Texas new PR discussion thread.

[Sync.]

Ok guys with Dojo reviving our Brawl scene I think there should be a revival of our PR system as well.

To be honest, our PR system is not really suited for the size of our state. It shows an inaccurate portrayal of the true threats of our state. Since our PR is attendance based it can be quite inaccurate. Many retired players are still on our PR falsely informing those who lurk our PR to know our state better. Also, top players can't always show up to every tournament and other players can climb the PR (which is a good thing for them but an inaccurate portrayal of our scene and my goal here is accuracy).

In sight of this I have thought of a new system:

1. We should keep ELO for regions of Texas. The regions of Texas could be N,S,E and W Texas or something to that effect. It's a lot more effective to use ELO with a smaller amount of people. The ELO system Zac is using is really good and should be used for this (including his policy on OoS). To make the ELO more dynamic as well and make our scene more competitive (TBH we are pretty easy-going here in Texas) I think we should also include MMs above $5. Since people of the same region of Texas plays each other often these ELO rankings will be quite accurate.

2. Now for a Texas wide PR I think we should have a panel like before. Obviously the ELO region rankings will have weight in these discussions and should be used as good reference tools. I'm thinking there should be 2-3 representatives from each region of Texas in the panels and should be pretty credible players. I'm thinking we could use a point system/criteria for rankings having the ELO rankings count as part of them as well.

I think this would be a great new way to usher in the new generation and revival of Texas smash. Of course it us up for discussion in this thread and subject to change upon the community's request. I think the ELO will promote competition and the panel PR will accurately display the true threats of our state as well with tangible evidence to use in discussion and will not be so attendance based. I'm still not sure about deterioration factor or where to start with the ELO system or if we should just start anew.

Once again, everything is up for discussion and I would really appreciate it if people put thought into this new idea.

 

[Big Boss]

Honestly, I think it's a great idea. You've got California split into NorCal & SoCal, and Texas has a multitude of cities dishing out 240+ players total. It would also be easier on Zac with all of the updates on tournaments because each region will be responsible for their own updates and whatnot.

Just my 2 cents.

 

[Suspect]

dont give a ****

 

[TheAlpacaMan]

I think this is a great idea since Texas is basically 4 states anyway, so splitting it up just makes sense.

 

[NeoBatou]

So I guess doubles and LT tournaments still won't affect the PR?

 

[Maharba the Mystic]

we could do wat some other states do and have a doubles PR as well. although i don't think that would work with team handicaps but if there was a way that would be sweet

 

[Xyro77]

no panel ever. thats why it was corrupt in the first place.

 

[Espy Rose]

Technically, Texas could split up into FIVE states if we ever did, Alpaca.

Also, please don't count Money Matches. Have you not seen the drama that occurs in some of the EC PRs because of it?

To be honest, the only change I'd even consider, would be to separate the regions of Texas, but still use the ELO System. There's nothing else especially wrong with it outside of a few inactive players.

 

[UTDZac]

Sync, out of curiosity, if I remove (or somehow mark) the inactive players on the current ELO PR, how accurate do you think the list is, or has been? Has there ever been a time when you thought "Man, this list just isn't right." ?

I'm not turning down your idea. I'm merely promoting discussion on the topic.


Again for discussion: From my experience when i was on the panel and during the time I ran the elo pr:
- You don't count MMs
- You don't count side events (apparently not even doubles =/)
- You don't count pools (TOO many people sandbag)

It's all strictly tourney results and who beat who.

 

[Denti]

I like this idea because the ELO system is a real reference on skill level when creating the panel. It wouldn't just be 1 or 2 people's opinions it would be evidence on top of representing our state's threats. I think that the regions should be Houston, DFW, SA/Austin, Rio Grand, and the West/Northwest region(whatever their name is).


Also like zac and espy said, don't count MMs of any sort.

 

[iLight]

Elo system is a good idea, don't need a panel in my opinion, just split into seperate areas

 

[Fogo]

The panel was never corrupt, Xyro. However, with the elo system in place I don't feel like we really need one. I like the idea of splitting up, and the inactive players should be marked either way.

