How do retreats work?

[Madroc]

I play PTW 1.27, and I have just begun my first Monarch game.
I have never really been a warmonger, but I am trying to employ REX strategy and I spent most of the AA at war (without accomplishing much, for a variety of reasons.)
I know that "fast" units (in this case, horsemen and chariots) have the ability to retreat when they are on their last health bar. But how does that work? I lost several chariots on that assumption, even though I was careful to leave one movement point left, and an open square to run to,and they were attacking a spearman (or other 1-move unit). But the chariot would keep on fighting the good fight until the bitter end. It seems to me that the retreat is totally random, but I know this is not the case. How does it work?

 

[Reno]

Chariot can only retreat into a sguare where there is a road or flat grond if there is no road or flat ground the chariot cannot retreat, When a 2 movement unit is at its last health it will retreat if the unit is attack again on the same turn it will not pull out and thus will die.

 

[IbnSina]

Units do not always retreat. The RNG is involved. When they get down to their last hit point, they roll the die, and then some fraction of the time they get a retreat. I forget the probability... Retreat does not require any special ground - no movement actually takes place - it just means that the unit stops attacking. You do not need to leave the movement point, though doing so allows you to pull the unit away, or fortify it in town after it is captured.

 

[boogaboo]

I play on C3C, and I think I understand that retreating is a POSSIBILITY for those units who theoretically can.
Not every retreatable unit actually retreats.
If I understand correctly, the chances of retreating are 34/50/58/66 percent for conscript/regular/veteran/elite, and are called "retreat bonus" in the editor.

Also, this applies both to retreating when failing to attack and to retreaing when failing to defend.
Please correct me if I'm wrong.

 

[maverick_mw]

Not entirley sure, but my understanding from fiddling with the editor is:

"Any unit can retreat (or has a random chance to do so) if it has remaining movement points."

So, if you move your horsemen one square (out of max movement 2) and THEN attack, it will not retreat as it has no movement points left.

But, if you attack with a horseman on it's first movement point, it CAN retreat, because it still has an unused movement point. Retreating is still random though, not sure what it depends on. Therefore, any defensive unit has not used any movement points and therefore is likely to retreat when down to 1 HP.

And this takes roads into consideration too, as they count for 1/3, a unit can still have 1/3 of a movement point left and still retreat. Not sure about effect of railroads.

Am I right?

 

[Smartsquid399]

I play on C3C, and I think I understand that retreating is a POSSIBILITY for those units who theoretically can.
Not every retreatable unit actually retreats.
If I understand correctly, the chances of retreating are 34/50/58/66 percent for conscript/regular/veteran/elite, and are called "retreat bonus" in the editor.

Also, this applies both to retreating when failing to attack and to retreaing when failing to defend.
Please correct me if I'm wrong.

You are correct. The retreat bonus is the chance of retreat. In my opinion though, I think the more movement points you have to spare the higher the chance should be.

 

[boogaboo]

Should and is are different issues. It SHOULD be also reflected by the road movement, but, hey, we didn't develop this software, just want to find out how it DOES work..

Thanks for reassuring me that this is the retreat bonus meaning..

 

[punkbass2000]

Also, units never retreat when the opponent has one hitpoint.

 

[boogaboo]

Also, units never retreat when the opponent has one hitpoint.

Good info!
To sum up:
Our unit will use the retreat bonus chance to retreat or not retreat in front of an opponent unit (attack or defence) if and only if all the following aplly:
1. The opponent unit has 2HP and upwards.
2. Our unit has 1HP only.
3. Our unit would have had at least 1/3 (...) movement point after the attack.

The only remaining question is weather it matters if the opponent unit has more movement points than our unit? If so, this should be added.
But I was playing Aztec too long not to notice things (but haven't seen a 2MP unit retreat in front of a 3MP unit, please tell me if you have)

 

[Pfeffersack]

Never saw a unit retreat which has more than one hitpoint, so I assume it has no influence how many hitpoints the enemies has, as long as it is has more than 1.

 

[JMK]

This article has been written by BomberEscort
http://forums.civfanatics.com/showthread.php?postid=1278822#post1278822
What we know about retreating...
When the Attackers retreat value is 50 percentage points greater than the Defender, the Attacker always retreats
When the Attackers and Defenders retreat values are zero, the attacker will never retreat
The slow-Defenders retreat bonus does factor into the ability of the fast-Attacker to retreat. The greater the difference in the attackers favor, the greater the chance of retreat.
There is no way to create a slow-unit that will always block a fast-units retreat.
The range for an elite fast-attacker is 55-75% retreat in valid cases.
The range for a veteran fast-attacker is 50-66% retreat in valid cases.
The range for a regular fast-attacker is 43-56% retreat in valid cases.
The range for an conscript fast-attacker is 29-38% retreat in valid cases.
The best-guess for the retreat formula is:

(100*Attacker Retreat Bonus)/(50+Defense Retreat Bonus)

Credit to Oystein for this discovery...

This formula is also modified by the chance of the fast-attacker/fast-defender reaching 1hp before the slow-defender/slow-attacker reaches 1hp... The larger the difference between A/D values, the less this effects the results. Testing above was done with a 2 Attack fast-attacker v. 1000 Defense slow-Defender...

 

[Madroc]

OK... so there is a random element after all... thanks for the info.
Does anyone know of any good articles on how to properly use "fast" units? So far I've just been using them as regular units that arrive on the scene faster, which is probably a waste. Although if I can't rely on a unit to retreat, I don't see how I can develop special tactics for them...

Edit:Sorry JMK... I posted at the same time as you... will check out your link... thanks

 

[boogaboo]

The range for an elite fast-attacker is 55-75% retreat in valid cases.
The range for a veteran fast-attacker is 50-66% retreat in valid cases.
The range for a regular fast-attacker is 43-56% retreat in valid cases.
The range for an conscript fast-attacker is 29-38% retreat in valid cases.
The best-guess for the retreat formula is:

(100*Attacker Retreat Bonus)/(50+Defense Retreat Bonus)

Credit to Oystein for this discovery...

I wrote in my first post here values of retreat bonus that seem to match the average of your experiments... "the chances of retreating are 34/50/58/66 percent for conscript/regular/veteran/elite".

I now have an excel file of myself, and this is a real math puzzle..

 

[Grogs]

OK... so there is a random element after all... thanks for the info.
Does anyone know of any good articles on how to properly use "fast" units? So far I've just been using them as regular units that arrive on the scene faster, which is probably a waste. Although if I can't rely on a unit to retreat, I don't see how I can develop special tactics for them...

Edit:Sorry JMK... I posted at the same time as you... will check out your link... thanks

Tactically, I don't pay much attention to the retreat ability of my fast units. Even under the most ideal of conditions, the fast attack unit won't return to the battle for 3 turns (1 turn to town w/ barracks, 1 turn healing, 1 turn back.) In most cases it takes 4-5. Usually by that time I've already taken the city and moved on to the next one.

On a strategic level though, I can assume I need ~50% fewer fast attack units needed for a campaign because of the retreat ability.

 

[rysingsun]

Also, units never retreat when the opponent has one hitpoint.

this explains a lot! it explains why my new units retreat much more often than my obsolete units. in case you are wondering about this statement ...

my obsolete units rarely attack anything that hasnt been reduced to one hitpoint with artillery. newer units often do.