circumspect since they only shoot 1 square. If possible they should approach the enemy from their flank, so the enemy cannot charge them.

[the_iron_duke]

After some more testing with various differentials, it looks as though the disrupted status of the elephants was making the difference as to whether the javelinmen would try and evade or not. Even disrupted, though, the elephants have much greater chances of victory (Impact: 22%/76%/2%; Melee: 16%/77%/7%). When being charged by steady elephants, the javelinmen were evading, but charged by disrupted elephants, the javelinmen would generally not evade, although they would sometimes, and seemed to be more likely to do so when there were friendly troops behind them to evade through.

[rbodleyscott]

After some more testing with various differentials, it looks as though the disrupted status of the elephants was making the difference as to whether the javelinmen would try and evade or not. Even disrupted, though, the elephants have much greater chances of victory (Impact: 22%/76%/2%; Melee: 16%/77%/7%). When being charged by steady elephants, the javelinmen were evading, but charged by disrupted elephants, the javelinmen would generally not evade, although they would sometimes, and seemed to be more likely to do so when there were friendly troops behind them to evade through.

As I say, the calculation is a combination of how badly they are outmatched vs their chance of being caught. Obviously they are less outmatched when the elephants are disrupted, so less likely to risk being caught by evading. (Which is much worse than standing, because it counts as an automatic full-effect rear attack).

[the_iron_duke]

Could we get more info on how units decide to auto-evade, please?

So, if a charging unit mouses over a potential target, the percentage combat odds are shown. If odds are equal, the unit that is the target of the charge will not evade. How bad do the odds have to be for a target to run away (assuming the target is two or more squares away)? What is the range of probabilities in which chance will be a factor (as opposed to the target always evading or always standing)? Are both the Impact and Melee odds taken into account?

[nikgaukroger]

Could we get more info on how units decide to auto-evade, please?

So, if a charging unit mouses over a potential target, the percentage combat odds are shown. If odds are equal, the unit that is the target of the charge will not evade. How bad do the odds have to be for a target to run away (assuming the target is two or more squares away)? What is the range of probabilities in which chance will be a factor (as opposed to the target always evading or always standing)? Are both the Impact and Melee odds taken into account?

Personally I think leaving that unknown, as "fog-of-war" is a better game mechanism.

[the_iron_duke]

Personally I think leaving that unknown, as "fog-of-war" is a better game mechanism.

The unknown should exist in the random element of the result, rather than a lack of knowledge of how evade mechanics work.

[rbodleyscott]

Could we get more info on how units decide to auto-evade, please?

So, if a charging unit mouses over a potential target, the percentage combat odds are shown. If odds are equal, the unit that is the target of the charge will not evade. How bad do the odds have to be for a target to run away (assuming the target is two or more squares away)? What is the range of probabilities in which chance will be a factor (as opposed to the target always evading or always standing)? Are both the Impact and Melee odds taken into account?

It depends on the combat chances, taking into account both impact and melee, and also on how likely the unit is to get caught. It also takes into account nastier enemy units that could charge this turn if the unit stays to fight the initial charger.

[MikeC_81]

Personally I think leaving that unknown, as "fog-of-war" is a better game mechanism.

The unknown should exist in the random element of the result, rather than a lack of knowledge of how evade mechanics work.

I am heavily in agreement with this statement. A dedicated player if they so choose could probably come up with a reasonable model if they were willing to grind out umpteenth hours of trials. There is no reason for this to be a black box to the player. It would be the same as hiding PoA/combat power calculations from the player and just telling them to go by "gut feeling".

If you want to have inherent chance in the game on this mechanic, that is perfectly ok but the player needs to be able to make some sort of educated decision on this. Right now it is a rather arcane process of gut feeling, past experience and guessing at the relative weight of factors involved.