Statistics of the Small World

Inspired by a comment that Chris Ingersoll made, I ran some statistics about how starting position affects ending placement in Small World.  Kind of like we thought, there does appear to be a correlation.

The first thing that I need to explain about this data is that it is only based on games where we tracked the player order for the game, which totalled only 7 games.  Two of these games were with 5 players, two were with 4, and the other three were 3-player games.

In this first chart, you can see the raw number of different finishes for each starting position.  Generally, there does seem to be a direct correlation to starting first and ending on top.  The biggest problem with this presentation is that you can’t see how many players were in each game, so while finishing 3rd might sound good, it might also be losing the game.   

Therefore, in this graph I simplified what I was looking at and only reported first and last place finishes for each starting position.  Here you can see that even more clearly that going first has definitely proven to be an advantage.  Specifically, no one that started first has ever finished last in this sample of games. Conversely, no one starting 4th has ever won.  

Of course, there are lots of things that I can’t account for in this data.  Most importantly, player skill will obviously have a huge impact on performance, but can’t be reflected here in any way.  For instance, both wins for 3rd position were from Chris Ingersoll, who is probably our “best” player at this time.  Does his performance then reflect on starting in that position or on his ability to play the game well?  Maybe a little of both, but my bet’s on Chris.  

So, I guess that my conclusion at this time is that there does appear to be an advantage to going first in Small World.  Both from these limited statistics and from the fact that it just makes freaking sense that it’s good to have the first choice at races and on where to start on the map, going first seems like a good thing to do.  If it really is that big of a deal, it makes me wonder if there might be a way to account for the advantage by changing the random (or “pointy ear” thing) start into some sort of auction.  Maybe give all the players 7 or 10 VP coins to start, and them have them bid for starting order.  It might be something to think about.

Anyway, I’ll keep tracking the data and report back on it in a few weeks.  Until then, what are you thoughts about starting order in Small World


  1. Britt

    So having pointed ears does have an advantage!

  2. Chris Norwood

    Yeah, as long as you can put up with the “being a freak” thing! 

    (but I guess you’re probably used to it by now…)

  3. Chris Ingersoll

    To be fair, at least one of the *last*-place finishes from a 3rd (and in this case last) starting position is mine as well, as that just happened this week. 🙂

    The other major factor that can’t be accurately analyzed is what race/powers are available to each player at each time (other than what is actually selected). First-pick Trolls of any power will go a long way to helping the 1st player not lose, for example.

    Still, it is an interesting little group of data.

Comments are closed.