We could give this a test run at least, and if we don't like it, it should be an easy switch back to regular style PR.

 

[Dakpo]

No panel please (even though it would help my own placings)
I like Zac's system for the most part. I feel if we simply divided it up into 2 or 3 parts it would be good, i mean north texas has Dojo, me, Van Jones, and....69dude?
also i hate that losing to someone like Lee Martin would count as a 1600 thats just gay. I would rather you just not count them

 

[Player-4]

No panel please (even though it would help my own placings)

Whaa huh ?

 

[TheJerm]

I support this and would like to see where it goes. I do think their or some minor problems with the current system, but noting too bad.

 

[Reizilla]

Definitely no panel. I can't see anything good coming of it. Houston has their own PR, that's pretty much your East Texas one, SA has a PR, and it's fairly easy to see where the Austin players would fit into that. RGV is too small/inactive for anything, as of now. I have no idea about North and West Texas, but I can't see it being too much hassle or too far off from the current ELO system.

I agree with Espy that the only change I can see working is having a region split in the ELO system, instead of everything bunched together. (North, South, East, West, Central?)

 

[UTDZac]

I support this and would like to see where it goes. I do think their or some minor problems with the current system, but noting too bad.

Enlighten me please. If the problems do not affect the accuracy of the list, would you agree they are not problems?

I agree with Espy that the only change I can see working is having a region split in the ELO system, instead of everything bunched together. (North, South, East, West, Central?)

I can do both, and plan on it shortly.

 

[Nike.]

I like the ELO system currently being used for the top half of the TXPR.

S-Tier
Gnes - 2141.381
UltimateRazer - 2072.563
Dojo - 2011.447

A-Tier
Trela - 1967.580
Espy - 1964.395
UTDZac - 1881.160
K Prime - 1847.936
Jerm - 1846.843
Dphat - 1837.801
Denti - 1821.803
Shlike - 1807.043

B-Tier
Infinity - 1783.453 /
Illmatic - 1781.025
Dakpo - 1779.463
Magik - 1774.412 /
D4Ba - 1761.593
MegaFox - 1749.764 /
Ozz - 1747.169 /
RoyR - 1745.333
Poltergust - 1740.574
FK - 1733.741
Zori - 1728.864
DMG - 1721.245
Fliphop - 1717.040

That's pretty darn accurate. These players rarely lose to C-Tier or lower so their points are basically from against each other.

However, I feel like the list gets less accurate the lower you go.

C-Tier
Fogo - 1689.237
Xyro - 1689.025
Van Jones - 1683.268
Bwett - 1683.147
Nike - 1681.609
Broly - 1678.191
Pax - 1677.280
Andilex - 1674.932
Smoom - 1674.566
Sethlon - 1669.354
Morningstar - 1668.297
Melee1 - 1668.207
Dao - 1662.689
PhantomX - 1662.424
Sky Pirate - 1654.782
Bigfoot - 1654.349
PJ - 1651.863
Ice - 1651.235
Santi - 1650.199

I think this is the borderline area are far as accuracy. It's obviously up to pure opinion.

D-Tier
Chuky - 1649.558
Vorguen - 1649.154
Esca - 1647.754
TheBoredOne - 1646.514
Chuy - 1646.412
JustinKamikaze - 1643.665
Derk - 1642.133
CY - 1636.749
Light - 1635.854
Foursaken - 1630.304
Phil - 1625.009
TakeYourLife - 1624.525
Tyson - 1623.238
69dude - 1622.716
Jon - 1622.596
ALSM - 1620.513
PhoenixAlpha - 1618.421
Rei-gun - 1617.335
KirbyKid - 1616.475
RT - 1616.031

