Critical Assessment of the Game

[herwin]

Well, I've now played four weeks PBEM, and have a better idea of the strengths and weaknesses of the game. I'm posting this comment in the hopes that it will influence the next game on the same subject. For the most part, the game models what it should model, but what it really gets wrong (and here I write as someone who has architected real-life command and control systems for operations at this level) is the operational tempo. That is, things happen at the wrong rate, sometimes too fast and sometimes too slow. Consider some aspects of this criticism.

The ground combat model is really, really bad. It doesn't allow for mobile operations, mobile defenses, delaying actions or retreats, and the reconstitution of reserves. During a pursuit, units can advance as much as 60 miles a day, four or five times what a foot-mobile force could do in reality. It doesn't model turning movements, which were what the Japanese actually used to unhinge defensive lines.

The occupation model is simply wrong. In reality, the Japanese Army was overstretched by the garrisons it needed to maintain control (and coal production) in Northern China. This drove much of the Army's strategic planning in 1941-42, and was the reason the Japanese delayed their operations in the DEI and the South Seas. The game gives the Japanese player every reason to do things the real commanders knew they lacked the resources to do.

On a positive note, the naval intelligence model is moderately good. It gives a good approximation to the 'pucker factor' that naval commanders feel in real life, although to see that, you need to use WiTPutility to show TF movements over a week or more.

The naval operations model is a bit too high-tempo for air task force operations. The cost of search and CAP is the time spent sailing into the wind to launch and recover aircraft. That can really reduce the combat radius of a task force.

The naval logistics model is simplistic. This results in TFs that can loiter for days, when in reality, they had to keep fueled up for high speed operations. It also does not force them to return to a major base for ammunition resupply. The only Japanese CVs that could strike Pearl without refueling were the Shokaku and Zuikaku, but the game allows the KB to sustain operations east of Pearl. The shoestring nature of that and other operations is missing.

The air operations model is a good try, but misses the real factors that air commanders were concerned with, and produces the wrong tempo of operations. In reality, commanders manage sorties, and they know what they have available at the beginning of the day. Their supplies, fuel, ammo, pilots, and operable planes come together to almost always give them less sorties than they would like to have--although they can usually put together a surge of about 3x the sustainable rate for one to three days--after which the devil has to be paid. This is particularly apparent when IJN attack aircraft redeploy to forward strips, and immediately launch major torpedo attacks--you can't do that in reality without prepositioned supplies, fuel, bombs and torpedoes, rested pilots, and maintenance organisations.

These deficiencies result in a game where both sides move much faster than they could in reality. Good enough for beer and skittles, but ultimately unsatisfying for the professional.

[kilowatts]

Yet for all that it's the best computer wargame I've ever played. And pretty close to the best wargame I've ever played. Have to agree with most of your points, but I feel I can respond to a couple...

The ground combat model is definitely very simplistic. However at this scale the tactical and operational considerations are abstracted. Considerations such as turning movements are reflected in the higher initial experience level of Japanese troops which allows them to be more effective on a man-for-man basis. There's no way to show a 5 mile flanking march in a 60 mile hex.

Not quite sure that your point on the logistics for the carrier TFs is valid past December 1941. Certainly the presentation of the Japanese re-fueling tanker on 12/7 as being full is at odds with reality and allows KB to linger around Pearl longer than was practical. But sustained operations in this area will require that the Japanese commander push a massive amount of supply and fuel forward. IRL this wasn't in the Japanese plan and therefore they didn't plan for it. A game player knowing what he's planning can have that supply moving up on day 1.

Unfortunately a truly accurate simulation of logistics would require the game to track supply stock down to the level of types of munition, food, water etc. That's simply beyond the capability of the AI, or game players, to manage. After all WitP is one of the few wargames that even separates fuel from other supply.

[moses]

Many points are valid and I know you're posting with best intentions towards future improvements[:)]:

I participated years ago, as a bit player, in some of the large computer wargames the US army does to identify future proplems/requirments. 40-50 officers, mostly Cpt-Col, and 15-20 computer modelers and analysts. Guess what??---from what I could see evryone was arguing about the problems with the model.

So I just always marval at the frequent posts by many that seem to expect that GG and company should be expected to have "solved" the Pacific War. It just seems to me to be such an impossible task.

