SpielByWeb

[Joshuaaaaaa]

Hey, for my 1st tourney, I'll take a 10th place finish.

[Zaui]

[Kanga]

40/42 done. The last two games dont affect the final placements.

tphon, almondralf, Kanga and Games play in the final.

tphon creates a v0 game, ralf the V1, I get the V2 and Games the V1,2 game.

Game name:
4player T15 Final A/B/C/D
password: final

Map: your choice, or, if you dont want to choose... here are some suggestions:

514
117
497
493
271

Standings:

[Dobinator]

Can't believe I don't make the final with four wins.

If I'm griping, using the margin of loss for tiebreaks is only one indicator of skill, and not always a telling one - I had one game here where I was blocked by several opponents, so I couldn't score anything, and ended up with a big deficit. Conversely, in another game, I won by like 15. It might be fairer to also give tiebreak points for margin of victory over the nearest opponent in wins.

Under the current system, you get lots of credit for coming in a close second, but no credit for a runaway victory, and both of those seem identical indicators of skill to me. The runaway victory even more, maybe, because often you achieve a narrow margin of loss if the other guy buys out animals to finish, even though you weren't really that close to winning.

Overall positions, like Gad uses in Amun Re, is maybe even better, if harder to keep track of.

OK, sour grapes mode off. Good luck to the finalists.

[Kanga]

[tphon]

[Kanga]

[Dobinator]

Margin of loss is also not fair if the maps and variants aren't the same for all players - some maps tend to be much higher scoring, and variant 2 cuts down on overall scoring by a lot. Variant 1 is often lower-scoring, too, I think, depending on the map layout.

I wasn't suggesting giving partial tournament points for 2nd or 3rd - just using those placings as the tiebreakers after number of wins, since ordinal placing seems to me to be a less-random measure of prowess than margin of loss. I don't see how that would get folks to play for second.

I agree that no system is perfect, and if the margin is easier to record than something else, then that's a significant plus. Thanks for running the tournaments.

[Knave]

I have an idea is almost certainly unfeasible, but I like it nonetheless...

Find the mean of the margin that each placing lost by. Then, every loss is increased by an amount equal to the mean of the loss by somebody in that placing.

Hypothetical example:

At the end of the round, the mean margins of victory over each placing are as follows

1st: 0
2nd: 13
3rd: 19
4th: 27


Now, Kanga, Tphon, Dobby and Knave play a game. The final score for that game is as follows:

Kanga: 180
Tphon: 175
Dobby: 149
Knave: 148


The margins of loss are as follows

Kanga: 0
Tphon: 5
Dobby: 31
Knave: 32

Now we add the placing penalty to these margins

Kanga: 0
Tphon: 18
Dobby: 50
Knave: 59


So, Dobby only beat knave by a point, but gets a 9 point spread in the standings because he squeezed into 3rd place. More importantly though, both Knave and Dobby got smacked around appropriately for their 30+ point loss.

The above would only be a tiebreak of course. I just invented it now, so I haven't fully thought through any flaws. Some easy adjustment is possible, you can multiply either the means or the game margins by some arbitrary constant to weight one or the other higher. You can square one of the measures if you really want to punish big losses and/or poor placings.

The main drawback I see is that it would be a nightmare for the tourney organizer, but a spreadsheet could probably handle the calculations... if that is what the tourney people use for these standings. I could make the spreadsheet , but I'm not entering scores or looking them up.

[GamesOnTheBrain]

[GamesOnTheBrain]

[Dobinator]

That is what I mean.

[Kanga]

I have actually paid attention to the above discussion and am toying with changing things for future tournaments.


This tournament has now concluded. Final standings:

[Kanga]

The final saw my luck finally return and allowed me to win 3 out of 4.

Winner: Kanga (3 wins)

Runner up: almondralf (1 win)

Finalists: GamesOnTheBrain, tphon