Early Blitz

Peter Stauffenberg · Post by **Peter Stauffenberg** » Fri Sep 02, 2011 11:19 am

I think this discussion just shows how hard it is to balance a game. Some people, like Morris, say that the Axis have no chance at all to win this game and his Allied strategy can crush most Axis players.

Others say that France can fall in 1939 or early 1940 if they go all out in the west, thus indication that we need to strengthen the Allies.

Regardless of what we do we will have players who find a strategy to better their chances and might get ahead of the historical progress. That is part of what's fun with the game. You hope to do better than the real commanders by not repeating their mistakes and instead try something else than they did.

Attacking westwards in 1939 is not guaranteed to give better results than a sitzkrieg attack. Let's say you fail to take Brussels in one turn. Then there wil be no surprise of the French forces and you have to bleed your units taking hex by hex in mud or winter weather.

To get maximum firepower against Belgium and Holland you need to neutralize Poland on turn 1 so you can take Warsaw on turn 2 with just some land forces. That means the Luftwaffe is in place in the west for a turn 3 attack on Belgum and Holland.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Fri Sep 02, 2011 11:27 am

Even if you succeed you will only get one extra turn with a good chance for fair weather. After that you get several turns of bad weather.

Let's say you have max luck and take Paris in February. It doesn't mean the British are guaranteed to lose in Sealion. The Allied player has to acknowledge that a blitz is taking out France early and make sure they build some extra land units and send the Canadian reinforcements to England instead of Egypt.

If Sealion comes anyway then the British can delay the Germans there enough to affect the German build-up to Barbarossa.

I think we should be careful about making alternative options so unattractive that people are forced to follow the historical path. Ronnie has collected fall of France stats for quite a bit of games using the latest rules. The conclusion is that the average fall of France time is about the same as before, maybe a turn later.

Even the real Germans thought about attacking the low countries in October 1939, but weather prohibited that. That's the who point. You can take a chance and launch the offensive and hope you get fair weather in October and November. If you get mud you will be stuck with a stronger French resistance and fall time similar to sitzkrieg.

I think it's fair that you can get rewarded by taking chances, but you may be penalized instead if you're unlucky. I think that is fine.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Fri Sep 02, 2011 11:34 am

We can NOT balance a game based upon how players like Morris plays. Such players exploit every little weakness in the game engine and will continue to look for more weaknesses if you try to plug one.

Even worse is that by trying to balance a game to extreme play you may actually unbalance the game to normal play. So we have to tell ourselves WHO are we making the game for. The answer is that we make the game for pretty experienced players who like to play normally. If the game is balanced for the majority of players who do normal things with the game then I'm happy.

We can't get the game balanced for novices because if you don't use the strength of your country then you will fall behind. E. g. it's very hard for a novice Axis player to win because you don't manage to take France early enough or get deep into Russia in 1941. But novice players learn from their mistakes and will eventually become experienced enough to feel that the game is balanced.

Normal players have little chance against players who have analyzed the game to death and tried out lots of extreme play styles (different blob strategies etc.). They might have failed with many strategies before they found one that worked. Then this strategy can be improved and be a killer against normal players. E. g. Morris has found a way to sacrifice the British to get the upper hand with the Russians later. Then the Germans are crushed like Plaid's Axis forces in Russia.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Fri Sep 02, 2011 11:44 am

The only way to beat such a strategy is to first learn about it and then try to counter it with different strategies. You will need many tries until you find something that works.

So you might end up with having several weird strategies that are more effective than normal ones. If you meet such a player you get crushed. These strategies are often not historical at all and hard to discover, but the game engine allows it so it's used.

We as GS v2.0 developers can't anticipate all of these strategies and have a counter against them. Trying to plug the holes is frustrating because new ones pop up all the time and you're back to square one. Even worse is that you complicate the game with special rules for no reason.

I think the BEST "medicine" against extreme play styles is to actually not play against these players and instead pick opponents you know will play pretty normal games. It's fine to have some surprises against your opponent like rejecting the French armistice offer or invading Spain or Canada. That's about strategy and not about exploiting the game engine.