E-Tier
Guero - 1614.524
ArturitoBurrito - 1613.799
Bad Karma - 1613.162
Panty Raider - 1613.101
Chaosgriffin - 1612.399
Afiari - 1612.328
Slaps - 1611.582
Neon - 1610.033
PyleDriver - 1609.774
Wotts - 1608.899
Big Boss - 1608.759
KDawg - 1607.755
Panic - 1605.386
Awestin - 1604.817
Cake - 1604.637
Zeta - 1604.263
Avoid - 1602.571
Ramom - 1602.086
David - 1601.994
Toxik - 1601.999
Big Papi - 1601.783
Noc - 1601.385
Sammy - 1601.255
Suspect - 1600.809
Pnes - 1600.697
ChaoticNightmare - 1600.410

F-Tier

Spoiler

Shuzaku - 1598.531
Peachguard - 1597.598
Slash - 1597.135
Syoudous - 1596.983
Castro - 1596.773
KingJ - 1596.595
Teh Brettster - 1596.307
Mess - 1595.667
Brax - 1595.637
Esca Jr - 1595.412
Dillon - 1595.379
Mario - 1594.939
Paw - 1594.579
Everis - 1594.260
FuriousDuffman - 1593.727
Mumble - 1593.608
CxC - 1593.035
Zane - 1592.889
Mike2 - 1592.661
Viper - 1592.088
CBF - 1591.655
KRD - 1591.634
Ikey Poo - 1591.429
Deviation - 1591.249
610 - 1591.234
Clinton Smith - 1591.199
Sergio - 1591.196
josh - 1591.173
Carlos - 1590.800
Kalo - 1590.500
Panda - 1590.415
Loko - 1589.523
Jack - 1588.931
Caraan - 1588.537
Riza - 1588.056
Cheedoh - 1587.241
P4 - 1587.115
King Jacob - 1586.740
Nothx - 1586.699
Donkey Bong - 1586.620
She Said that - 1586.546
Khaoz - 1586.434
Tesh - 1586.233
Sonic King - 1585.014
JTD - 1584.839
Martin - 1584.760
MxT - 1583.829
Kilik - 1583.595
Turdox - 1583.506
Xero - 1583.508
Isaac - 1583.242
Zdax - 1582.918
Mike - 1582.914
M - 1582.715
Tree - 1582.617
ryan - 1582.518
Danny - 1582.477
Endrek - 1582.350
Sync - 1582.218
Ginko - 1582.207
CrossJeremiah - 1581.965
Matt - 1581.929
Chute - 1581.805
Anti - 1581.575
Slush - 1580.705
C Spyker - 1580.647
****r - 1580.287
**** - 1579.197
Kingdom Come - 1578.977
FHero - 1578.638
PsychoAce - 1578.336
Caleb - 1577.165
Sven - 1576.776
DDM3K - 1576.473
Bio - 1576.217
GreyFox - 1575.562
JS - 1575.045
Haptic - 1574.811
Gova - 1573.908
Dime - 1573.618
The Winged - 1573.568
Alpha - 1573.401
C - 1573.263
Zolo - 1573.219
KoN - 1573.183
Chic - 1573.034
Jew - 1572.953
Bombchu - 1572.821
Robert - 1572.447
RAM! - 1572.104
Candy - 1571.878
Red Shadow - 1571.577
Slate - 1571.188
ZtA - 1570.621
Sigma - 1570.417
Gamma - 1570.233
Bokkie - 1570.182
Riot - 1569.551
Cody - 1569.199
Tro - 1569.047
Kadin - 1568.692
JiveTurkey - 1568.158
Zigsta - 1566.881
Mikey - 1566.880
Baka - 1566.680
Raikou - 1566.364
Gator - 1566.193
Kates893 - 1566.170
Serious Cat - 1565.889
Rave - 1565.861
Echo - 1564.331
Turbo - 1563.503
Cloud - 1563.419
NPC - 1563.144
Mike3 - 1562.770
Ponyo - 1562.698
Mr.T - 1561.734
Rsn - 1561.681
Omega - 1561.361
Sandman - 1559.060
Shugo - 1558.856
Alejandro - 1558.771
Max10n - 1558.700
Chrispriter - 1558.445
Sama - 1555.930
Tallman - 1554.737
Tosm - 1553.634
JUJ - 1553.147
Graf_nik - 1552.309
RS - 1552.109
Since - 1550.992
JRUGS - 1548.656
Alex - 1547.004
Furbs - 1546.661
Kirby - 1544.034
Kyro - 1540.140
Soul - 1539.836
NeoBatou - 1537.397
Diego - 1536.769
Kovu - 1536.586
Problemo - 1535.619
Wingless - 1530.364
Leaf - 1523.161
Zook - 1519.946
Geoberos - 1515.223
Zero - 1514.007
Laurel - 1491.800
Tofer - 1491.561

