Poor odds attacks

[HansBolter]

Why do Japanese attacks at poor to abysmal odds routinely result in greater casualties for thier opponents than for themselves?

[herwin]

ORIGINAL: HansBolter

Why do Japanese attacks at poor to abysmal odds routinely result in greater casualties for thier opponents than for themselves?

HISTORY

Historically, the winner of a land battle was almost always the side with the lower percentage casualties. So if you attacked at 3-1 and lost two of your side for every casualty on the other side, you still won.

See R L Helmbold's paper on this.

Note added--I remember the WWII data that I worked with (about 500 1-day battles in all theatres of operations) showed the weaker side taking lower actual casualties unless the combat power ratio was 6-1 or higher.

[FatR]

    Because for the first several months of GC they generally fight inexperienced troops? 

[erstad]

Also, note that many of those casualties may be occurring the fire phase, not the assault phase. In many cases, they have a strong numerical advantage in the fire phase, both in terms of number of troops involved and in the amount of firepower/troop.

[jwilkerson]

As far as casualties are concerned - it is perhaps best to think of land combat as three separate phases:

01 - Fire Phase

02 - Assault Phase

03 - Retreat Phase

The fire phase involves shooting by both sides - and determines all the casualties unless there is a retreat. This shooting phase does not use any consideration of odds.

The assault phase is where the odds come in - basically you are looking for a (modified) 2 to 1 odds for the attacker to retreat the defender. The is no shooting or casualty causing activity during this odds comparison. The odds comparison is for the purpose of determining whether there is a retreat.

The retreat phase will cause casualties to the defender if the defender is retreated.

And BTW this is the same as it has always been in this system - nothing new to AE.

[HansBolter]

ORIGINAL: jwilkerson

As far as casualties are concerned - it is perhaps best to think of land combat as three separate phases:

01 - Fire Phase

02 - Assault Phase

03 - Retreat Phase

The fire phase involves shooting by both sides - and determines all the casualties unless there is a retreat. This shooting phase does not use any consideration of odds.

The assault phase is where the odds come in - basically you are looking for a (modified) 2 to 1 odds for the attacker to retreat the defender. The is no shooting or casualty causing activity during this odds comparison. The odds comparison is for the purpose of determining whether there is a retreat.

The retreat phase will cause casualties to the defender if the defender is retreated.

And BTW this is the same as it has always been in this system - nothing new to AE.

Thank you for bursting my bubble. [:(]

Let me get this straight.......attackers can NEVER take casualties in the assault phase?

I'll repeat.....attackers can NEVER take casualties in the assault phase?

What on earth were the people responsible for designing this game thinking?[&:]

What you are really telling me is that the assault phase is really NOT an assault phase, it's a Retreat Determination Phase.

If it were an assault phase, then by all tenents of warfare, it would be causing greater casualties to the attacker than to the defender.

You are also telling me that the so called combat odds are not combat odds at all.

The way these so called "odds" are reported is extremely misleading

[herwin]

ORIGINAL: jwilkerson

As far as casualties are concerned - it is perhaps best to think of land combat as three separate phases:

01 - Fire Phase

02 - Assault Phase

03 - Retreat Phase

The fire phase involves shooting by both sides - and determines all the casualties unless there is a retreat. This shooting phase does not use any consideration of odds.

The assault phase is where the odds come in - basically you are looking for a (modified) 2 to 1 odds for the attacker to retreat the defender. The is no shooting or casualty causing activity during this odds comparison. The odds comparison is for the purpose of determining whether there is a retreat.

The retreat phase will cause casualties to the defender if the defender is retreated.

And BTW this is the same as it has always been in this system - nothing new to AE.

Sounds like design for effect using a breakpoint model. In reality, we know breakpoint models are invalid.

In 1971, Robert L. Helmbold published Decision in Battle: Breakpoint Hypotheses
and Engagement Termination Data (Rand R-772-PR, June 1971). This paper is a
watershed in battle termination theory, for in it breakpoint hypotheses of
whatever nature are finally eliminated as an explanation of battle termination.

Helmbold challenged analysts to develop an adequate battle termination theory,
identifying seven properties such a theory should possess:
1. The theory should have simplicity and naturalness.
2. It must reproduce the observed quasi-exponential shape of emperical casualty
distribution curves.
3. The winner's casualty-fraction distribution curve must lie above and to the
left of the loser's casualty-fraction distribution curve.
4. The theory must address the separate casualty distribution curves observed for
the category I and category II battles. (In category I battles, both sides have
the option of disengaging; in category II battles, one side is not in a position
to disengage. Category I battles are typically resolved by the time the loser has
lost 10% of his force. Category II battles are typically resolved around 25% losses
for the loser. Winners typically lose half the percentage casualties that losers
lose. 90% of the time, losers lose a higher percentage of their force.)
5. The theory must define the relationship between the two side's casualty
fractions in accord with actual empirical data.
6. The theory should explain why the loser's and winner's casualty fraction
distributions are approximately the same independently of who was the attacker
and defender.
7. The theory should be useful for simulation.

