Loading...
  OR  Zero-K Name:    Password:   

Ladder parking

12 posts, 204 views
Post comment
Filter:    Player:  
sort
Over a year ago, there was a period of time where highly ranked players would drop by for one game on the Matchmaking Ladder to re-cement their position display, without then going on to play more games. This behaviour led to a phenomenon where the most active players vying for advancement would fight amongst themselves indefinitely without the opportunity to climb, because they were merely passing rating back and forth despite any skill growth gained from the sessions.

At the time, I wanted to imagine a change that would solve this problem and incentize more activity. After much consideration, this is my suggestion:

Instead of gaining 1 month of visibility per game played: Winning a game nets the player 1 additional day of ladder visibility, and losing nets 2 additional days (a consolation boon). The period of visibility on the ladder is capped to 15 days banked. (Numbers provided are 'feel based' and could be changed to address arising concerns)

With this system, title holders are incentivized to risk their heavily banked rating on staying visible. The competing bracket then either gain a chance at those spoils or are advanced by default of inactivity on the ladder parking player's part.

As a natural side effect, the Matchmaking Ladder would have much greater temporal accuracy and provide a better read of who's active and up and coming.

Any thoughts or context I'm missing with this idea?
+8 / -1
29 hours ago
nice idea.. ill add my avitar to your cause

but i would go even further and make the elo degradation for absence more agressive
+2 / -0
26 hours ago
Maybe such a system would be beneficial also for the Casual Ladder, although probably the parameters could be different there.
+0 / -0
18 hours ago
I don't have time to play that much. How much I play depends more on my real life than on a visibility incentive anyway. Knowing that I cannot play often enough to stay on the ladder consistently would motivate me even less.

Half of the current top50 would vanish despite being active. The ladder would become completely inaccurate, showing only a fraction of the players.
+0 / -0
The exact numbers could be tweaked but I definitely think you should need to remain active to remain on the ladder for an extended period; the only mechanism by which inactive players have been purged in the past year has pretty much just been CNrankXNTWSAD accidentally cooking the server every few weeks by uploading malformed mods (thanks for your service keeping the ladder fresh in that way!).
+0 / -0
I think it would be a good time to consider implementing seasonal ranking resets. For example every 3 months all players have their elo reset. Gives time for players to build it back up and also provide a more accurate list for active players.
+2 / -0

9 hours ago
we could enable a scheduler for CNrankXNTWSAD to take action.
+0 / -0
quote:
Half of the current top50 would vanish despite being active. The ladder would become completely inaccurate, showing only a fraction of the players.


Is this in reference to the casual ladder? If so keep in mind that my suggestion is for the Matchmaking ladder.

I'd posit that the casual ladder is already, on some level, quite inaccurate and nebulous due to the wide array of games and game sizes that it encompasses. My own casual rating is a prime example of this issue - I really have no talent or taste for large teamgames, but have a massively inflated casual rating due to FFA successes. I don't feel comfortable or as though I deserve to retain my spot on that ladder as it is, but I don't really have a lot of incentive or opportunities to go back to FFA. Should I, like others, forever keep other more experienced casual ladder players down as a result of not passing off my rating and taking their rightful spot?

As USrankStuart98 rightfully clarifies, the numbers could so easily be adjusted to something that would feel right.
+0 / -0
I think ladder parking is something that naturally happens due to busy life and other distractions but it's also a symptom other issues with the current system:

A- player's ladder position/achivement vanishes if they grind then stop playing for a month

B- (iirc, may be wrong) mm rating references shift over time, sometimes to lower values than the player had, so being very active fighting evenly skilled players will lower their rating over time, leading to avoidance (they only experience the drop IF they play)


to mitigate issue A, player's profile could show something about player's past achievements
- year, month and best rank color, top% the player achieved, ever
(reward intense participation at some point)

- year, month and best rank color, top% the player achieved and was able to consistently hold for 3 months, 6 months, a year?
(reward recurrence/consistency)


to mitigate issue B (if true), when ladder references shift at the end of each month or whatever, shift player's scores immediately, and/or do explicit MM rating rescalings or resets, that way people won't feel punished when they eventually get back to playing.
+0 / -0


4 hours ago
quote:
B- (iirc, may be wrong) mm rating references shift over time, sometimes to lower values than the player had, so being very active fighting evenly skilled players will lower their rating over time, leading to avoidance (they only experience the drop IF they play)


Check this out.



USrankUltraGodzilla got done dirty by that shift over time. I don't even get it.
+0 / -0

3 hours ago
Apparently the graph shows current rating, not ladder rating. I think current rating can be rescaled and adjusted over time without playing, but the graph only shows points that coincide with days where they played at least one battle.

Maybe it should show vertical lines highlighting rescaling events and points with the corresponding adjustment.
+0 / -0
quote:
USrankUltraGodzilla got done dirty by that shift over time. I don't even get it.

The longer you go without playing, the higher your uncertainty is, which means the change in rating from subsequent games will be very high. Ultragodzilla went 3.5 years without any games so had very high uncertainty, so when he went on a losing streak after returning that caused his rating to go down dramatically. If he had won those games, his rating would be significantly higher than it used to be (see DErankHoppili's competitive rating for an example of that).
+0 / -0