This whole area is clearly inaccurate. There are alot of players in these sections that are so much better than what their rank is. But until they pull a major upset from the very top of the PR, they will never get out of this mess.

If I take out every player in San Antonio I would get this:

Espy - 1964.395
K Prime - 1847.936
MegaFox - 1749.764
Nike - 1681.609
Morningstar - 1668.297
Santi - 1650.199
PyleDriver - 1609.774
Big Boss - 1608.759
KDawg - 1607.755
Toxik - 1601.999
Big Papi - 1601.783
Suspect - 1600.809
Pnes - 1600.697
Castro - 1596.773
Brax - 1595.637
Paw - 1594.579
CBF - 1591.655
610 - 1591.234
Panda - 1590.415
Riza - 1588.056
Sonic King - 1585.014
Anti - 1581.575
Slush - 1580.705
GreyFox - 1575.562
KoN - 1573.183
Turbo - 1563.503
Shugo - 1558.856
Max10n - 1558.700
JUJ - 1553.147
JRUGS - 1548.656
Alex - 1547.004
Kyro - 1540.140
Soul - 1539.836
Zook - 1519.946
Geoberos - 1515.223
Zero - 1514.007
Laurel - 1491.800
Tofer - 1491.561

If you know anything about the SA brawl scene, you would know that this is a complete disgrace. Everything below top6 (Ironically where C-Tier ends) is so wrong it hurts. Players like Paw, Turbo, JRUGS, Slush, KoN, and Castro are all way better than their seed is. However, their stuck in their area because the players that they beat in bracket give them literally fractions of points. There's also multiple players higher up the list that are ironically the worst players in SA as of right now. Have you guys even heard of KDawg, Pnes, or Brax?

If you include Laurel, San Antonio makes up the 6 worst players in Texas. That is not accurate in my mind (troll bait).

If you look at RGV, they are greatly hindered by this same problem.

When people say "Use ELO and separate the whole list by regions," this is the problem you'll find. In case you've never known about it, San Antonio has it's own Power Rankings. We use a panel. Since everybody plays each other and go to the same tournies, we can make a very accurate list. If we use the ELO to represent our top10, KDawg and Toxik would be on there.

Now, some people will say "Nike, who cares about the bottom 3/4 of Texas, all that matters are the top players." Okay, then if that's the case then why even have a list of the bottom 3/4 of Texas if nobody cares about them?

C-Tier and above, I really like the ELO system. I don't think there's any reason to make changes for that part, outside of reorganizing it by region. However, if the goal is to represent everybody in Texas, something about the bottom half has to be done.

 

[TheJerm]

Well, most of us agreed that things like magik getting so high up there is wrong.. then he went to hobo and proved that was wrong. Depending on where you live, its alot easier or harder to gain points.

And people complain about how OoS players have 1600. But again, these are things that arent that bad and can be easily looked over, which is why I said minor problems.

 

[UTDZac]

Thanks for that Nike. I appreciate the feedback.

I would like to point out one thing about that reasoning (and im not saying you are wrong in anyway). I want to point out something you may not have realized:

The ELO system strives on head-to-head comparisons and not "what place did you get." Similarly, saying somebody is the 6th worst in TX doesn't hold any meaning using this ELO system. Instead it looks at comparisons between players. See chart below.

