Updates from Garrick van Buren Toggle Comment Threads | Keyboard Shortcuts

  • Garrick van Buren 8:21 pm on February 16, 2013 Permalink  

    @swedenssons – you spent the Loppet observing teams, any interesting new game play you witnessed?

     
    • Eric Anderson 8:24 pm on February 16, 2013 Permalink | Log in to Reply

      Tossing at the king to lose on purpose to start game two when time limits are short.
      The rescue play I saw and how it unfolded.
      A lot of young guys with the look in their eyes that they could get into the sport part of it.

  • Garrick van Buren 2:46 pm on November 19, 2012 Permalink  

    I’ve been trying to find a stat that helps quantify the value of an advantage line. This time around, I’m comparing first baton blasting with and without an advantage line.

    average difference by # kubbs inkast:
    1: +10.5%
    2: +10.9%
    3: +16.5%
    4: +14.3%
    5: -20%
    6: +4.5%
    7: +13.7%
    8: +16.2%
    9: +2.75%

     
  • Garrick van Buren 2:49 pm on November 14, 2012 Permalink
    Tags:   

    Number of Base Kubbs Knocked Over in Winning Turn:

    0B: 37%
    1B: 37%
    2B: 23%
    3B: 1%
    4B: 1%
    5B: 0%

     
  • Garrick van Buren 4:28 pm on November 13, 2012 Permalink
    Tags:   

    New graph – Percent Chance of Clearing All Inkast Kubbs w/ 1st and 2nd batons.

     
  • Garrick van Buren 3:46 pm on November 12, 2012 Permalink  

    Distribution of # of kubbs inkast

     
  • Garrick van Buren 3:42 pm on November 12, 2012 Permalink  

    What are the chances of an advantage line resulting in immediate loss?

    Notice there are only 2 places (@ 2 kubbs & @ 10 kubbs) where leaving an advantage line doesn’t result in immediate game over.

     
    • Garrick 9:20 pm on November 12, 2012 Permalink | Log in to Reply

      taking another look at these graphs…

      Assuming all kubbs in the data are thrown short – these stats support the idea that inkastares should throw 1 long – especially when they have 6, 8, or 9 to throw in.

      Assuming that 1 remaining field kubb is the one thrown long, the chances of the game being immediately over drop by 15%, 10%, and 40% (!!!!) respectively.

  • Garrick van Buren 2:13 pm on November 12, 2012 Permalink  

     
  • Garrick van Buren 10:25 am on November 7, 2012 Permalink
    Tags:   

    last stats for a while.

    The distribution of number of turns across all the games in the wiki

     
    • Dobbie 1:53 pm on November 7, 2012 Permalink | Log in to Reply

      Average is around 5 rounds of play. Can you add a separate charts for opening teams and answering teams to identify any trends between the two (as in, if you open, you better win in 10 turns or your chances of winning drop)

      • Dobbie 1:54 pm on November 7, 2012 Permalink | Log in to Reply

        Is there some way to download something (like a .csv file) that has all the data in excel? How does that work?

        • Garrick 2:01 pm on November 7, 2012 Permalink

          right now, I’m pulling out the data from each game by hand for each chart. I’ll look into building a meaningful csv file.

        • Jamie Thingelstad 8:08 pm on November 9, 2012 Permalink

          All of the semantic data in the Planet Kubb Wiki can be queried and downloaded. Here is the query form for games with turn counts. That can all be downloaded as CSV or even queries via API and get JSON data back.

          You can get a sense of the data that is available for a team by looking at the property page for a game.

  • Garrick van Buren 9:50 am on November 7, 2012 Permalink
    Tags:   

    Of the 46 tournament games in the wiki, 32 have an advantage line somewhere game play.
    The percentage chance that leaving an advantage line results in a loss? 100% 88% (see @chrishodge’s correction below)

     
    • Chris Hodges 10:10 am on November 7, 2012 Permalink | Log in to Reply

      That isn’t true – see these as examples:

      Twin Cities Winter Tournament 2012 Final Game 1
      Rockford 2012 Championship Head Slap v. Tad Kubbler (Game 2)
      US Nationals 2012, Finals, Knockerheads v. Kubbsicles, July 15 (Game 1)
      Kubb Open 2012, Finals, Kubb’Ings vs Fortschritt99 Blue (Game Two)

      In each of the above the winning team left an advantage line at some point but their opponents returned the favor later in the game.

      • Garrick 10:24 am on November 7, 2012 Permalink | Log in to Reply

        @chrishodges – thanks for double checking the work. Looks like in each of those games, each team left an advantage line. this lowers the chances to losing to 88%.

        • swedenssons 10:32 am on November 7, 2012 Permalink

          Is this in the next turn, or in general?
          Is there a way to run a regression, or something, and see how quickly you lose a game based on the number of field kubbs in play when you allow the other team to move up?