Its a great game that has held my interest for years now. As a soldier I know that there are some things that are just wrong. But I do have just enough programming experience to appreciate the difficulty of the task, and enough research experience to know the enormity of getting everything about the Pacific war right.

So I accept it for what it is. The greatest epic wargame ever.[:D][:D]

[mlees]

"Beer and Skittles"?

[Feinder]

Skittles (chewy fruit candies) - for those who don't like Pretzels.

Actually, I'm more of a chicken-wings man myself. But the sauce would stain the little counters, so I always had to play the Russians because they're always red or brown already... [;)]

-F-

[herwin]

ORIGINAL: kilowatts

Yet for all that it's the best computer wargame I've ever played. And pretty close to the best wargame I've ever played. Have to agree with most of your points, but I feel I can respond to a couple...

The ground combat model is definitely very simplistic. However at this scale the tactical and operational considerations are abstracted. Considerations such as turning movements are reflected in the higher initial experience level of Japanese troops which allows them to be more effective on a man-for-man basis. There's no way to show a 5 mile flanking march in a 60 mile hex.

It's worse than that--controlling a 60-mile front with a company is just imaginary. To prevent infiltration, you would need at least a corps and desirably an army. On the other hand, Japanese turning movements were a problem for the Allies into mid-1944, so experience is not the right way to model it. We know (or at least I know) how to model ground combat operations on that scale, but it would involve reducing the operational tempo to one hex movement every 4-5 days.

Not quite sure that your point on the logistics for the carrier TFs is valid past December 1941. Certainly the presentation of the Japanese re-fueling tanker on 12/7 as being full is at odds with reality and allows KB to linger around Pearl longer than was practical. But sustained operations in this area will require that the Japanese commander push a massive amount of supply and fuel forward. IRL this wasn't in the Japanese plan and therefore they didn't plan for it. A game player knowing what he's planning can have that supply moving up on day 1.

Unfortunately a truly accurate simulation of logistics would require the game to track supply stock down to the level of types of munition, food, water etc. That's simply beyond the capability of the AI, or game players, to manage. After all WitP is one of the few wargames that even separates fuel from other supply.

My concern is less the individual stocks and more the failure to model the constraints imposed by logistics.

The prevalence of unrealistically high tempo operations in the AARs indicates that there is a serious problem here. And what's sauce for the goose is sauce for the gander--the Japanese can expect to lose unrealistically quickly, too.

[herwin]

ORIGINAL: moses

Many points are valid and I know you're posting with best intentions towards future improvements[:)]:

I participated years ago, as a bit player, in some of the large computer wargames the US army does to identify future proplems/requirments. 40-50 officers, mostly Cpt-Col, and 15-20 computer modelers and analysts. Guess what??---from what I could see evryone was arguing about the problems with the model.

So I just always marval at the frequent posts by many that seem to expect that GG and company should be expected to have "solved" the Pacific War. It just seems to me to be such an impossible task.

Its a great game that has held my interest for years now. As a soldier I know that there are some things that are just wrong. But I do have just enough programming experience to appreciate the difficulty of the task, and enough research experience to know the enormity of getting everything about the Pacific war right.

So I accept it for what it is. The greatest epic wargame ever.[:D][:D]

Yes and no. The operational modelling is twenty years out of date, even just considering the open literature.

[Yamato hugger]

In reality the War in the Pacific was a bunch of tactical battles that is nearly impossible to tie together in a strategic campaign. A ground combat system that works in China would be useless on Wake Island and vice versa for example. If you are going to be detailed enough to keep track of individual tanks and infantry squads, then you should also keep track of which supplies are bullets and which are food. The lack of stacking limits, especially in the eastern Pacific lends itself to abuses. How many people can you put on an island before they get in each others way? And yes, the pace of the game is way too fast. How could it be slowed down?

The pace in my humble opinion is due to the fact the Jap player KNOWS what the allies are really capible of and what they are not. He knows the Dutch are near worthless, that the allied air is impotent and he knows WHEN the allies start getting "good" airplanes in suffient numbers. All things that will be a problem in ANY game on the subject. So how exactly do you "fix" that? You cant, simple as that.