(Note the percentages are actually a bit higher with my system)

Now let's compare a few of the people from San Antonio:

Espy - 1964.395
Nike - 1681.609
Paw - 1594.579
Sonic King - 1585.014
Tofer - 1491.561

Compare Espy and Tofer using the chart.
Compare Espy and Nike using the chart.
Compare Paw with Tofer using the chart.
Compare Nike with Sonic King using the chart.

These feels a lot more accurate than if you said "Oh well Nike is 20th best in TX and Paw is 20th worst in TX." This is merely a reminder that the ELO system doesn't work with placings or ranking people by numbers (hence why i setup tiers), it works by comparisons.

To be honest, technically C tier, D tier, and E tier should all be as one (as well as some other changes to S, A, and B). I split them up to make it easier on the eyes.

 

[Ojih~Sama]

will this new system get me out of F - Tier?? o.o

 

[Zori]

I took the bait

 

[Thebest1pj]

zori should be higher

 

[z00ted]

K Prime should be lower (no offense), Zori should be higher, and Magik should be lower.

I'm not going to discuss myself.

When you enter C, D, E and F tier it gets a bit ridiculous because of what Nike has already stated.

Jerm was also right when he said "depending on where you live it may be harder/easier to get points".

I really like what we currently have right now - it just has a few problems, in my opinion.

 

[sheikamaru]

doooooooo it!

 

[KoNfucius]

the top 30 or so seems pretty accurate, its just the lower half like nike said. if we could help zac find a way to solve that problem then i would prefer to keep it the current method.

 

[UTDZac]

the top 30 or so seems pretty accurate, its just the lower half like nike said. if we could help zac find a way to solve that problem then i would prefer to keep it the current method.

I think people are skipping my post.

ELO system is not accurate because it tells you who is 1st, 2nd, 3rd and so on.

 

[z00ted]

Zori more than 50% of the time will beat K Prime

Zori more than 50% of the time will beat Magik

 

[Reizilla]

I'll put $20 on Magik over Zori more than 50% of the time.

 

[Sync.]

Sync, out of curiosity, if I remove (or somehow mark) the inactive players on the current ELO PR, how accurate do you think the list is, or has been? Has there ever been a time when you thought "Man, this list just isn't right." ?

I'm not turning down your idea. I'm merely promoting discussion on the topic.


Again for discussion: From my experience when i was on the panel and during the time I ran the elo pr:
- You don't count MMs
- You don't count side events (apparently not even doubles =/)
- You don't count pools (TOO many people sandbag)

It's all strictly tourney results and who beat who.

Ok I understand about MMs and side events and pools. That causes a lot of drama and innaccuracy (sandbagging and what not).

Yeah, like Nike said I think the bottom half is totally inaccurate but that's not the main problem I'm trying to correct. When I see the top half it's fairly accurate but the fact that people can climb to the top by just staying in their city and taking only their city's points I think is wrong and a wrong portrayal.

Also for if inactivates were taken out (I know it's kind of tough in the ELO system because you must always have a certain average of points) it would help with accuracy and not weigh people down who are potentially top 10.

The reason I moved for a panel was to project a view of all of our regions put together. Panels are opinionated but I thought using the ELO system for each region of Texas would help make it less opinionated by the panelists.

 

[Sync.]

I'll put $20 on Magik over Zori more than 50% of the time.

Please don't let the thread degenerate to stuff like that. We have tournaments to see the results of this.

 

[UTDZac]

Yeah, like Nike said I think the bottom half is totally inaccurate but that's not the main problem I'm trying to correct. When I see the top half it's fairly accurate but the fact that people can climb to the top by just staying in their city and taking only their city's points I think is wrong and a wrong portrayal.

This is a problem, and it isn't. You are correct in that some players may crawl up to the top like that, but that doesn't mean the system is inaccurate. As soon as those players participate in a single regional or w/e, they will be put against everyone else and their points will scale down (if they are bad) or scale up (if they are good).

Imagine we used a PR panel. Assume NO BODY ever left their city to travel and play against other people not in their city. How does the PR panel solve this dilemma? The instant players leave towns and fight others, the system adjusts the points and averages them out according.

