Share on facebook
Share on twitter
Share on reddit
Share on email

Power Stats

In response to questions at Game No. 88, here’s a look at how each of the powers has performed, through Game No. 88, in terms of average center count. These stats exclude Game No. 55, which was a non-standard game.


Power Average Count
France 5.697
Turkey 5.359
England 5.343
Germany 4.888
Russia 4.544
Italy 4.025
Austria 3.889

And here’s a look at elimination frequency, again excluding Game No. 55.


Power Eliminations Pct
Austria 40 45.8
Russia 29 33.3
Germany 27 31.0
Italy 26

29.9

Turkey 23 26.4
England 18

20.7

France 16 18.4

Join the discussion!

Find out more about an upcoming event or article, talk smack before a game, brag about your board top, or most likely, ask what on earth your fellow Weasels were thinking!

This Post Has 8 Comments

  1. Anonymous

    Interesting the position of Russia. I wonder what the correlation is between the club’s tendency to deny Russia Sweden in Fall 01 and the results as it affects the other countries?

    For example, if in the games where Russia takes Sweden
    does both Russia and Germany do better and England do worse?

  2. Jim O'Kelley

    That would be helpful to know. However, the data on neutrals exists on our supply center charts, not in our database, which is unfortunate. Maybe our technical gurus can think of a way to fix that…

    I will say, having just topped a board as Russia, that it really helps to have a reasonable German player. A lot of our players feel that it’s Germany’s [i]duty [/i] to bounce Russia out of Sweden. A reasonable German player may position himself for the bounce, but his decision whether to bounce will be based either on his Scandinavian agenda or his intelligence gathering in the south.

    In our game, Peter Yeargin played Germany and opened to Sweden. Although the Turks were in the Black Sea and positioned to deny me Rumania, Peter decided to bounce because he had learned that Austria would be providing a necessary support for my move to Rum. He didn’t want me building twice, so he bounced me.

    If he had thought I wouldn’t be getting Rumania, he would have allowed me into Sweden. That’s all a Tsar can ask.

  3. Eric Brown

    how about board-tops by power (shared tops split into fractions)?

  4. Jim O'Kelley

    [quote]how about board-tops by power [/quote]

    This would be an interesting metric to track.

    First, let me say that we’d only want to go back through 2008-09, since I don’t think board tops were as relevant in the draw days. Agreed?

    Second, I like to think that I’m a sophisticated Access user, but this may be beyond my programming capabilities. We have the necessary data in the database, but I’m not sure how write a query that will automatically determine board toppers. My solution would be to add a field that would allocate one point based on board-topping, and then sum that field by country, but that would require additional data entry by me.

    Anyone know how to caluclate this via query?

  5. Anonymous

    Hi Jim,

    If all your minions are bouncing in Sweden as you seem to mention in your comment, then there is the EF v GR discrepancy in the SC averages for you.

    I know that when I am playing E and there is a good player in G, I don’t tend to bother wasting time encouraging a Sweden bounce because it just isn’t going to happen.

    Peter (with an r).

  6. Jim O'Kelley

    If you’re interested, I could also post player ratings, based on average center count and for players with a minimum of, say, five games. Worth seeing?

  7. Mike Morrison

    [quote]If you’re interested, I could also post player ratings, based on average center count and for players with a minimum of, say, five games. Worth seeing?[/quote]

    Sure!

Leave a Reply

White article icon

More Articles.