The art of "averaging out"
Posted: Fri Aug 26, 2011 6:43 am
I have seen a lot of posts again lately, even by people who should know better,
who take historical data as a measure for ingame results and question those results
based on exceptional real life single events.
Historical data is only useful if you take everything with influence on the events into account.
If you just look at a historical map snippet, at the the military hardware present in this snippet,
and the extreme battle result on that snippet, you will never have data to start a discussion on in
game mechanics encompassing a much broader spectrum of action-reaction variables.
The two examples that stood out lately were:
- The Midway discussion
- The Pearl Harbour damage discussion
I´d like to evaluate shortly why one has to be careful when assessing
a single historical battle as representative:
Midway:
Midway did not happen on a 500x500nm square, it did not start on July 4th, and the outcome is not
deducable only from the forces present at the day the battle took place.
The battle of Midway started over half a year before events unfolded.
From a Japanese perspective it was a compromise decision to satisfy warring high commands, the
strategic goal and the reason for the operation were vagualy outlined at best, many compromises
weakened the forces available for conduction in advance, and the operation itself relied
on unrealistic prerequisites and was doomed to failure as soon as those prerequisites were not met.
And this is just the beginning.
Based on the above there were a lot of decicions made on a tactical level which improved the
situation for the USN and slowly but continuousely worsened the position of the IJN.
As anybody else who has read it (though I disagree with some of Parshall&Tullys conclusions) I can
reccomend Shattered Sword as a very in deep read about how Japane stumbled into disaster.
Pearl Harbour:
Pearl Harbour did not happen on a 550x500nm square as well. But here the explanation is a bit
different to why it is not representative.
Since Scen1/2 starts at the night before the war, the setup is pretty accurate in terms of
firepower and decisions made leading to the result.
But the behaviour of the US on the day of the attack was relaxed to the point of willful negligence.
The reason for this is understandable, noone expected a large scale carrier raid because
the general belief still held that the queen of the seas was the BB.
Looking at "Battleship Row" the position of the ships begged for the most disastreous result
possible. It was an invitation to cause maximum damage (combined with many of the other factors
working against the Americans). Look at the pic below. Doesn´t it say "please sink me"?
Any change in the positioning of the BBs may have resulted in an improvement, any small decision
to reduce the exposure against air attack would have had the potential to lead to less gutting
losses.
WitPAE does not simulate "position of ships at anchor" individually. It abstracts the position
by adding randomness, which partly can be interpreted as different outcome because of different
target layout.
Midway is on the extreme upper scale of how an offensive battle can go wrong.
These things happen in game. Just not when you expect them.
Pearl Harbour is at the slight upper scale of how an offensive battle can go right.
These things also happen in game. Just not when you expect them.
What I want to show with the above is:
Yes, AE tries to represent and even replicate events in WWII, on every scale from grand logistical
down to single pilot, inf squad, and individual ship crew.
But it does not recreate history. This is a misunderstanding. It provides you with similar capabilities
and sets up a similar initial situation and general option for the parties involved.
It does not replicate every out of average battle result wittnessed in WWII. And it shouldnt.
Be aware that it does not suffice to estimate an averaged out game result. You have to average
out the real life event as well. Only then you can begin to compare the result and come to a conclusion
about the ability of the game to recreate situations on the scale you are interested in.
My personal conclusion when looking at the specific events and comparing to the historical battle is:
In game, on a small scale, with the same setup, Midway should be quite unlikely to end
as a decisive victory for the USN, PH should be a bit less likely to be as successful
for Japan as both events were historical.
who take historical data as a measure for ingame results and question those results
based on exceptional real life single events.
Historical data is only useful if you take everything with influence on the events into account.
If you just look at a historical map snippet, at the the military hardware present in this snippet,
and the extreme battle result on that snippet, you will never have data to start a discussion on in
game mechanics encompassing a much broader spectrum of action-reaction variables.
The two examples that stood out lately were:
- The Midway discussion
- The Pearl Harbour damage discussion
I´d like to evaluate shortly why one has to be careful when assessing
a single historical battle as representative:
Midway:
Midway did not happen on a 500x500nm square, it did not start on July 4th, and the outcome is not
deducable only from the forces present at the day the battle took place.
The battle of Midway started over half a year before events unfolded.
From a Japanese perspective it was a compromise decision to satisfy warring high commands, the
strategic goal and the reason for the operation were vagualy outlined at best, many compromises
weakened the forces available for conduction in advance, and the operation itself relied
on unrealistic prerequisites and was doomed to failure as soon as those prerequisites were not met.
And this is just the beginning.
Based on the above there were a lot of decicions made on a tactical level which improved the
situation for the USN and slowly but continuousely worsened the position of the IJN.
As anybody else who has read it (though I disagree with some of Parshall&Tullys conclusions) I can
reccomend Shattered Sword as a very in deep read about how Japane stumbled into disaster.
Pearl Harbour:
Pearl Harbour did not happen on a 550x500nm square as well. But here the explanation is a bit
different to why it is not representative.
Since Scen1/2 starts at the night before the war, the setup is pretty accurate in terms of
firepower and decisions made leading to the result.
But the behaviour of the US on the day of the attack was relaxed to the point of willful negligence.
The reason for this is understandable, noone expected a large scale carrier raid because
the general belief still held that the queen of the seas was the BB.
Looking at "Battleship Row" the position of the ships begged for the most disastreous result
possible. It was an invitation to cause maximum damage (combined with many of the other factors
working against the Americans). Look at the pic below. Doesn´t it say "please sink me"?
Any change in the positioning of the BBs may have resulted in an improvement, any small decision
to reduce the exposure against air attack would have had the potential to lead to less gutting
losses.
WitPAE does not simulate "position of ships at anchor" individually. It abstracts the position
by adding randomness, which partly can be interpreted as different outcome because of different
target layout.
Midway is on the extreme upper scale of how an offensive battle can go wrong.
These things happen in game. Just not when you expect them.
Pearl Harbour is at the slight upper scale of how an offensive battle can go right.
These things also happen in game. Just not when you expect them.
What I want to show with the above is:
Yes, AE tries to represent and even replicate events in WWII, on every scale from grand logistical
down to single pilot, inf squad, and individual ship crew.
But it does not recreate history. This is a misunderstanding. It provides you with similar capabilities
and sets up a similar initial situation and general option for the parties involved.
It does not replicate every out of average battle result wittnessed in WWII. And it shouldnt.
Be aware that it does not suffice to estimate an averaged out game result. You have to average
out the real life event as well. Only then you can begin to compare the result and come to a conclusion
about the ability of the game to recreate situations on the scale you are interested in.
My personal conclusion when looking at the specific events and comparing to the historical battle is:
In game, on a small scale, with the same setup, Midway should be quite unlikely to end
as a decisive victory for the USN, PH should be a bit less likely to be as successful
for Japan as both events were historical.
