using chariots to stifle your opponents?

[Watiggi]

I personally have completely underestimated the withdrawl promotions. I guess I have never considered the withdrawl rate as being anything worth factoring in to the equation. Having a 30% withdrawl meant to me that if I go against another strong unit that gives me bad odds, then I have 30% chance of surviving (plus the odds of winning). It just didn't say much to me. I might have to look into the withdrawl a bit more. But in giving them withdrawl promotions, they are missing out on other strength/odds of winning enhancement promotions.

I am actually surprised as to why the Keshik didn't have a lowered strength (even reduced below par to 5 or 4) and have a rather high withdrawl rate added. Afterall that was their kind of strategy.

I wonder what the ramifications would be if the odds of winning calculator were replaced by an odds of surviving calculator. I think that would be interesting. You would be looking at combat not from whether you could win the fight, but whether you could survive it (where winning would be the unknown element). That could leave it open for more unresolved fights between units too (like the musketmen defeating a gunship, where in reality, if the musketmen were so defensively powerful, it wouldn't kill the gunship, it would just mean that the gunship couldn't kill the musketmen. Iow, just because a modern unit cannot defeat an old unit, it doesn't mean the modern unit deserves to die as a result. hmmm, maybe the gunship should have a 50-70% withdrawl rate - especially considering its advantages).

Watiggi

 

[scottcstoness]

I personally have completely underestimated the withdrawl promotions. I guess I have never considered the withdrawl rate as being anything worth factoring in to the equation. Having a 30% withdrawl meant to me that if I go against another strong unit that gives me bad odds, then I have 30% chance of surviving (plus the odds of winning). It just didn't say much to me. I might have to look into the withdrawl a bit more. But in giving them withdrawl promotions, they are missing out on other strength/odds of winning enhancement promotions.

I wonder what the ramifications would be if the odds of winning calculator were replaced by an odds of surviving calculator.
Watiggi

Unfortunately, a 30% withdrawl rate doesn't help nearly as much; flanking only helps out a lot for units which have some sort of natural withdrawl rate.

In terms of an odds of survivial calculator, it is pretty simple.

If OddsWin represents the odds of winning as a fraction (e.g. 35% = 0.35, 78% = 0.78), and Withdrawl represents the odds of withdrawing if you don't win the combat, then the odds survival are:

OddsWin + Withdrawl * (1 - OddsWin)

That is, if you have a 40% chance to win and a 30% withdrawl rate, you have a 0.4 + 0.3*0.6 = 0.4 + 0.18 = 0.58 = 58% chance to survive. With a 60% withdrawl rate, on the other hand, you have a 0.4 + 0.6*0.6 = 0.4 + 0.36 = 76% chance to survive. Huge difference.

 

[Watiggi]

I recon that odds of survival calculator could be really useful when deciding whether to attack or not.

It's a pitty the combat system generally requires a winner and looser and not just allow damaged units with no clear winner or looser (considering they in essence become a fraction of a division or battalian or something instead of being a unit that looses it's strength).

It would make more sense with the musketmen vs gunship and as such, an odds of survival calc would become more appropriate than an odds of winning. Damn I wish I could mod, even just to test and see. Simply allow units to fight a random number of preselected rounds and then disengage. Have the odds of survival calc instead and see what happens.

Watiggi

 

[scottcstoness]

But in giving them withdrawl promotions, they are missing out on other strength/odds of winning enhancement promotions.

Let's consider an Immortal with 2 promotions against a fortified archer in a city.

The archer's strength is 3 + 25% for fortify, +50% for the city for a total
strength of 3 + 75% + 1 first strike.

An immortal with 2 combat promotions is 4 + 20% base strength is 4.8, with +50% against archers and 30% withdrawl. According to thread 137615,
and the site http://c4combat.narod.ru/c4c.htm, the odds are 71% for a win, with 30% withdrawl on the other 29%, which is about 9%. So, you'd survive 80% of the time.

With flanking 2, it is 4 base str with +50% against archers and 60% withdrawl. Here, the odds are 60% to win, + 60% of 40% or 24% to withdraw, so 84% chance to survive. Actually, I forgot that flank 2 ignores first strikes. That makes the odds 66% to win, + .6 * .34 = 20% withdraw, or 86% chance to survive.