    • Garrick 10:51 am on November 7, 2012 Permalink | Log in to Reply

      @swedenssons – here’s another pass at the advantage line stats:

      32 games have an advantage line
      31 games had the final turn on the advantage line
      66 turns have advantage lines (so it’s tough to clear the advantage line once it remains)

      • Garrick 11:04 am on November 7, 2012 Permalink | Log in to Reply

        This data tells me that an advantage line almost always results in an immediate loss.

        Primarily because of the difficulty in recovering from leaving an advantage line.

        There’s a slim chance the losing team may be able to hold on for a couple more turns, maybe even recover from that initial advantage line, but they’ll leave another one – and lose immediately.

  • Garrick van Buren 8:36 pm on November 6, 2012 Permalink
    Tags:   

    In case you were wondering about the chances you might leave an advantage line here’s a handy chart:

    from the 46 tournament games currently in the wiki

     
    • Dobbie 9:00 pm on November 6, 2012 Permalink | Log in to Reply

      Sweet, sweet stats. Under 50% to over 90% from 9 to 10 kubbs. Interesting.

      • Garrick 10:16 pm on November 6, 2012 Permalink | Log in to Reply

        I kept hearing Terry Ekelöf saying ‘Get Better’ as I was putting this graph together.

        • Dano 9:16 am on November 7, 2012 Permalink

          I also hear Terry saying “we play not to lose” when he was talking about throwing some kubbs long when there were more then seven to toss. Makes a lot of sense after seeing this chart.

        • Dobbie 9:55 am on November 7, 2012 Permalink

          @dano – perhaps for 6 player teams throwing long in certain situations makes sense. Not sure that strategy is valid for 2 or 3 person teams…..yet.

    • Jamie Thingelstad 10:45 pm on November 6, 2012 Permalink | Log in to Reply

      Great graph @garrick! I’m looking forward to automating a graph like that right in the scoring system. This is a key stat for teams to know of their own roster.

    • swedenssons 8:55 am on November 7, 2012 Permalink | Log in to Reply

      This is awesome stuff. A couple quick ideas/questions. Can you have horizontal lines across it so one can see the exact(ish) percentages. Also, I would be very curious on (if it is possible) the percent you lose a match in the turn you let the team move up AND the second turn after you let them move up. Is that possible? That would then tell a person how often you let the team move up AND what the result of that is. I have a feeling that everything is possible for you two guys though. I will also just add one more thing. I think this is for a lot of late in the day games, right? The top teams playing in a tournament. I would be really curious what these results would be like if they were for the “average” game. My thought is that the percentages would go up. Perhaps that is a very easy statement to make, but it would be neat to see how much. Another reason to score more random matches throughout the day. The other cool thing would be to see the results for two-person team tournaments vs. three-person team tournaments. Again, this is really cool. Not sure how long it took you to run the numbers, but whatever it was, thanks for doing it.

    • Garrick 9:16 am on November 7, 2012 Permalink | Log in to Reply

      @swedenssons – here are the exact percentages:
      1:5% (admittedly, this is all me)
      2:0%
      3:9%
      4:10%
      5:19%
      6:15%
      7:23%
      8:31%
      9:47%
      10:100%
      Of course – these percentages and all the other overall stats will be much much more accurate with more games. With only 46 games outliers are much more prominent than they should be.

  • Garrick van Buren 9:35 pm on October 16, 2012 Permalink  

    I enjoyed the Bavarian themed kubbs @ Dallas so much that I started looking around for painted kubbs around the internet. My favorites thus far:

    Grain Belt!

    Monsters!

    Moose!

     
  • Garrick van Buren 7:59 pm on October 16, 2012 Permalink  

     
  • Garrick van Buren 3:06 pm on October 16, 2012 Permalink  

    I just grabbed more than 200 turns out… 

    I just grabbed more than 200 turns out of the games in the Planet Kubb wiki where 100% of the field kubbs were hit.

    Each one of those baton throws on average hits 1.3958 field kubbs.

    Think of that the next time your inkastare gives you a tempting cluster of 6.

     
    • Jamie Thingelstad 8:09 pm on October 16, 2012 Permalink | Log in to Reply

      @garrick and I were talking about a calculated value that would be sort of like team efficiency. Something like:

      ( (field kubbs hit on first baton) + (field kubbs hit on second baton) ) / (kubbs thrown in)

      Only valid when Kubbs thrown in is greater than 4.

      So, you throw 5 Kubbs and get 2 in the first throw and 2 more in the 2nd throw, you have:

      (2 + 2) / 5 = 0.80

      I’m intrigued by this.

  • Garrick van Buren 5:32 pm on September 27, 2012 Permalink
    Tags:   

    Planet Kubb ‘Unite People and Create Peace’ T-Shirt – $15 

    The first run of Planet Kubb Unite People & Create Peace T-shirts will be in Dallas. $15 + $5 shipping.



    If you’d like a different size (womens, kids, etc) – email me

     
c
Compose new post
j
Next post/Next comment
k
Previous post/Previous comment
r
Reply
e
Edit
o
Show/Hide comments
t
Go to top
l
Go to login
h
Show/Hide help
shift + esc
Cancel