This has multiple angles I’m coming from and wish to sort out...
It’s cool that you RL guys are listening to the players and sorting the scoring issues (and more). But I can only find topics and threads about rectifying the SP scores... I'm more concerned with the O/L MP boards and how that scoring system works,
What factors are currently used
How these factors are converted to a score,
How it all adds together (or hasn’t in some cases)
How it relates to the Players EXP Level… (what point the exp lvl actually serves)
If there’s a problem and if it’s going to get or has been fixed (if only in time for the next game [looking forward to it btw])
And can I have my MP rank rectified? Please
I am currently #247 with a score: 2,366,388 1040/1739 wins/games (this works out to [wins/games] x 100 = 59.8% win average) and has taken 3d12h21m to get this far
I can understand that solely using the win/game ratio for ranking is a flawed system for 2 reasons:
1) winning on your first O/L game would equal #1 rank as [1/1] x 100 = 100%
2) people who only exclusively play against nubie ranks to cheat the system and increase their average (watch out for my next topic of annoyances where u get to name and shame your favourites)
As there are ways round these flaws this stat should surly be a major contributing factor counting towards the overall rank position in some way right?
What I can’t understand is why a player who:-
a) has played less games
b) has had less wins
c) has a lower win ratio/percentage/average
d) a far lower playtime
e) I can consistently thrash
f) combinations of the above
g) all of the above
have somehow managed to get a higher score and therefore a higher rank.
EG the guy just above me (hopefully not for too long)
#247 - FcT Gotze - 2,368922 -130/320(40.6%) - 10h 43m
#237 - symo80 - 2,459,936 - 99/414(23.9%) - 22h 18m
I can however understand how this can happen, IF some of these players are being scored by a newer system introduced since I started playing (had my exp lvl reset quite a few times before I learned I had to exit to title screen in order to properly save my progress… I know u’v fixed it but I still do it out of habit)
Now, if the MP score system were to work the same as the SP ranks by converting best time to score I see another rather large flaw:-
A fluke run that the player manages to hit close to or better than WR would superficially raise their rank even tho they have never and probably will never come close to doing such a time again.
(would also pi… upset them slightly as this time is not linked into the SP boards… this opens a whole other discussion topic)
Of course the argument to this is the player who can post consistent fast times would deserve more points and therefore the higher rank. However this is quiet the major factor contributing to the overall win/game ratio/%/average because being consistently faster significantly increases the win average
But it doesn’t work like that either:-
E.g. people who are consistently faster than I am, but are lower in rank:
#278 - x SiniStar x - 2,55,565 - 899/1384(64.95%) - 2d 13h 7m
#284 - x High DEFF x - 2,029,490 - 173/208(83.2%) - 7h 30
(maby he’s using the ‘nubie only’ win-average cheat? LOL)
Recording a players average level time… (not for ALL a players races just the last 3,5 or 10) Adding this into the overall equation would help generate a more accurate rank scoring system (also a factor when calculating how many points are awarded when winning/loosing against higher/lower ranks)
Understandably there are many factors/variables that would count towards to overall rank so instead of bi…. moaning about it all and telling you guys “you need to sort it out”
I’d like to help…
Firstly I/we need more info on the current system so I and others (wishing to add their 2pennith on this topic) can:-
Help identify the factors and variables that should go into determining (MP) rank
(1st 2nd 3rd place %’s, average times, beating higher ranks, games/win streaks in a continual multi game session, DNF’s, etc)
Identify flaws and loopholes in the system
(exiting out of session to reset ‘matched players stats’ I.E. the nubie only win-average cheat)
Ways to circumvent these flaws and loopholes
(geting more rank points for coming a close second to a high ranked player than ‘DNF-ing’ nubies. While DNF-ing a higher ranked person should be worth a sh… shed-load of points)
Identify statistic boosters and detractors
(never / always quitting when matched against much higher ranking opponents)
Help formulate better equations / algorithms to produce a more accurate ranking system
(i'm no mathematician but I’m sure there’s plenty out there willing to chip in once we’ve identified the input variables and how they should work)
Help improve the ‘Matched Player skill’ system
Or scrap the whole thing and start again
Either way 2 heads are better than 1, and with a forum like this finding 2 (or more) heads to come up with a comprehensive solution to this vast and complex problem shouldn’t be too difficult.
It would also alleviate some stress, freeing you RL guys up to concentrate on other things like developing an awesome game with cool and inventive level designs (even though you’ve already outsourced some of this to the players anyway) and help speed development
I’ll leave opening the multitude of relevant forums for info, ideas and equation generation / gathering to you guys (unless you wish to add me to the payroll?

). We need to start with the small equations for the different factors and their corresponding variables and build the overall rank equation from there. Good luck ppl
I look forward to every1s input and further discussion(s).
Wishing everyone Many Happy Ninja-ing’s
z FruitBat z
(Com’on guys…. Sort it out

)