349 replies to this topic

[Jarl Dane]

Scuro the Black-Shot of the Isengrim created this website for all of you! It is fully automated to parse and update the leaderboard from each season (monthly)

To make the leader board more competitive and more accurate Scuro has set the weight of the last three seasons to have five times the impact on the score as well as (with Tarogato's help!) a formula to reduce your scores if you have less than 350 games.

The leader board will be an ongoing project and Scuro plans to have leader boards for each weight class as well as a score system that makes the leader board more competitive.

Credit is also given to Chimmy for helping create the CSS code!

ENJOY:

http://leaderboard.isengrim.org/

[Angel of Annihilation]

Now give us a filter that removes all players less than 100 games then give us a ranking percentage.

As it stands now I am ranked 4498 out of 139,790 players which means I would rank roughly in the top 3% of all players in this game. Honestly I just do not think I am that good of play, not in my wildest dreams. I think that is being skewed by the probably thousands of players who haven't even broke 100 matches let along experienced players that have played at least 1000 matches.

[Thorqemada]

Nice and quick...

[Champion of Khorne Lord of Blood]

Still looking for a leaderboard that makes a distinction between solo and group queue.

[MischiefSC]

Viktor Drake, on 06 September 2017 - 05:51 PM, said:

Now give us a filter that removes all players less than 100 games then give us a ranking percentage.

As it stands now I am ranked 4498 out of 139,790 players which means I would rank roughly in the top 3% of all players in this game. Honestly I just do not think I am that good of play, not in my wildest dreams. I think that is being skewed by the probably thousands of players who haven't even broke 100 matches let along experienced players that have played at least 1000 matches.

Don't sell yourself short. Human skill curve is actually more like a cliff. The vast, vast majority of people suck on ice. Probably 18 out of 24 people in a T1 match play lime they're using a joystick (or may actually be doing so) or are in builds that should probably have a clown nose on them.

So the top 2-10% are all shockingly close to each other in skill and while better than the wide brown sea of 4xlrm deck, I only play lore builds, my Atlas is great with an XL hordes of window licking professional terribads the top 1% has a larger skill gap between them and the 2-10s than the 2-10s have over the 20-80s.

You're a good player. You know what a good build is, you know not to go off alone and if someone gets out of position in front of you then you can punish them for it. You've got a good command of the game and when I drop with you in a match i can hang close to you and expect to be in a reasonable position and that you won't be lurming rocks or wandering out of bounds.

A good 80% of the games population, in T1, doesn't deserve that trust.

However you see the 1% folks and watch them put up 2k with 2 raw potatoes and a streak 2 and think "my god, I suck."

No, you're a solid performer and if you played 40 hours a week on TS with a top tier team you'd likely develop 1% skills.

It's no coincidence that stuff like this comes from people like Isengrim - they put in the time and energy to cultivate the talent to perform.

So, yeah. You're probably accurately around the 3% mark. Just that the next 2 % requires human sacrifice, an energy drink IV and a lot of human desocialization to bridge.

[Y E O N N E]

MischiefSC, on 06 September 2017 - 06:15 PM, said:

So, yeah. You're probably accurately around the 3% mark. Just that the next 2 % requires human sacrifice, an energy drink IV and a lot of human desocialization to bridge.

Hey...I only did one of those things!

(Rank 1789)

[El Bandito]

I'm rank 139. Hey, not bad at all!

Dakota1000, on 06 September 2017 - 06:12 PM, said:

Still looking for a leaderboard that makes a distinction between solo and group queue.

I agree.

Viktor Drake, on 06 September 2017 - 05:51 PM, said:

Now give us a filter that removes all players less than 100 games then give us a ranking percentage.

This too.

Edited by El Bandito, 06 September 2017 - 06:42 PM.

[Nik Reaper]

As someone who does well in slow lights and mediums it kinda seems that the extra weight on those scores kinda skewers those players up the ranking.

I really don't feel like the 1760-th player of this here game...

Also, you know we want it, lets get some graphs going to garner wide appeal to the topic! .

Edited by Nik Reaper, 06 September 2017 - 06:44 PM.

[Snowbluff]

ONly 17000 D:
Althought my K/D and W/L are doing favorably. Not great but better than average. This is what I get for playing lights and mediums. Compare to Senor Bandito who is heavily heavy.

WTB rewarding gameplay for lights.

EDIT: Looking at my mech stats I wonder why I spend so much time in mechs I'm bad with.

Edited by Snowbluff, 06 September 2017 - 07:01 PM.