Most players don't have the time and energy to playtest exploits enough to find what works and what doesn't. Therefore you don't have time to find the counter strategies either.

I want playing to be FUN and not about who has spent most time in the analyzing room. I don'twant to feel that I'm in a chess match for the world championship where you know your opponent and his team has spent thousands of hours analyzing your playing style and games to find weaknesses they can exploit. You know you don't have a chance to win the match unless you spend an equal amount of time in the analyzing room. That's a choir and not fun.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Fri Sep 02, 2011 11:53 am

The conclusion is that I think we should just give up trying to balance the game against playing styles performed by Morris and instead try to make sure we don't imbalance what's already working.

So if some elite players manage to make a 1 turn conquest of Poland regularly and be in Paris in January then I suggest we just let these players be with their strategy. It's not a problem until regular players can replicate the strategies and end up with the same results.

E. g. I've tried to replicate what players like Morris and Supermax do and always fail to get the same results. I therefore believe that's not only about finding an effective strategy, but also about HOW you implement it on the battleground. I simply have to accept that players like Morris, Supermax, Neil and Joerock are better than me to get results with the resources they have available. That's true regardless of the strategies they choose.

So we should not pull the emergency break by analyzing games done by elite players. They manage to get results the rest of us can only dream about. My suggestion is to just stop tweaking the game balance by looking at such games and instead get out GS v2.1 to the public if we feel that us "normal" players can get balanced results.

E. g. I'm playing the Axis against Pionurpo, Jim, Ronnie and Duncan and my progress to the end of 1941 is almost the same in all games. They play differently and I get to the historical line and survive the winter pretty well. That's encouraging because it shows that with normal play GS v2.0 seems pretty balanced at least till 1942.

Schnurri · Post by **Schnurri** » Fri Sep 02, 2011 11:42 pm

Did some testing on 1 turn fall of Warsaw.

34 tests yielded 1 successful fall of Warsaw. 11 times (of the 34) I couldn't bring 2 ARM to bear on the capital. If you can bring 2 ARM and 1 TAC and 1 FTR to bear the best results you can get are 11 steps - 8 from the 2 ARM and 2 from the TAC and 1 from the FTR. It is pretty rare to get at least three units to max out in order to take the capital. Statistics aren't meaningful here as I was only successful once and if I got a successful attack on the 35th attempt they would change dramatically but I think we are in the range of RKR of 5-7% max. Unless Richard has some other way but the max you can get is the 2 ARM, TAC and FTR. Really isn't a problem that needs to be addressed as it is rare enough and forces the Axis to use all their troops. Only a problem when combined with the weak low countries combined with fortuitous good weather.

rkr1958 · Post by **rkr1958** » Sat Sep 03, 2011 12:10 am

Schnurri wrote:Did some testing on 1 turn fall of Warsaw.

34 tests yielded 1 successful fall of Warsaw. 11 times (of the 34) I couldn't bring 2 ARM to bear on the capital. If you can bring 2 ARM and 1 TAC and 1 FTR to bear the best results you can get are 11 steps - 8 from the 2 ARM and 2 from the TAC and 1 from the FTR. It is pretty rare to get at least three units to max out in order to take the capital. Statistics aren't meaningful here as I was only successful once and if I got a successful attack on the 35th attempt they would change dramatically but I think we are in the range of RKR of 5-7% max. Unless Richard has some other way but the max you can get is the 2 ARM, TAC and FTR. Really isn't a problem that needs to be addressed as it is rare enough and forces the Axis to use all their troops. Only a problem when combined with the weak low countries combined with fortuitous good weather.

Statistics aren't meaningful here as I was only successful once and if I got a successful attack on the 35th attempt they would change dramatically but I think we are in the range of RKR of 5-7% max.

