Play Balance in China

[Dr Deo]

ORIGINAL: Red Prince

It's been a long time since I took statistical analysis. Perhaps I need a refresher. [:)]

Wikipedia helps... it did for me! [;)]

The reason why the math is so simple is that each roll is independent, so the total mean is just the sum of the expected values of each roll.

[Red Prince]

ORIGINAL: Dr Deo

ORIGINAL: Red Prince

It's been a long time since I took statistical analysis. Perhaps I need a refresher. [:)]

Wikipedia helps... it did for me! [;)]

The reason why the math is so simple is that each roll is independent, so the total mean is just the sum of the expected values of each roll.

I'm trying to build an actual model using some program I just pulled off the net. Frankly, it would be easier and faster just to run the numbers in a spreadsheet to get the distribution, which is what I'm most interested in. Just the basics. I'll get back to you when I've run this correctly. I didn't intend to put a huge amount of time into this. I'll work on it some more later.

Thinking about it, I don't see why a Monte Carlo would be any better than pure number-crunching. The only uncertainty is in how many cities more than the minimum required to conquer China. All the chances are already defined: 20%, 30%, or 40% . . . depending on the cities chosen, and since all of the factory cities have to be taken, this creates no need for a range beyond total cities required.

At this point I'm not looking to find the "expected" value. Only the probability breakdowns.

[paulderynck]

You don't need a program or simulation. It is a standard Binomial distribution. Define a success as a chit generated and you can produce the set of probabilities versus number of chits from an excel spreadsheet in about ten minutes.

[Shannon V. OKeets]

ORIGINAL: Red Prince

ORIGINAL: Dr Deo

ORIGINAL: Red Prince

It's been a long time since I took statistical analysis. Perhaps I need a refresher. [:)]

Wikipedia helps... it did for me! [;)]

The reason why the math is so simple is that each roll is independent, so the total mean is just the sum of the expected values of each roll.

I'm trying to build an actual model using some program I just pulled off the net. Frankly, it would be easier and faster just to run the numbers in a spreadsheet to get the distribution, which is what I'm most interested in. Just the basics. I'll get back to you when I've run this correctly. I didn't intend to put a huge amount of time into this. I'll work on it some more later.

Thinking about it, I don't see why a Monte Carlo would be any better than pure number-crunching. The only uncertainty is in how many cities more than the minimum required to conquer China. All the chances are already defined: 20%, 30%, or 40% . . . depending on the cities chosen, and since all of the factory cities have to be taken, this creates no need for a range beyond total cities required.

At this point I'm not looking to find the "expected" value. Only the probability breakdowns.

I taught college level statistics. I also have decades of experience with Monte Carlo simulation. Take it from me, Dr. Deo is correct in his posts. Paul is also very knowledgable about statistics and Monte Carlo simulations (and Nils knows theoretical statistics enough for this 'problem' should he add his opinion).

[brian brian]

The issue with additional cities in China isn't the US Entry hits. For me at least, putting additional cities in China just continues to make the Chinese into the biggest 'fantasy' part of the game, far more than Italy. It doesn't affect play balance, to me.....WiF is the most perfectly balanced game ever with the bidding system for victory conditions. If you feel that the Axis is completely doomed and can't even hold a single victory city at the end of 36 turns, you could always bid how many turns it takes for the Allies to dismantle them completely and you still have a great WWII grand strategy game.

Anyway, China needs a large re-write in the game. In the interests of having the possibility of there being a Chinese player, the Chinese are outfitted with all of the fun pieces to build that everyone else gets, but are freed from the many historical & realism constraints that make many of China's achievements in many games more than a bit of a farce all too often. In the interests of keeping China from being conquered too easily in the unified global simulation of the rest of the game, their economy is jumped up too far (again more than Italy) and then they get to build all kinds of fantastical units that the real China could quite simply never ever have produced.

