Best allround players

[LortGob]

    Since I was kind of bored at work, I decided to put all the new 1v1, 2v2 random and 2v2 teams in an Excel sheet and tried to figure out who would be in the top 10 best players if you combine all the categories.

What I did is look at the rank of a player, and then award him/her with a maximum of 25 points for 1v1, max 10 points for 2v2 random and max 5 points for each 2v2 team ranked. As an example, the #1 in 1v1 gets 25 points, #2 gets 24,75 points and #3 gets 24,50 points.

Of course the points are arbitrary (and open for discussion!) but they feel about right and do not favour either 1v1 or 2v2 (2v2 has fewer players so I have the feeling this would be easier to achieve though).

Any suggestions would be welcome

EDIT:
Copied it into Google Drive for easy viewing:
https://docs.google.com/spreadsheet/pub?...utput=html

[alexjiang1]

^Some people that play 1v1 don't play as much 2v2 and vice versa. So the list MIGHT be a tad bit off. But overall, great idea!

[Necrocat219]

Looks interesting and its a really good idea but as mentioned I think that you can definitely score better, you've weighted 1vs1 games as being directly being 5 times better than someone with only one 2vs2 team - not many people have the time to have 5 different partners ^^''. Also not many people have ranked in 2vs2 random so far (although I suspect that will change in the next few weeks) Here are a few propositions to improve it:

-give points equal to the place, 1 point for 1st, 2 for 2nd, etc. The lowest points gets highest rank and so on. This just follows what we already have and will make it easier to rank too - you might even be able to just simply export the top 100's to excel and data sort to make it easier (maybe)
-do an avarage of 1vs1, 2vs2 and 2vs2 random treating them equally - all people will value each have preferences so that would just suit to them better.
-Choose only the best 2vs2 team ranking OR an avarage of all their teams. I would choose the first as it wouldn't affect the top 2vs2 players from choosing to play with players who might be slightly worse.
-Only include players that have taken part in all 3 gamemodes (after a few weeks). This would probably be the only rule that I personally think is debatable but the top list for 2vs3 random should encourage players to take part - in a month or so more players would rise into it as they are encouraged to play.

Also I know everything I suggested would take more effort sorry about that - I can make a sample table later?

[LortGob]

(03-06-2013 11:39 PM)alexjiang1 Wrote:  ^Some people that play 1v1 don't play as much 2v2 and vice versa. So the list MIGHT be a tad bit off. But overall, great idea!

Hehehe. Of course it's off. Couple of examples that stand out:

# 20 - dowkstar (has 4 teams ranked in 2v2, giving him a lot of extra points)
# 49 - Dharma (doesn't play 1v1, highest ranked player outside 1v1 top 200 and only one in the top100)
#246 - Gurleyman & krogoth300 (both play 2v2 with 1 team exclusively)
#250 - alexjiang1 & Mr. Banger (both play 2v2 with 1 team exclusively)

[LortGob]

Thanks for the feedback Necrocat!

Here's some thoughts:
- I agree to your first comment: just giving rounded points would probably be easier. Let me work something out
- That's a valid point. I weighted 1v1 more heavily since the amount of players in 1v1 is so much higher (there are less than 1500 teams according to OSN). It must be more difficult to get in the 1v1 toplist, so I feel that should be awarded.
- I agree the current system is now skewed towards having as many 2v2 teams in the list as possible. However, getting one team at #1 in 2v2 probably requires more wins than getting three teams in the top #200. I think it would be best to award the highest ranked team more points than subsequent teams. This would bump the two pure 2v2 teams mentioned above.
- I would still like to include everyone, since, like you said, it's a player preference to either play a mode or not.

[lawtai]

Maybe not weight 1v1 as much? 15 for 1v1, 10 for 2v2 Random, 5 for 2v2?

I do think 2v2 random should be weight slightly more than having a known partner.

[LortGob]

I've played around with the numbers. Interestingly enough, the top 6 stay the same.

Max Points
1v1 25 (was 25)
2v2 Random 12,5 (was 10)
2v2 Team1 12,5 (was 5)
2v2 Team2 7,5 (was 5)
2v2 Team3 5 (was 5)
2v2 Team4 2,5 (was 5 - only 2 players apply)

This new point system favours:
- 2v2 play in general
- 2v2 players with only 1 partner

Significant changes in rank due to new point distribution:
- s4mbita (was #24, now #15) - 2 high ranked 2v2 teams
- Dharma (was #49, now #25) - 3 high ranked 2v2 teams and high ranked 2v2 random
- Gurleyman, krogoth300, Mr. Banger, alexjiang1 - bumped from #250s to #140s.

Looking at this, focusing on 1 team and be very high ranked nets you #140, while playing only 1v1 with very high rank (like joelduque and pokhs) net you around rank #50.

Hmmmm..

---

LOL whatever I do, the top 5 stays the same

Current setup:

Max Points
1v1 17,5
2v2 Random 12,5
2v2 Team1 10
2v2 Team2 7,5
2v2 Team3 5
2v2 Team4 2,5

This is most beneficial to Dharma, who get's ramped up to #6.
Other highlights of this setup:
- Highest rank based on two components is #17 (burnorod, 1v1 and one 1v1 team)
- Highest rank based on one component is #55 (joelduque, 1v1)
- Highest ranked only ranked in 2v2 random is #108 (austinoconner)
- Highest ranked only ranked in 2v2 teams is #68 (WorldFamous, three teams)

[Necrocat219]

Aww so the majority don't get to be on?

[LortGob]

(03-07-2013 01:11 AM)Necrocat219 Wrote:  Aww so the majority don't get to be on?

You'd be ranked #79, which is lower than your 1v1 rank but higher than your 2v2 rank. Sounds about right

I'm keeping those numbers for now:
https://docs.google.com/spreadsheet/ccc?...6aFE#gid=0

[John3:16]

Although too difficult (and impossible given the incomplete data) to factor in, best all-around would take into consideration if the person used different teams, right?