Actually they are ... my Master's is in Operation's Research with a focus on statistics and I do love statistics. Without going through the gorey details if we assume that each trial was independent and that the metric we're measuring is binomial (i.e., Warsaw falls or it doesn't) then we can get bounds on your results. For enough trials (15+) the estimate of p, or p(estimate), which is the probability that Poland falls in 1-turn, is a normal random variable with expected value equal to the true p(true) and a standard deviation equal to sqrt[(p(true)*(1-p(true))/N)], where N equals the number of samples used to derive p(estimate). Of course we don't know p(true) but we can use p(estimate) to get the sigma on our estimate, which is what I've done below.

You got 1 success out of 35 trials is p(estimate) = .03 (i.e., 1/35). The +/- 2-sigma (or 95%) confidence interval around that estimate is
0% <= p < 6%.

I basically got 1 success out of 15 trails, which gave a p(estimate) = .07, with a 95% confidence interval of 0% <= p < 14%

If we combine the data from our two tests then p(estimate) = 2/50 = .04, with a 95% confidence interval of 0 <= p < 7%.

If we ran the tests correctly (i.e., used the optimum) sequence I think we do have strong statistical results that tell us the chance to knock on Poland in 1-turn is no better than 6 to 7%.

By way, when I was in graduate school over 30-years ago we had a saying about people who use statistics, which was "There are two types of people who use statistics. Liars and damn Liars."

Schnurri · Post by **Schnurri** » Sat Sep 03, 2011 12:14 am

Isn't that what I said without knowing why?

If we subtract the 11 tries where I couldn't get 2 ARM on Warsaw (due perhaps to faulty play) then we have 23 tries with 1 success and it doesn't matter how good you are, the best you can get is 2 ARM, 1 TAC and 1 FTR.

rkr1958 · Post by **rkr1958** » Sat Sep 03, 2011 12:15 am

Schnurri wrote:Isn't that what I said without knowing why?

If we subtract the 11 tries where I couldn't get 2 ARM on Warsaw (due perhaps to faulty play) then we have 23 tries with 1 success and it doesn't matter how good you are, the best you can get is 2 ARM, 1 TAC and 1 FTR.

Yes, and you said much more succinctly that I did. That seems to happen to me a lot for some reason.

richardsd · Post by **richardsd** » Sat Sep 03, 2011 3:03 am

are you guy's using the sequence as per Supermax's AAR?

As far as I know it is only a TAC, FTR and two ARM that you can get into contact with Warsaw, from memory that gives standard projections of

TAC 2-0
FTR 0-0
ARM 4-1
ARM 4-0

I don't think I have succeeded if the TAC doesn't get the two steps, but in my last game I took Warsaw because the first ARM did 5 steps! (can't remember ever seeing that before)

I have also known the TAC do 2, the FTR 1 and the ARM's have failed to get seven between them

I think the weather plays a big part, in the game where I have just taken Warsaw in one turn, the next turn was mud giving the effect of a two turn conquest, but then it was fair all the way to winter and no more mud, Paris will likely fall February.

rkr1958 · Post by **rkr1958** » Sat Sep 03, 2011 3:12 am

richardsd wrote:I think the weather plays a big part, in the game where I have just taken Warsaw in one turn, the next turn was mud giving the effect of a two turn conquest

I'm not following you. The second turn of the game is September 21, 1939 and you're guaranteed fair weather on all fronts.

Schnurri · Post by **Schnurri** » Sat Sep 03, 2011 4:25 am

I think it is the same sequence, but regardless of the prelim you can only get the 2 ARM 1 TAC and 1 FTR. Their theoretical max steps is 11 (although the FTR says it won't take any steps it often takes 1). To be successful you must have 3 of 4 units max out their attacks. Given the latest rules I think it is worth the gamble to try for the one turn fall of Warsaw as, even if it fails, you aren't in much worse position than my standard move which involved no attacks on Warsaw but decimated the other land forces and sent 1 MECH to the western front. In 2.0 it made more sense to do a 2 move assault on Poland and take out Holland on the third turn but with the new rules, unless the weather gods are very generous, you need to wait to build sufficient forces to take out the low countries in 40. Maybe the solution to putting the Allies in such a fix with a one turn fall of Poland and a 3rd move take out of Belgium is to have a stiffer penalty for rail (less rail points) in 39?

richardsd · Post by **richardsd** » Sat Sep 03, 2011 7:24 am

so it must have been mud the turn after railing troops to the west, much worse!

either way I lost a turn due to mud to balance the 1 turn conquest

what I am really saying is that its not how long it takes for Poland - but what weather you get that counts (I think this was schnurri's point as well)

Morris · Post by **Morris** » Mon Sep 05, 2011 2:41 pm

Stauffenberg wrote:The conclusion is that I think we should just give up trying to balance the game against playing styles performed by Morris and instead try to make sure we don't imbalance what's already working.

