A summary of thoughts on scoring methods for tournaments

greenspun.com : LUSENET : MAME Action Replay : One Thread

On the eve (well soon anyway) of the 4th MARP Tournament I just thought I'd give my updated view on the scoring system we use at the MARP tournaments versus the Deca2000 tournaments.

Having been a person who had a significantly better score on a game than the 2nd a following places (well 2 acutally Galaga and Gyruss on T1) I was of the opinion that perhaps % based scoring may give a fairer outcome for the tournament. We did have some discussions back then about how to give some recognition to a game that was significantly better than the other players.

However, having taken part in the Deca2000 I can see now that if someone totally owns a game (such as I did with these games and hopefully will again with Tapper before the end of Deca2000) then they can significantly impact on the final score for another player. If you get several games that people own, then someone who say owns two games can get such a significant advantage out of these 2 games that it makes it harder for anyone who's reasonable on most of the games to make up lost ground.

Who's the better player, someone who can get 2nd/3rd place (or close to it) on alot of games or someone who happens to be good at just a couple? I think the 1st of these propositions is the better one and even though it meant I had to fight to find points at T1 there was some satisfaction in improving just enough to gain an additional placing. There's not much reward currently for someone to improve by a few thousand points at Deca2000 on several of the games that are into multiple hundred thousand (or multiple million) points for 1st place already.

So, having been an undecided fence sitter for a while, I think I can safely say that the MARP means of scoring adds an extra incentive for games that someone has totally owned such that a minor improvement can still improve your leaderboard points. Conversely, on a game that has a low 1st place score you may not improve enough to claim an extra point in MARP whereas with a % system that may have been worth a significant amount.

All things considered, I think that the benefits of the MARP Tournament scoring system while penalising to some extent a significantly better 1st place than 2nd place score does result in a closer and more hard fought end result.

Now, if only I could get enough good recordings at Deca2000 to get back into 2nd or 3rd place I could definitely happily join in at T4... although it will definitely show how much I suck at fighting games so I would expect to be if not the last player then close to it on the Street Fighter game. ;-0

Well, I think that's more than my $0.02 so I'll leave it for now and may the best player win at T4 and may we all be happy with our own personal bests - after all, it's not just the winning but the fun (and aggravation/disappointment) of the tournament that counts.

You all have fun out there and may the games commence... soon..

BeeJay.

PS: Only 1 month until I can more seriously consider returning to MARP.

PPS: The only IRC server on EFnet that will let me connect is still permanently net-split from the rest of EFnet so I still can't catch up with you guys in the #marp chatroom. :-(

-- BeeJay (mamester@i4free.co.nz), May 31, 2000

Answers

You might want to try irc.home.com, port 6667, beeJay. It's a cable server, and I don't know how widespread it is, but I've had good luck with it.

Looks like we've both slid a little on our original scoring viewpoints, BeeJay. And just when we were approaching the one year mark on the famous 'elitist' war. I had originally planned to have around 1000 recordings at MARP by this time, and damned be my average. Funny how things change. :)

Cheers, Q.T.Quazar (who is eagerly awaiting yours and JG's return)

-- Q.T.Quazar (qan@home.com), May 31, 2000.


*If* I enter T4 BeeJay you will not be last on StreetFighter.....That and lack of time is the main reason I probably won't enter, I can't abide fighting games....

Phil

-- Phil Chapman (pchapman@qspinc.com), May 31, 2000.


Positional scoring is without question an improvement over percentage based scoring. I'm in the Deca2000 tourney as well and have been somewhat discouraged from entering scores. I could submit what I believe to be average scores, yet I'll probably only receive less than 1 or 2 points since the top score is so astronomical. A score that embarassingly low should be reserved for last place only!!

For MARP's T4, the scoring will be even closer as there are only 8 games played instead of 10. This will certainly make for a very close and exciting finish!!

-- Pat (laffaye@ibm.net), June 07, 2000.


Moderation questions? read the FAQ