Helmbold's results suggest that force commanders take into account both casualty
rates and relative force strength and position. In relatively balanced situations,
commanders base their sense of whether they are winning or losing on the relative
percentage casualty rates. In unbalanced situations, the positions and strengths
of the forces are taken into account. In balanced situations, the battle continues
until the commanders are convinced that the relative casualty rate (percentage
casualties per unit of time) favours one or the other side. In unbalanced situations,
the commanders don't bother to assess the relative casualty rates. The commander's
decision process is driven by the incurring of casualties. Situations then leave the
zone of acceptable force ratios at a constant percentage rate relative to
percentage casualties, which produces a quasi-exponential casualty distribution curve.

In gaming terms: both commanders begin by deciding whether the battle is worth
fighting and then assess the balance of forces continuously. Every few percentage
own casualties, they reassess the situation using both the current force ratio
and the relative casualty rates. If the battle is no longer worth fighting,
they break off; otherwise they continue. Fog of war plays a role in this,
producing an unexpected victory for the losing side about 10% of the time.

The game is not too far off, though, if we assume the commanders (the players) are making the decision on a daily basis. It's definitely better than the usual wargame, so keep it. The only tweaking the game designers really need to do is making sure the daily casualty rates are reasonable, and providing the players with a mechanism for ordering an orderly retreat.

[stuman]

ORIGINAL: HansBolter

ORIGINAL: jwilkerson

As far as casualties are concerned - it is perhaps best to think of land combat as three separate phases:

01 - Fire Phase

02 - Assault Phase

03 - Retreat Phase

The fire phase involves shooting by both sides - and determines all the casualties unless there is a retreat. This shooting phase does not use any consideration of odds.

The assault phase is where the odds come in - basically you are looking for a (modified) 2 to 1 odds for the attacker to retreat the defender. The is no shooting or casualty causing activity during this odds comparison. The odds comparison is for the purpose of determining whether there is a retreat.

The retreat phase will cause casualties to the defender if the defender is retreated.

And BTW this is the same as it has always been in this system - nothing new to AE.

Thank you for bursting my bubble. [:(]

Let me get this straight.......attackers can NEVER take casualties in the assault phase?

I'll repeat.....attackers can NEVER take casualties in the assault phase?

What on earth were the people responsible for designing this game thinking?[&:]

What you are really telling me is that the assault phase is really NOT an assault phase, it's a Retreat Determination Phase.

If it were an assault phase, then by all tenents of warfare, it would be causing greater casualties to the attacker than to the defender.

You are also telling me that the so called combat odds are not combat odds at all.

The way these so called "odds" are reported is extremely misleading

But it is just semantics to a point. Forget the terms used for a moment and just focus on the average results. Seems to work quite well to me.

[HansBolter]

Thanks for the replies gentelmen.

It should be obvious that the point I was trying to make in my overly dramatic way is that the Assault phase is simply mislabeled.

A basic litmus test for labeling a phase as "Assault" should be whether or not casulaties are being incurred.

The idea that an assault can take place with a guarantee of NO casulaties to the attacker screams out to me that it SHOULDN'T be labeled an assault phase.

It apparently would be more correctly and properly labeled as a Retreat Determination Phase.

Semantics, if you wish, but how things are perceived is often important so the nuances of semantics do sometimes matter.

[stuman]

ORIGINAL: HansBolter

Thanks for the replies gentelmen.

It should be obvious that the point I was trying to make in my overly dramatic way is that the Assault phase is simply mislabeled.

A basic litmus test for labeling a phase as "Assault" should be whether or not casulaties are being incurred.

The idea that an assault can take place with a guarantee of NO casulaties to the attacker screams out to me that it SHOULDN'T be labeled an assault phase.

It apparently would be more correctly and properly labeled as a Retreat Determination Phase.

Semantics, if you wish, but how things are perceived is often important so the nuances of semantics do sometimes matter.

Valid point.

[treespider]

Ok forever henceforth we shall no longer refer to the "Assault" Phase....

We will simply refer to the "Odds Calculation Determination Phase"... or "OCD" Phase to keep it short.[;)]

[stuman]

ORIGINAL: treespider

Ok forever henceforth we shall no longer refer to the "Assault" Phase....

We will simply refer to the "Odds Calculation Determination Phase"... or "OCD" Phase to keep it short.[;)]

Hmm, OCD, I know that stands for something else. What could it be, what could it be [:D]

[FatR]

ORIGINAL: HansBolter

Thank you for bursting my bubble. [:(]

Let me get this straight.......attackers can NEVER take casualties in the assault phase?

I'll repeat.....attackers can NEVER take casualties in the assault phase?

And? Attack modifies their casualties during the fire phase.

[HansBolter]

ORIGINAL: treespider

Ok forever henceforth we shall no longer refer to the "Assault" Phase....

We will simply refer to the "Odds Calculation Determination Phase"... or "OCD" Phase to keep it short.[;)]

Well, since it's those of us cumpulsively obsessed with getting things right who, more often than not, actually succeed in doing so, I'll take your hopefully good natured ribbing in the spirit of jest I hope it was offered in.

Thanks to both of you for the complement.