[kilowatts]

ORIGINAL: herwin
It's worse than that--controlling a 60-mile front with a company is just imaginary. To prevent infiltration, you would need at least a corps and desirably an army.

That is definitely a very big problem. The way that zones of control, and by inference are handled in the game is definitely inappropriate for the scale.

ORIGINAL: Yamato hugger
The lack of stacking limits, especially in the eastern Pacific lends itself to abuses. How many people can you put on an island before they get in each others way?

As I see it the way to 'enforce' stacking limits is to increase the effectiveness of barrages as more people fill the hex. As it stands right now bombardments seem to inflict less casualties on a bigger unit(s), the reverse should be true. If you added an indicator of the land area in a hex (for the smaller islands) and used the squad density in the bombardment resolution you'd go a long way toward removing the 100,000 man defense of Wake problem.

[herwin]

ORIGINAL: Yamato hugger

In reality the War in the Pacific was a bunch of tactical battles that is nearly impossible to tie together in a strategic campaign. A ground combat system that works in China would be useless on Wake Island and vice versa for example. If you are going to be detailed enough to keep track of individual tanks and infantry squads, then you should also keep track of which supplies are bullets and which are food. The lack of stacking limits, especially in the eastern Pacific lends itself to abuses. How many people can you put on an island before they get in each others way? And yes, the pace of the game is way too fast. How could it be slowed down?

The stacking limits and assault force limits need to be defined on an island-by-island basis. Too large an attacking force merely increases casualties with no increase in combat power.

The pace in my humble opinion is due to the fact the Jap player KNOWS what the allies are really capible of and what they are not. He knows the Dutch are near worthless, that the allied air is impotent and he knows WHEN the allies start getting "good" airplanes in suffient numbers. All things that will be a problem in ANY game on the subject. So how exactly do you "fix" that? You cant, simple as that.

The Japanese knew most of those things in reality, which is why the forces were so well tailored to their missions. There is a maximum pace even when the opposition is nominal. The game is well outside the 95% confidence limits.

[niceguy2005]

This game is absolute perfection.

[Japanese_Spirit]

Considering the other amount of war games out there that do have "problems" (take Hearts of Iron II, for example), I'd rather stick with WitP. Sure, there are problems but then, tell me of one single computer game that has never had any problems? Simple fact, nearly all do.

I respect and appreciate your comments but if the game works in a "realistic" model (which it does, in comparison to others) I have no complaints.

[moses]

"The stacking limits and assault force limits need to be defined on an island-by-island basis. Too large an attacking force merely increases casualties with no increase in combat power. "

This is the type of comment, that while completely accurate, totally underestimates the practical difficulties.

Sure, the designers should have done an evaluation of every hex on the map. How many land hexes are there????

" There is a maximum pace even when the opposition is nominal. "

Another comment that underestimates the difficulties. Movement rates are highly situational. 60 miles per day is very slow for a unit actually loaded onto a train. 15 miles per day on a road is extremly fast for an Army Corps but for a single battalion might actually be slow. The effect of opposition on movement is also situationally dependent. 100 troops along a jungle trail might slow a division to a crawl. A bde of infantry against an armored division in the desert might be nothing more then a speed bump.

Operational pace is even more situational as it depends in large measure on the prior preperation and planning that is barely modeled in the game.

Sure there are planning factors in various staff officer field manuals but these are only guides. The reality is messy, but a game has to have rules that players can understand, that avoid anomalies like ZOC issues, and that curtail gamey actions.

I just don't think its so easy.

[AmiralLaurent]

Herwin, WITP is not a simulation, it is a game.

All of what you say is true, and I guess most of the people here agree with you, but nevertheless WITP is the best PC game I ever had, and I had spent probably thousand hours playing WITP, or on this furm or thinking about it... Does it improve my understanding of WWII or military logistics ? Absolutely not, of course, but anyway there is a real logistic model in WITP, contrary to most games where it is abstract, and even if the game is allowing logistics to run far better than they did, or even that they do now (to go from Saigon to Thailand by car today in December will be slowest that what is is for a fully army in WITP), the logistic dimension is the most important of the game.

In fact I think we have the same interrogations as the real commanders, but the time scale is divided by 2 or 3.

Then, I can't imagine a game that will handle correctly at the same time the politics and huge armies in China, and what happened in the Pacific, where WITP allows you to manage individual barges, PT and so on.

