Thanks for posting crafty

of course i didn't guess right. i was told there were surprises, so i picked two middling spots that might have been the big shockers. but there werent really any big shockers. guessing that central america, scand, ant, the russias, indo, east china and s america are good isnt that surprising. guessing that canada, east russia, europe, and the americas are bad isnt really shocking either. and guessing the exact order is irrelevant since it will almost certainly change over time.

its interesting data. theres definitely some skew. a country thats good when done right - like east africa - might be better when you get another africa, but a trainwreck when your opponent does. a country that's better as a counterpick than an actual pick - like western US - may show up as winning more or less than it should. and it may be a self fulfilling prophecy where the top players tend to copy each other, and people who are paying attention tend to copy people who beat them, so the ones the top players pick may do better by the fact that the skill level of the guys playing them is higher. but if we can set aside the difference between the 4th best and 5th best countries and focus on the top ones, the data does seem to bear out what i think most of us would have expected. and that's good to know.

Very interesting data -- thank you crafty.

I was about to post Middle East for my candidate for worst pick but the data was already posted. Caucausus was my other choice for worst pick. West Russia as the very best is quite interesting too.

whoops, I meant to say aussie as my worst, not ant, lmao

hi, i'm the huge bonus that borders all of the faster better bonuses on the medium earth map. i take several turns to get and i'm not that much more valuable than the smaller, easier, and more defensible bonuses around me. you might remember me because i was overpowered on the original earth map but i totally suck on this one. if you're planning on getting me, maybe you should start in the russias and work your way down instead of wasting a starting spot. if you guessed that i'm caucusus, you're right.

lol @ bostenfred

I'm not surprised. My guess for the shocker number 1 was going to be Indonesia. I thought Antartica would be a lot lower than it is. I find, generally speaking, that it's the worst out of the 3 armies for 4 territory bonuses (Central, Scandinavia, India), but I guess the data suggests otherwise.

I'm not too surprised by Caucasus either. It's value significantly decreased with the Russia's becoming much stronger. It's now more of a counter pick to an East or West Russia rather than something you legitimately expect to get early.

To counter all the criticism.

As far as I checked it, this analysis is completely sound and as unbiased as possible.
However, I'd like to add a column with the standard errors, confidence intervals and tests though, it gives a lot more information! (inference versus just descriptive statistics)

**Error**: the standard errors of the win%, this depends on the sample size, in this case the number of times occupied, and the success probability itself (win%).
**95% Conf. interval**: lower and upper boundaries of the 95% confidence interval. Interpretation: we are 95% sure that the population wise true (unknown) win% (over an infinite number of games) falls within these borders. For the blue and red territories, 50% is not included, the blue territories score better than 50%, the red worse.
**P-value**: Test to check whether the underlying probability of West-Russia and another territory are different. Interpretation of the %: how likely is it that when West-Russia and the given territory have the same underlying probability, that we get these data? When this value is smaller than 5%, this is very unlikely and we conclude that the given territory has a significantly lower probability. For the bold territories, we CANNOT conclude this. The territory with the underlying best probability is thus one of the 8 territories in bold.

If anyone has any more questions, please ask.


Fizzer, I think it would be interesting to use a new map configuration at some point. I think just bumping the bottom 10 or 11 bonuses from this list up by 1 would make this map even more interesting...

Mine were pretty spot on: "Scan/Antartica and Mexico. After that it's the choice 4s: East China/West Russia/E&W Africa."


I got 5 of the top 7. It's probably 6 of the top 8 too, since Mexico isn't on the chart.

Had I thrown the other 3 and 4 in (Indo and India) I'd have had them all. doh.

BTW -- my critcism was based on my misunderstanding of the term "occupied". My point about touching starting points was entirely obviated once I was told it meant picks.

Mexico = Central America (9th)

Any requests for what type of data/stats I should look in to next?

I'd love to see a ladder with autopicked spots based on a running analysis of the highest win % spots. Randomly select the starting spots in pairs where the win % is within 1.5 or maybe 2%. Everyone's ability to infer where their opponent was starting out would be exactly even instead of dependent upon the interplay between first pick luck and pick order...

How bad luck on the first couple turns translates into wins and losses is something that I would greatly like to see.

Thanks for the work you've done!

I would like to know who wins more often, the player who attacks on first contact or the player that defends

Average income after turn 1, 2, 3, 5, 10.

|>"I'd love to see a ladder with autopicked spots based on a running analysis of the highest win % spots. Randomly select the starting spots in pairs where the win % is within 1.5 or maybe 2%. Everyone's ability to infer where their opponent was starting out would be exactly even instead of dependent upon the interplay between first pick luck and pick order..."

Sounds like you are like me -- I am not a big fan of manual distribution. But what I would prefer to see, rather than what you suggest, is a more balanced map and random distribution (with wastelanded bonuses not included in the random distribution).

I know, that would be great in a perfect world, but I think that any map that isn't purely symmetric will end up being imbalanced in one way or another, so doing autopicks based off of %win per starting spot statistics is a way to provide a starting balance to an asymmetric map.