Great going guys! We’ve played 16 (actually 17, but one was a mistake) games between us in 9 days and in so doing have completed the first round of the Last Meeple Standing. There now follows a quick summary of all the action so far (bring tea)...
Firstly, I’d just like to say a big thank you to jma03 and MrNumbers for breaking my spreadsheet (kind of) with their tied game. Yes, tied games are a possible outcome that should be allowed for. And yes, I’m incompetent with Excel. But I was so pleased with myself for putting together something that actually seemed to work and calculated all of the scores for all four scoring models, that it hadn’t even occurred to me that I wasn’t allowing for draws. Doh.
We’ve scored 3087 points between us, with wins ranging from 80 to 130 points and with losses ranging from 46 to 124. Jungleboy has achieved both the lowest-scoring win and the highest-scoring loss, with highest-scoring win being won by Rosco (against MrNumbers of all people
) and lowest-scoring loss going to Decar (against Jéré).
Next up, here’s a chart of win/ draws/ losses and overall PD as if we were playing an ordinary league:
(there’s no ranking, we’re ordered alphabetically)It’s good to see that all the coolest players have at least 3 wins and an overall PD of 90 or more
. Other than this, things seem quite even with most players winning 1 or 2 games albeit with some surprisingly wild fluctuations in PD. I’ve included “Heavy Losses” (20 point loss or worse) because one of the scoring models uses this. Half of us have suffered at least one heavy loss although Jéré has mostly recovered from his (
) and Paul has been on the receiving end of 3!
In terms of individual PD per game, jungleboy and Rosco are by far our most consistent players with standard deviations of 4 and 6 respectively. At the other end of the scale is wild-card Decar with an SD of 23, owing largely to the fact that his PD ranged from -59 against Jéré to +49 against jma03.
So, with all that in mind, let’s consider the following (you may need more tea by this point):
It can be seen that LP adjusted after each game leads to the biggest difference in LP and leaves Paul and jma03 in the Danger Zone, with Decar hanging on by the skin of his teeth. However, although this scoring method has its benefits and certainly seems to be one of the more exciting models, the problem still exists with the unintended significance of the order in which games are played. Game order must be irrelevant to overall scores in order to keep things fair and equal, which kind of rules out this method unless everyone sticks to a randomly pre-determined order of games which risks slowing things down with the potential for missing games.
So then, in the next column we have scores based on LP adjustment at the end of the round rather than the end of each game. This means that wins and losses can cancel each other out in terms of LP regardless of individual LP at the start of each game. This makes it fairer, but easier with it, and could lead to much longer tournaments unless other factors are taken into consideration that can be used to restrict LP. On the most part however, this model has brought the majority of us closer together although Paul is still in the Danger Zone and jma03 and Decar are further down the table despite avoiding the DZ.
Adjusting the LP cap (initially 50) by -5 points per loss seems like a good way of getting round the “never-ending competition” problem. However, it brings new problems with it in that players who suffer a loss of less than 5 points could end up being penalised twice; once for the loss and again when their LP cap is reduced. This aside, it does seem to work well as it has reduced the bunching at the top (previously half of us were left with max LP) and pushes everyone who was previously on a comfortable 50 down towards the DZ a little since nobody won all of their games.
The final suggestion was that LP cap could be adjusted by -5 points for heavy losses only. The definition of a “heavy loss” is still up for debate, but initially it was suggested that a loss by 20 points or more seemed appropriate so this is what I’ve used. This wouldn’t actually make any difference to Paul or Decar as all of their losses were heavy ones, but it does mean that there’s no extra penalty for light losses which seems like a good thing.
I’ll leave my detailed analysis at that for now, but please post any questions or feedback you may have. I’m all ears. Not literally.
All being well, I’ll aim to post the Round 2 fixtures later today. These will including specific start players so we can ensure that each player gets to place the first tile in 2 of their 4 games.
Share Topic
June 15, 2015, 03:45:31 PM