To give the Chinese additional cities is to give them additional bases and logistical capabilities; a city in WiF is a very powerful construct. The real Chinese were hobbled by the dominant fact that the Japanese were doomed to lose the war once they poked the USA into the war, and everyone is China was merely passing time until the real war started ... the Chinese Civil War. WiF players will never play the game the way Chiang made his decisions. No matter how much FDR poured over The Hump, his fantasy of a bottomless pool of Chinese manpower atttritioning down the Japanese to save American lives was just that, a fantasy, which was quickly shattered in 1944 when the Americans finally began using China as a base to bomb Japan and the Japanese promptly overran all of the bases despite four+ years of Allied military aid and training of various sorts. The Japanese were hobbled by their own very poor logistics system and their inability to well secure the parts of China they did conquer, due to their racism. WiF makes the situation in China into an OK game, but not a very good simulation, and there is a whole lot of room to improve the simulation of WWII in China in future editions of the game.

I am glad that the additional cities are optional, that is very good. I don't think playing without them will lead to an easy conquest of China at all as long as the Chinese 'player' has half a brain.....Chungking is a long, long way from the Japanese start lines. I would argue against the additional cities to help keep Japan in the game all the way to J/A 45, as the new map magnifies their U-boats-or-Barbarossa problem of dealing with a simultaneous Chinese land offensive and the USMC leaving port for a major invasion due to the action limits system. Giving the Chinese all those additional bases will just lead to more game misadventures such as Stilwell leading the Chinese Marines ashore in Japan and other such WiF only fantasies.

[lomyrin]

My past experiences with CWiF had China conquered or very severely mauled by Japan in most games. That is likely to be true in MWiF as well if the optional Chinese cities are not in use. WIth the extra cities, and modifed US entry rolls for entering the cities, the Chinese have been holding out better in my trials and the Japanese are likely to have more problems in the mid and late game than in either CWiF or the boardgame.

Lars

[Centuur]

ORIGINAL: lomyrin

My past experiences with CWiF had China conquered or very severely mauled by Japan in most games. That is likely to be true in MWiF as well if the optional Chinese cities are not in use. WIth the extra cities, and modifed US entry rolls for entering the cities, the Chinese have been holding out better in my trials and the Japanese are likely to have more problems in the mid and late game than in either CWiF or the boardgame.

Lars

If there are more problems for the Japanese than in the boardgame, then IMHO there is a problem. Japan can't have a more difficult job in China due to the addition of these cities, if you compare it with the boardgame, since that means that Japan will be in a more difficult position when the war with the Wallies starts...
If this is the result of play testing, than the extra Chinese cities balance the game to much towards the Allied side. That is not the way it should be.
The extra cities should compensate the unified map (since CWIF made conquest of China to easy) to get the balance back, not to unbalance the game towards the allied site...
If this is happening, than some of those extra Chinese cities should be erased from the map...

[lomyrin]

Keep in mind that the extra Chinese cities are an option that can be turned on or off when you decide on the game options to use.
When the option is off those cities are not there.

Lars

[Centuur]

ORIGINAL: lomyrin

Keep in mind that the extra Chinese cities are an option that can be turned on or off when you decide on the game options to use.
When the option is off those cities are not there.

Lars

But if that means a choice of two bad things, it isn't good. The conquest of China should be just as hard in the boardgame as in MWIF. If no cities means it is easier and with the cities it is harder than on the board (or gives the Chinese to much opportunities to gain terrain in the later years of the war), there is an imbalance eitherway. IMHO the extra cities should balance play back to the boardgame, not making things difficult for any side...

[lomyrin]

I suppose only a lot of games played after MWiF is released will reflect the actual balance the players encounter in complete games.

Lars

[paulderynck]

ORIGINAL: lomyrin

I suppose only a lot of games played after MWiF is released will reflect the actual balance the players encounter in complete games.

Lars

It was a little more interesting to calculate with the realization that the chances for a chit would be 40% for Chungking, 30% each for the other 6 factory cities and 20% each for the added ten cities that are assumed to require conquest. Here's the results. The "19 Cities" plot would be MWiF with the added 10 cities taken and the "9 Cities" plot is WiFFE.

It is consoling to see the probability of the allies getting 19 chits out of it is: 1.1944 times 10 to the minus 12.

