Grand Strategy Victory Stats and Summary

pk867 · Post by **pk867** » Tue Mar 09, 2010 10:43 am

Well all I can say if a player reports a win then that means there was a corresponding loss that should be
put in the stats. You do not need to know that players name. You can put it under anonymous .

rkr1958 · Post by **rkr1958** » Tue Mar 09, 2010 10:51 am

pk867 wrote:Well all I can say if a player reports a win then that means there was a corresponding loss that should be
put in the stats. You do not need to know that players name. You can put it under anonymous .

That's a good idea. Should we also include a "win" under anonymous for every loss reported? Also, if we did this some games would get reported twice as two wins and two losses if both players reported their games. What do you think? I can certainty add anonymous to the stats.

pk867 · Post by **pk867** » Tue Mar 09, 2010 10:56 am

It is just like a balance sheet.
I reported 1 axis ultimate loss and 2 allied ultimate victories, so there would be a corresponding under anonymous 2 axis ultimate losses
and 1 Allied ultimate victory.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Tue Mar 09, 2010 2:02 pm

I agree that we can plot both wins and losses when a player reports his results. You can keep internally who played who so you know when to change an anonymous loser result into a true name result and vice versa with winner results. Then you will always get an accumulated 100% of wins and losses.

Please notice you that have a DIV0 error message in the cells containing draws. This can be solved by making the value in the cell calculated via an if statement similar to:
=IF(A1<>0; B1/A1;0). This means that the cell will show B1/A1 if A1 is not equal to 0 and 0 if A1 = 0.

rkr1958 · Post by **rkr1958** » Wed Mar 10, 2010 3:26 am

Added Anonymous, who really has had a tough go of it with a 0-10 record.

Also, took care of the divide by zero.

rkr1958 · Post by **rkr1958** » Sat Mar 13, 2010 4:40 pm

Stats updated.

rkr1958 · Post by **rkr1958** » Sun Mar 14, 2010 4:48 pm

Added Neil and our first draw.

rkr1958 · Post by **rkr1958** » Mon Mar 15, 2010 2:49 pm

Updated Stats.

rkr1958 · Post by **rkr1958** » Sun Mar 21, 2010 12:48 am

Updated Stats.

massina_nz · Post by **massina_nz** » Sun Mar 21, 2010 9:27 pm

Anyone seeing a disturbing trend here? More Allied wins than Axis. Not a huge sample size tho'.

rkr1958 · Post by **rkr1958** » Sun Mar 21, 2010 9:47 pm

massina_nz wrote:Anyone seeing a disturbing trend here? More Allied wins than Axis. Not a huge sample size tho'.

Note that margin of error still includes 50%/50%.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Sun Mar 21, 2010 10:15 pm

We haven't even received any finished games with the latest balance tweaks for the east front either.

You need maybe 1000 games or more to get a certainty within a few percent about the game balance. The numbers we have now show about 40% Axis victory. That's not a small number. If we had seen 10% Axis victory and 90% Allied victory then we really had a problem on our hands.

Making a game completely balanced is not possible. There are too many variables. The goal must be to give both sides a fair chance to win and I think the numbers show exactly that.

One conclusion we can draw is that it seems that it's a bit more challenging playing the Axis. So you need to have a very sound strategy in order to get the upper hand. You can afford to make more minor mistakes with the Allies and still have a chance for victory. So maybe playing the Axis is like playing black in chess?

I'm sure Kramnik or Carlsen would beat me with both white and black in chess. It will be the same with GS. I think most players will lose against aces like Supermax regardless of which side they choose.

rkr1958 · Post by **rkr1958** » Sun Mar 21, 2010 10:56 pm

Added massina_nz.

ncali · Post by **ncali** » Mon Mar 22, 2010 3:10 am

Stauffenberg wrote:We haven't even received any finished games with the latest balance tweaks for the east front either.

You need maybe 1000 games or more to get a certainty within a few percent about the game balance. The numbers we have now show about 40% Axis victory. That's not a small number. If we had seen 10% Axis victory and 90% Allied victory then we really had a problem on our hands.

Making a game completely balanced is not possible. There are too many variables. The goal must be to give both sides a fair chance to win and I think the numbers show exactly that.

One conclusion we can draw is that it seems that it's a bit more challenging playing the Axis. So you need to have a very sound strategy in order to get the upper hand. You can afford to make more minor mistakes with the Allies and still have a chance for victory. So maybe playing the Axis is like playing black in chess? I'm sure Kramnik or Carlsen would beat me with both white and black in chess. It will be the same with GS. I think most players will lose against aces like Supermax regardless of which side they choose.

I don't think you necessarily need that huge of a sample size. Victories versus defeats aside, I think it is interesting there is still only 1 reported ultimate Axis victory while there are 9 reported ultimate Allied victories.

Peter Stauffenberg · Post by **Peter Stauffenberg** » Mon Mar 22, 2010 8:59 am

Getting an ultimate Allied victory is easier against a not so good opponent than an ultimate Axis victory. If the Axis player runs out of oil too soon (poor play) then you can be in Berlin before the Summer of 1944. In order to get an Ultimate Axis victor you need to have Berlin, Rome, Paris and either London or Moscow at the end of May 1945. That's not easy unless the Allied player truly messed up.

So ultimate victories don't tell us much. Look at minor and major victories instead.

Most losses are from "anonymous". So we don't know the difference of quality of the players. If I could score an ultimate victory as the Allies regularly against Ronnie, Jim or Joe then something would be wrong. But it seems the good players usually get minor or major victories against each other. I only got a strategic Axis victory against Ronnie because he made a big mistake and I later prevented him from landing in France in 1944. It should have been a minor victory.

You can use statistics for anything and it's easy to draw the wrong conclusion because of the data you analyze. So I think it's better to look at games where we know both sides and not where an experienced player has beaten a novice.

joerock22 · Post by **joerock22** » Mon Mar 22, 2010 2:37 pm

Stauffenberg wrote:One conclusion we can draw is that it seems that it's a bit more challenging playing the Axis. So you need to have a very sound strategy in order to get the upper hand. You can afford to make more minor mistakes with the Allies and still have a chance for victory. So maybe playing the Axis is like playing black in chess? I'm sure Kramnik or Carlsen would beat me with both white and black in chess. It will be the same with GS. I think most players will lose against aces like Supermax regardless of which side they choose.

I second that, and I don't think having the Axis side be more challenging is a bad thing. They lost the war, and to win you have to change history. I think the best way to test game balance is to have two people play each other in two games at the same time, with one as each side. This is what supermax and I have been doing, and I am probably going to win both games (finally...no more 0-4 record! He's still the king of vanilla CEAW though).

Considering that, I'd say the game is fairly well balanced.

rkr1958 · Post by **rkr1958** » Sat Mar 27, 2010 11:58 am

Updated stats.

rkr1958 · Post by **rkr1958** » Sun Apr 04, 2010 3:06 pm

Added Dave.

rkr1958 · Post by **rkr1958** » Wed Apr 07, 2010 2:36 am

Updated Stats.

rkr1958 · Post by **rkr1958** » Thu Apr 08, 2010 12:27 am

Updated Stats.