Not sure if you knew this, but that's why I'm throwing it out there.

Also for if inactivates were taken out (I know it's kind of tough in the ELO system because you must always have a certain average of points) it would help with accuracy and not weigh people down who are potentially top 10.

Explain how this is a problem? Recall how the ELO system works. It's head-to-head comparisons. If Santi never plays in a tournament for 6 months his points will remain the same and everyone else is unaffected by him. He doesn't play so how can others be in head-to-head with him to trade points? Inactives don't influence the system in the slightest.

Again, I don't mean to say that your ideas are wrong or anyhting... I'm just trying to enlighten you on things you may not have known or may have missed.

 

[Sync.]

This is a problem, and it isn't. You are correct in that some players may crawl up to the top like that, but that doesn't mean the system is inaccurate. As soon as those players participate in a single regional or w/e, they will be put against everyone else and their points will scale down (if they are bad) or scale up (if they are good).

Imagine we used a PR panel. Assume NO BODY ever left their city to travel and play against other people not in their city. How does the PR panel solve this dilemma? The instant players leave towns and fight others, the system adjusts the points and averages them out according.

Not sure if you knew this, but that's why I'm throwing it out there.

Yes I know if people travel out and play in tournaments with other players it would adjust but not by much. It's not like tournaments were there is full representation always happen though and the Pr would adjust to how it was before. If I recall, the max someone can win is 32 points per match which probably won't impact the PR that much (for the upper levels at least the bottom levels it impact a lot).

Explain how this is a problem? Recall how the ELO system works. It's head-to-head comparisons. If Santi never plays in a tournament for 6 months his points will remain the same and everyone else is unaffected by him. He doesn't play so how can others be in head-to-head with him to trade points? Inactives don't influence the system in the slightest.

Again, I don't mean to say that your ideas are wrong or anyhting... I'm just trying to enlighten you on things you may not have known or may have missed.

The only way it's a problem is sometimes inactivates don't like staying inactive. They come to a tournament once in a blue moon and are marked as "active" again. The reason why I say it's a problem is because although we know they're inactive people viewing the PR WILL overlook it because the fact of the matter is people are pretty lazy and get the interpretation that someone from 08 is better than a current 2011 person is...

 

[UTDZac]

The only way it's a problem is sometimes inactivates don't like staying inactive. They come to a tournament once in a blue moon and are marked as "active" again. The reason why I say it's a problem is because although we know they're inactive people viewing the PR WILL overlook it because the fact of the matter is people are pretty lazy and get the interpretation that someone from 08 is better than a current 2011 person is...

I agree with you here. That is why I've taken the step to mark those players as "Retired" on the current TX PR. This should alleviate any confusion non-TX people have when viewing our extended PR list.

 

[Big Boss]

This is a problem, and it isn't. You are correct in that some players may crawl up to the top like that, but that doesn't mean the system is inaccurate. As soon as those players participate in a single regional or w/e, they will be put against everyone else and their points will scale down (if they are bad) or scale up (if they are good).

Imagine we used a PR panel. Assume NO BODY ever left their city to travel and play against other people not in their city. How does the PR panel solve this dilemma? The instant players leave towns and fight others, the system adjusts the points and averages them out according.

Not sure if you knew this, but that's why I'm throwing it out there.

After some sleep and reading this, I have to agree with Zac. Having multiple regions w/each having their own panelist of 2-3 people can get messy with regionals like HOBO.

 

[defsithe]

All that matters is I'm above cloud, at a game neither of us plays

 

[Magik0722]

Zori more than 50% of the time will beat K Prime

Zori more than 50% of the time will beat Magik

Pretty sure i beat zori in the last tournament we played

i might be mixing up my olimar players though.

 

[KoNfucius]

You want to move up the tier list win win win win!

Its called doing the megafox.

 

[Zori]

Magik we played like 2 years ago.... But let's not make this thread about me lol thanks for the hype guys

Magik pm me

 

[Magik0722]

Not entirely sure what you want me to pm you about.