[Centuur]

The balance of the US entry can be calculated, that isn't my concern here. The boardgame has a delicate balance concerning the Chinese front. This is reflected by two things. The first are the early war years, where Japan could try to conquer China, or make it so small, it isn't going to be a concern for the rest of the war. The second are the late war years, when China should be able to grab some territories back and perhaps even drive the Japanese out of Manchuria, if he is very lucky or the Japanese didn't push the Chinese to hard in the early war years.
The CWIF map has, according to the players here, the disadvantage of making the defence of China very difficult, if not impossible due to supply problems and the large gaps in the frontlines...
The MWIF map now has, according to one here, the disadvantage of making it to easy for the Chinese to punish the Japanese in the later war years, due to more supply sources for the Chinese troops and the large gaps in the frontline... Don't forget that the Chinese will take only land impulses, which gives the opportunity to tactically exploit the gaps in the Japanese lines (since Japan has to take a lot of naval impulses against Uncle Sam...). Gaps are almost never there on the boardgame map, so there is less need for the Japanese to move land units in the boardgame.
Question is: how many playtesting has been done, using the new cities? Is this enough to come to the conclusion that the cities are doing a proper job, not only in making it more difficult for the Japanese to conquer China, compared to CWIF but also not giving the Chinese to much opportunities in the later war to crush the Japanese to early, compared to the boardgame?
If the conclusion is that there is a problem with the new cities in the late war years, shouldn't it be fixed somehow before MWIF goes on the market?

[Shannon V. OKeets]

ORIGINAL: Centuur

The balance of the US entry can be calculated, that isn't my concern here. The boardgame has a delicate balance concerning the Chinese front. This is reflected by two things. The first are the early war years, where Japan could try to conquer China, or make it so small, it isn't going to be a concern for the rest of the war. The second are the late war years, when China should be able to grab some territories back and perhaps even drive the Japanese out of Manchuria, if he is very lucky or the Japanese didn't push the Chinese to hard in the early war years.
The CWIF map has, according to the players here, the disadvantage of making the defence of China very difficult, if not impossible due to supply problems and the large gaps in the frontlines...
The MWIF map now has, according to one here, the disadvantage of making it to easy for the Chinese to punish the Japanese in the later war years, due to more supply sources for the Chinese troops and the large gaps in the frontline... Don't forget that the Chinese will take only land impulses, which gives the opportunity to tactically exploit the gaps in the Japanese lines (since Japan has to take a lot of naval impulses against Uncle Sam...). Gaps are almost never there on the boardgame map, so there is less need for the Japanese to move land units in the boardgame.
Question is: how many playtesting has been done, using the new cities? Is this enough to come to the conclusion that the cities are doing a proper job, not only in making it more difficult for the Japanese to conquer China, compared to CWIF but also not giving the Chinese to much opportunities in the later war to crush the Japanese to early, compared to the boardgame?
If the conclusion is that there is a problem with the new cities in the late war years, shouldn't it be fixed somehow before MWIF goes on the market?

No. In the lifetime of WIF, the designers have always made an honest effort to gets things balanced correctly but inevitably have had to make changes later to "fine tune" game play.

There are too many interacting systems and subsystems to hope to exhaustively test all cases. Even trying to cover all the primary cases is beyond feasible. Who declares war on whom when, who builds what units when, which optional rules are in effect, and how the die rolls go on certain critical attacks are just a few of the variables with which we have all seen have dramatic effects on the tide of war.

Indeed, no one wants to know in advance that doing A-B-C will produce result X. That is what playing the game is all about.

So, let me phrase your concern as being that you do not want doing A-B-C to always produce result X. Rather, there should be sufficient alternatives in the opponent's tool kit to maybe cause result Y, when you do A-B-C. What is presently unknown is whether there are enough alternative strategies and tactics available to both the Chinese and the Japanese when playing MWIF. Ideally, either side should be able to find responses to counter their opponent's game play (and luck).

We'll really only know that after hundreds of games have been played to completion by a wide variety of players using different rules, and strategies, and tactics, and die rolls, and, ...

But I can say we have made a serious effort to play balance the war in China and the Pacific for MWIF. So far we haven't found any obvious egregious flaws. They might be there[&:], they might not be there.[&:] More than that I don't think is possible. In the short term, the numerous optional rules let players 'adjust' the play balance rather significantly.