So if some elite players manage to make a 1 turn conquest of Poland regularly and be in Paris in January then I suggest we just let these players be with their strategy. It's not a problem until regular players can replicate the strategies and end up with the same results.

E. g. I've tried to replicate what players like Morris and Supermax do and always fail to get the same results. I therefore believe that's not only about finding an effective strategy, but also about HOW you implement it on the battleground. I simply have to accept that players like Morris, Supermax, Neil and Joerock are better than me to get results with the resources they have available. That's true regardless of the strategies they choose.

So we should not pull the emergency break by analyzing games done by elite players. They manage to get results the rest of us can only dream about. My suggestion is to just stop tweaking the game balance by looking at such games and instead get out GS v2.1 to the public if we feel that us "normal" players can get balanced results.

E. g. I'm playing the Axis against Pionurpo, Jim, Ronnie and Duncan and my progress to the end of 1941 is almost the same in all games. They play differently and I get to the historical line and survive the winter pretty well. That's encouraging because it shows that with normal play GS v2.0 seems pretty balanced at least till 1942.

I do not agree with your conclusion ! I just discovered the weakness of the game engine which after my AAR , anyone may will find it & use my way to defeat their oppoent . Maybe I am the first one who found it , but I am sure I will be not the only one who will use it . Anyplayer who find a way to win the game ,he will learn & use it no matter it is normal play or not .

Actually , the armor bloc is not the weakness of the game since it is very difficult to run it well , but you had done a lot changes to against it ( although it is still undefeatable before the production limited patch !). But the present problem which will show in the three AAR is really the weakness of the game . If everyone learn my way to defeat Axis , no one will play Axis in the future ! then the game will fail .

Blathergut · Post by **Blathergut** » Mon Sep 05, 2011 2:42 pm

Morris...I don't suppose you could decsribe the strategy in detail here? I think many are waiting on the AARs but they may take quite a while.

Morris · Post by **Morris** » Mon Sep 05, 2011 3:05 pm

Blathergut wrote:Morris...I don't suppose you could decsribe the strategy in detail here? I think many are waiting on the AARs but they may take quite a while.

fact is the best way to show . I have tried to argue about this for several times ,I failed in the discussion because of my limited English & time . I really do not have enough time on the forum to do the job . I am a business man , I love & enjoy the game , but this is not my job ! Maybe there is different culture from some others . In China , we'd always like to play game with the one better than me sothat we can learn & improve , If I can meet an elite who defeat me , I won't lose interest of the game , It will be great hornor to have opportunity to play with the elite who taught me a lesson . If anyone lose the interest of a game after defeated by another,He is not good player , He is a loser !

zechi · Post by **zechi** » Mon Sep 05, 2011 3:40 pm

@Morris
I think you should take your time to explain your strategy in detail. Firstly, it could help to better understand your point and the balance problems you claim exist. Secondly, I do not think that the current AARs will prove anything, if you do not share your point of view as well (through an AAR or an explanation). For example from your game with Plaid I do not think that we could really learn anything as he gave up prematurely and got some bad results (like losing two ARMs in France) as well.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Mon Sep 05, 2011 3:55 pm

Morris wrote:
Blathergut wrote:Morris...I don't suppose you could decsribe the strategy in detail here? I think many are waiting on the AARs but they may take quite a while.
fact is the best way to show . I have tried to argue about this for several times ,I failed in the discussion because of my limited English & time . I really do not have enough time on the forum to do the job . I am a business man , I love & enjoy the game , but this is not my job ! Maybe there is different culture from some others . In China , we'd always like to play game with the one better than me sothat we can learn & improve , If I can meet an elite who defeat me , I won't lose interest of the game , It will be great hornor to have opportunity to play with the elite who taught me a lesson . If anyone lose the interest of a game after defeated by another,He is not good player , He is a loser !