[Nikademus]

ORIGINAL: moses


"The stacking limits and assault force limits need to be defined on an island-by-island basis. Too large an attacking force merely increases casualties with no increase in combat power. "

This is the type of comment, that while completely accurate, totally underestimates the practical difficulties.

Your 100% correct. Stacking limits were discussed by the old design team on more than one occasion. There was no practical way to implement it and the suggestion that each island be defined on a case by case basis was financially impossible. Hard stacking limits would only impose another set of headaches for players.

However a soft or indirect stacking limit factor is possible by reworking the load/unloading routines. 100,000+ troop concentrations on islands are made possible primarily by the ability to offload supply and/or troops with no bottleneck factor in place. PzB and Andy's AAR recently re-highlighted that factor. Initially PzB had insufficient troops for his defense, but a combo of mass air transport and sea transport quickly solved his dilemma and right in the face of his opponent's forces. Thats what makes 100-ship TF's so great...even with port size limits as they stand...100 ships unloading at the same time amounts to alot of troops and supply unloaded....same goes for air transport. 600 transports working every day adds up quickly. Put into place a logistical bottleneck (and perhaps even a hard CAP on supply/fuel levels based on port size) and restrict overbuild capacity (so that the port/airfield values actually mean something) and I think this would greatly improve the feel of the game.

P.S. WitP is an awesome game....warts and all [:D]

[anarchyintheuk]

Lcu stacking limits seem kinda pointless to me. For example, island or atoll invasions, what's the correct ratio of stacking for attacker vs. defender? If the ratio is too small, invasions that happened historically will fail in the game. If too small, they'll be a walkover. Besides, even if you took away a variable like stacking, people would just taking into account another variable, like exp or disruption.

It's all about force allocation. Cramming an island full of lcus just shows you care. [;)]

[anarchyintheuk]

ORIGINAL: Nikademus

However a soft or indirect stacking limit factor is possible by reworking the load/unloading routines. 100,000+ troop concentrations on islands are made possible primarily by the ability to offload supply and/or troops with no bottleneck factor in place. PzB and Andy's AAR recently re-highlighted that factor. Initially PzB had insufficient troops for his defense, but a combo of mass air transport and sea transport quickly solved his dilemma and right in the face of his opponent's forces. Thats what makes 100-ship TF's so great...even with port size limits as they stand...100 ships unloading at the same time amounts to alot of troops and supply unloaded....same goes for air transport. 600 transports working every day adds up quickly. Put into place a logistical bottleneck (and perhaps even a hard CAP on supply/fuel levels based on port size) and restrict overbuild capacity (so that the port/airfield values actually mean something) and I think this would greatly improve the feel of the game.

P.S. WitP is an awesome game....warts and all [:D]

*wonders where Ron is*

[qgaliana]

Some features that could address some land combat limitations, without messing much with the model.
 
Very small units shouldn't project a zone of control - although this could depend on the terrain type. Holding an atoll is easier than holding a city.
 
Other than logistical problems, another reason for limiting stacking should be vulnerability to enemy fire. I don't see much sign that large troop concentrations are vulnerable to bombing or shelling (although I may be wrong). It seems forts are the only consideration. And they're independent of the number of units.

[moses]

"Does it improve my understanding of WWII or military logistics ? Absolutely not."

Don't be so quick. I think I've learned quite a bit. Lots of the learning taking place through the action of arguing on the board and debating game mechanics. The game serves as a basis for discussion which we can compare to the reality. It does not have to be perfect.

[ChezDaJez]

I participated years ago, as a bit player, in some of the large computer wargames the US army does to identify future proplems/requirments. 40-50 officers, mostly Cpt-Col, and 15-20 computer modelers and analysts. Guess what??---from what I could see evryone was arguing about the problems with the model.

You bring up a good point. Moses. We are all able to identify the problems associated with the various models but coming to a consensus on how to fix them would, how would you say, be somewhat "problematic." How many people frequent this forum... 100, 200? Each and every one of us can be guaranteed to have a different opinion of how to fix it.

And like someone said, what will work in China isn't going to work on Wake.

I personally think this is one of the best games that I have ever owned, warts and all. It certainly has made me cut back on purchasing other games the last couple of years!

Chez