In this case, it is pretty close -- one has a better chance to win, the other has a better chance to survive.

But now consider a archer with garrison 2 in a city with 40% cultural defense. Now archer is 3 + 25%F, +50%C, + 45% Garrison, +40% culture, for base str 3 + 160%.

Attacking with a combat 2 immortal is 4.8 + 50% against archers, so we have 4.8 vs. 3 + 110% = 6.3. Chance to win is 18%, another 25% to withdraw, total survival chance 43%, the archer will be left with 58 HP on average

Attacking with a flanking 2 immortal is 4 vs. 6.3, but no first strike. Now chance to win is only 5%, but chance to withdraw is .6 * 0.95 = 57%. So, survival chance is 62%. The archer will be left with 65 HP on average.

Here, clear advantage for the flanking 2 immortal; does almost as much damage when it attacks, and is almost 50% more likely to survive the attack. It also gets an XP for retreating.

It gets even more striking if the units survive to get 10 XP (guess which is more likely to do so!). Then the 5.2 str immortal has a 21% chance to win, 24% chance to withdraw, for 45% survivial chance. The 4.4 str flanker has an 11% chance to win, 53% chance to withdraw, 64% survival chance. Both archers are left with about 53 hps on average.

Of course, don't get your flankers caught out in the open. For city attack, best to run around with a combat/pinch promoted axeman to take out attacking spears, and maybe a medic spearman to handle opposing mounted units. That is, of course, if you haven't plundered their resources yet.

You can even be effective against longbowmen, garrison 1, fortified, culture 60%.

Attack    LB str    Win%     Surv%   LB HP    LB HP remaining
1            13.8       0           60       100         91
2            12.6       0           60        91          81
3            11.2       0           60        81          66
4             9.1       3           61         66         47
5             6.5       25          70        47          29 (if it survives)
6             4.0       65          83        29          13 (if it survives)
7             1.8       99          99

Assuming that the calculator at http://c4combat.narod.ru/c4c.htm is correct, and I'm using it properly, on average you'll need to throw 5 flanker 2 immortals against the fortified, garrison longbowman to get to decent combat odds. But, on average, 3 of those 5 will survive through withdrawing. Its a lot of units running around to do it, but with about 20 immortals you can safely take out a city with a couple of fortified garrison longbowmen, and have enough troops left to secure the city and protect the wounded while they heal.

Obviously not a good long-term strategy -- 50 immortals to take out 3 cities, even if 40+ of them survive, is not a good deal given how long they take to heal. But a quick 3-city push to secure a resource or good city sites, followed by a peace treaty, wait to heal, repeat, can be pretty effective if you can afford to keep a standing army of 50-60 immortals!

 

[scottcstoness]

I recon that odds of survival calculator could be really useful when deciding whether to attack or not.

It's a pitty the combat system generally requires a winner and looser and not just allow damaged units with no clear winner or looser (considering they in essence become a fraction of a division or battalian or something instead of being a unit that looses it's strength).

It would make more sense with the musketmen vs gunship and as such, an odds of survival calc would become more appropriate than an odds of winning. Damn I wish I could mod, even just to test and see. Simply allow units to fight a random number of preselected rounds and then disengage. Have the odds of survival calc instead and see what happens.

Watiggi

A great mod would be to allow the ability to break off an attack after taking a certain number of losses -- a tactical withdrawl of sorts. Of course, you'd have to give the defender the same ability. For example, units could try to retreat after sustaining 50HP damage, which would probably represent about 10-25% actual real-life battle casualties; healing would represent collecting and re-arming scattered ineffective troops as much as anything.

Since ancient battles were much more likely to result in massacres once one side started losing, you could give various units a % chance to break off after dropping to 50HP, say 10% for ancient era, 30% for medival, 50% for renaissance, 70% for industrial, 90% for modern. The flanking could give an additional percentage to break off at 50HP, and then another chance at 0HP for the surviving cadre to flee to fight again.