[Scurro]

Viktor Drake, on 06 September 2017 - 05:51 PM, said:

Now give us a filter that removes all players less than 100 games then give us a ranking percentage.

There is already a very hefty reduction in points for having games less than 300. I use this formula for my adjusted score: (Last three seasons have 5 times the weight for average)*(-(1.015^(-GamesPlayed)-1))

[Nameless King]

Very interesting

[chucklesMuch]

A leaderboard with actual good players filling the top spots... what kind of space magic is this.




Nice work!!!

[SFC174]

Hmm, maybe PGI could take a hint from this in revamping the matchmaker......

I ran into a huge streak of horrible losses today (like 9 in a row with 3 12-0 stomps). Comparing the average player rank (for the entire team) and adjusted average match scores for the winners vs. the losers showed a substantial difference with the winners having a 10-15% higher score average......

Nice work.

[Tarogato]

SFC174, on 06 September 2017 - 09:54 PM, said:

Hmm, maybe PGI could take a hint from this in revamping the matchmaker......

I ran into a huge streak of horrible losses today (like 9 in a row with 3 12-0 stomps). Comparing the average player rank (for the entire team) and adjusted average match scores for the winners vs. the losers showed a substantial difference with the winners having a 10-15% higher score average......

Nice work.

If you like looking into things like that, then you'll enjoy this post I made a while back: https://mwomercs.com...is-of-the-12-0/

[SFC174]

Tarogato, on 06 September 2017 - 09:57 PM, said:

If you like looking into things like that, then you'll enjoy this post I made a while back: https://mwomercs.com...is-of-the-12-0/

Yes, I read that thread with great enthusiasm when it was fresh. I found it very interesting and, while I agree that statistically more data was needed, I thought the criticisms were short sighted (only seeking to tear it down instead of trying to gather more info). I only started pulling data on the last 5 matches I lost today before I logged off in disgust, but the winning teams averaged a match score of 257 vs. 226 for my losing teams (this season only). Interestingly enough, the winning teams also had at least one player with a substantially higher avg match score than anyone on my losing team. In one match the winners had 2 400+ match score players while my team had none above 330 (and a lower average to boot). If you showed 100 players those numbers before the match started, I'm pretty sure 99 of them would have predicted my team would lose.

Edited by SFC174, 06 September 2017 - 10:47 PM.

[Eindridi]

If only this game had a legitimate competitive outlet as opposed to ranking based on how often people queue in geometrically correct laser vomit heavies and assaults, and to a lesser extent Clan versus IS.

An active tournament scene would reward players who improve because they are willing to take risks with playstyles and (possibly) builds. Leaderboard is just a cookie for people with the emotional stamina to play all day in a BLR or EBJ.

PS: I did not mean to exclude LRM assaults from my opening statements. Having further browsed the index and found a player who runs nothing but LRM SNV and TBR builds in the top 10k, I admit my mistake.

[meteorol]

Great effort, gratz on the outcome!

Can't be too accurate in ranking top tier players though, since i'm on 28.

And i'm a filthy casual who plays a few matches a week in solo queue, most of them pretty alcoholised. If i even get to play at all.
My stats add up to something like 1.6 matches per day, which comes down to spending like an average of 10 minutes per day on this game. I don't think this little investment does make me one of the games best players. I'm far from it.

But in the end, this is not this leaderboards fault. You guys can only work with the data PGI delivers.

[Davegt27]

I don't see this guy he says he is the best

edit: vid removed for COC

Edited by Davegt27, 05 June 2018 - 05:29 PM.

[Champion of Khorne Lord of Blood]

Davegt27, on 07 September 2017 - 12:43 AM, said:

I don't see this guy he says he is the best

That feel when he can pilot a Locust right infront of people and not have them do anymore than wave lasers around near him while I get focused by the enemy LRM boat whenever I bring an Urbanmech.

Spoiler

Locust PB is the god of light mechs though if you're looking to do some shenanigans like this.

[Fox With A Shotgun]

:/

How is a Sh*tfox at #229? Is there too much weight being given to the latest few seasons post-Skilltree?

Must be the 1% of games that I played in my Fatfox (direwhale).

Also lol.
98% 1% 0% 2%
Lights-mediums-heavies-assaults. 98 + 1 + 0 + 2 = 101%. Dat rounding.

Also lol, my alt is at #4. Lights OP, amirite?

Edited by Fox With A Shotgun, 07 September 2017 - 02:15 AM.