Pointless Singles - Arms Race! (Finished - Winner=wrexham)

[DaveH]

Thanks for the feedback. The original idea was to try and make it more of a level playing field for differently ranked players, though I appreciate that the disparity can be difficult to make up if the diifference in points is greater than the difference in win ratios - which it clearly is given the results table.

Perhaps points based just on on ranks rather than CC points?

I'll have a closer look at the wins and points and see how they compare.

[DaveH]

Just to follow up on the last few comments, if you assume that everyone was an equal rank so that 20 points would have changed hands on each game, the number of assumed wins correlates to within 1 of the actual number of wins. I'll do the math again at the end of the tournament and publish the results in one of the newsletters.

One additional thing I love about CC is the way in which there are so many different formats to these tournaments - and trying out new ideas - though I accept that some may not work too well!

[Dukasaur]

I think this would be better if you put a cap on the losing. For example--the most points you could lose is 25 or 28 Something like that. You start getting up in the 30s when other players are only losing 10-12 and they gain a 75% advantage. Something to think about.

I have to be considered one of the best on this map and I only win at a 66% win rate.

Other maps that you do this for you have to consider 60% as the factor of a good player in a one vs one--that is actually very good. 55% is good.

If you factor those numbers in to make things fair you have to cap it.

Just a thought for the future of this series if you continue it.

Great idea though

Yeah, I think that might be the basis of an idea.

Points won = (8 + (actual points won)*0.60)/20

This still fulfils the original goal of rewarding one for beating a higher-ranked player, but the difference is much less, to account for the huge luck factor in 1v1 games. The correction factor is HA's 60%, and a constant of 8 is added so that it still returns the intuitive value of 20 for equal ranked players.

[DaveH]

Points won = (8 + (actual points won)*0.60)/20

This still fulfils the original goal of rewarding one for beating a higher-ranked player, but the difference is much less, to account for the huge luck factor in 1v1 games. The correction factor is HA's 60%, and a constant of 8 is added so that it still returns the intuitive value of 20 for equal ranked players.

Very well thought through - I like it! 30 difference gives 24; 10 gives 14 - very good! I'll use that in the next one.

[Dukasaur]

Well, I'm almost last in the standings, but at least I won my last game, so I'm going out on a high note...

Thanks for hosting this, Dave!

[DaveH]

Congratulations to wrexham who won by just 3 points from ianphull who was just 4 points ahead of Mr. CD!


There were players who had more wins, but none who had more than 1 win below the next player - and the specification of this tournament was to try and equalise, to some extent, the experience and expertise of players of higher rank by awarding points based upon rank difference (as CC does) rather than straight number of wins. However, with 1 v 1 on a map such as Arms Race! together with starting first, the luck factor has led to large points differences awarded in a few cases that made it very hard for highly ranked players to compete.

For this reason, future pointless tournaments will use the formula "Points won = (8 + (actual points won)*0.60)/20" which has a 60% factor on points difference that increases points below 20 and reduces them above 20. With thanks to Dukasaur and HA for their input on this - I hope you join in the next pointless tournament according to wrexham's map specification!