I don't understand that argument at all. It takes you maybe 30 minutes to write down the essence of your strategy, maybe even less if you explain only the major parts. Playing the game against another player actually takes numerous hours of your time.

Even more important is that by not explaining why you feel the Axis have no chance then we will need many weeks in real time before we even get to the point in the game where we see the problem. That means we can't fix the problem until them. Then we have to delay the GS v2.1 to the public or simply release without fixing your suggested game balance issue.

Morris · Post by **Morris** » Tue Sep 06, 2011 6:21 am

Stauffenberg wrote:
Morris wrote:
Blathergut wrote:Morris...I don't suppose you could decsribe the strategy in detail here? I think many are waiting on the AARs but they may take quite a while.
fact is the best way to show . I have tried to argue about this for several times ,I failed in the discussion because of my limited English & time . I really do not have enough time on the forum to do the job . I am a business man , I love & enjoy the game , but this is not my job ! Maybe there is different culture from some others . In China , we'd always like to play game with the one better than me sothat we can learn & improve , If I can meet an elite who defeat me , I won't lose interest of the game , It will be great hornor to have opportunity to play with the elite who taught me a lesson . If anyone lose the interest of a game after defeated by another,He is not good player , He is a loser !
I don't understand that argument at all. It takes you maybe 30 minutes to write down the essence of your strategy, maybe even less if you explain only the major parts. Playing the game against another player actually takes numerous hours of your time.

Even more important is that by not explaining why you feel the Axis have no chance then we will need many weeks in real time before we even get to the point in the game where we see the problem. That means we can't fix the problem until them. Then we have to delay the GS v2.1 to the public or simply release without fixing your suggested game balance issue.

Hi Borger : It will not only take 30 minutes for me to explain all details of my point . I donot agree that nothing prove from the AAR between Plaid & me . even if Plaid add two more panzer ,it still could not change the situation there .
But there is one opportunity coming : just at this turn (July12th 1941) with Supermax , he launched a smart landing in North Africa , & it will probably cause the final
failure of UK's attempt in Africa to help Russia .

So if Supermax gain this advantage ,maybe he will show you a prove that Axis has a chance to defeat Allies on the base of the present 2.01 ; but if Supermax unfortunely lose the game by carrying such great achivement in 1941 ( ocuppy England &Scotland ,whole Vichy France，& maybe plus Giberoto & Spain ), Will you accept that the game is unbalance at present ?

zechi · Post by **zechi** » Tue Sep 06, 2011 7:40 am

Morris wrote: Hi Borger : It will not only take 30 minutes for me to explain all details of my point .

Even if it takes a little bit longer, it would be really helpful if you would explain your strategy in detail. You are part of the Beta-Testing group, i.e. you should give your best to improve the game.

I donot agree that nothing prove from the AAR between Plaid & me . even if Plaid add two more panzer ,it still could not change the situation there .

Perhaps, but you can’t know it for sure. Furthermore, there are other things in the game which were greatly in your favour, like the invasion of Sicily. You got lucky that you captured Palermo so easily and quickly, even without any supply source. Would have Plaid repelled the invasion, then the game would have taken another direction. Plaid also did not play perfectly when he switched to the defence in the east (placing wrong units in the first line) and he could have defended Sicily much better.

So if Supermax gain this advantage ,maybe he will show you a prove that Axis has a chance to defeat Allies on the base of the present 2.01 ; but if Supermax unfortunely lose the game by carrying such great achivement in 1941 ( ocuppy England &Scotland ,whole Vichy France，& maybe plus Giberoto & Spain ), Will you accept that the game is unbalance at present ?

From my point of view this will prove nothing at all, neither if Supermax wins nor if you lose. Supermax is an extreme player who takes extreme risks, so his playing style is not very representative (but very entertaining, interesting and full of surprises).