Pick'em Leagues: THE BIG SHOW CASUAL BEST OF THE REST Single Event PvP: FANTASY POOLS Betting Leagues: THE BIG SHOW BEST OF THE REST

My input on how fight camps should be ranked

Print  
How should we rank fight camps?
3 different pages for Win%, fight end %, Average Team money 6 32%
One ranking page with a combined score out of 100 for each team 6 32%
Something else entirely 1 5%
I dont care! 1 5%
Rickson by arm-bar 5 26%
Posted By Message

calcifer

Learning to Sprawl

calcifer Avatar
 
 
 


 
 
Posts:20
Career:58-28
Joined:Jan 2007
Chips:
2
Ok so as promised, here is my input on how I think teams should be ranked.
-----------

Possible stats to base rankings off of:

Team win loss percentage - this is the most important thing, the ability to pick winners

Bonus points (ability to pick how the fight ends) - This is less important than picking the winner

total money - this is how good you are at picking bets. this is an interesting one to base rankings off of. however this can be manipulated by people making multiple accounts and transferring money between them. furthermore, it can be manipulated by donations purchasing more money. this one is also not a percentage like the other scores, so can continue increasing as the game goes on. we have to be careful how we use this in our rankings.

We should limit ranking to a fight camp with 3 or more people in it. This way, 1 person camps cannot dominate the rankings. besides it makes more sense to have a team be at least 3 people.

----------

so I'll propose a couple different ways to work the scoring system and we can have some discussion on what people think is best, or if anyone else has other ideas i'll be glad to give input on it. without further adieu:

Ranking Method #1: Team rankings are based only on team avg win loss percentage. Perhaps add a seperate ranking page for money, and 3rd page for team that is best at picking what round the fight ends in etc.

Ranking Method #2: Composite Weighted ranking out of 100. We combine all 3 of the scores from above and weight them, to get a number out of 100 that is your score.
Win-loss percentage is worth 50% of your score out of 100, bonus points are worth 25% of your score, and money is worth 25% of your score.
Example:
Team A has an average win percentage of 75%, they pick the way the fight ends correct 40% of the time, and the average money their team members have is $600, and the team with the best average amount of money per person in the whole league is $1000. Therefore:
(0.75*0.50 + 0.40*0.25 + (600/1000)*0.25)*100 = 62.5 out of 100.
This method makes sure that teams are looking for people that are good at picking winners, and it doesnt really matter when you joined, teams will still want you so long as your percentages are good and you have some money.

DISCUSS

Post #1   1/19/07 4:10:54PM   

titorican

MMA Regular

titorican Avatar
 
 
 


 
 
Posts:50
Career:819-516
Joined:Jan 2007
Camp: Ground'N'Pound
Chips:
2
I agree that obviously winning percentage is the most important. It would also be cool if camps were able to challenge each other for specific events. For example- Team Gronud n Pound challenges Sick Camp for the UFC 67 event. Whoever's winning percentage is higher for that single event wins, and the camp as a whole gets a victory and is now (1-0) and the loser gets a loss. This would be an interesting way to possibly develop Camp 'Champions'. If it's done with a ratings system then I think it should heavily depend on Winning percentage (maybe even 75-80%) and 20-25% on how the fight ends specifically. It would be nice to add in the 'money' part with the individual wagering but it does seem as thought here are too many loopholes there.

Post #2   1/19/07 6:00:16PM   

DoTheMMAth

MMA Sensei

DoTheMMAth Avatar
1
 
 
 


 
 
Posts:1,561
Career:1,439-884
Joined:Dec 2006
Chips:
355
Ok, here's some (quite) long-winded random thoughts on the formation of a camp ranking system as I begin Friday happy hour...

For now the camp directory is showing all the camps, even if it only has 1 member in it, so that people who are telling and inviting friends to the site can hopefully connect and get camp'd up without having to search around too much. As things get going we'll change this so that camps won't be displayed unless they have at least 2 members. Sorry Matt Hughes and all Godsmack-filled government army commercials... army of one just ain't a fight team.

If camp size and/or age doesn't affect overall scoring we could run into a potential situation like this down the road:

Camp "A" has been around for a few fight seasons on MMA Playground, they have lots of members and these guys (or girls, wishful thinking lol) as a whole really know the fighters and have a gift for picturing how two fight styles will mesh in the real world in a ring/octagon. Let's say they have an overall win percentage of 75%.

Camp "B" has just been formed. They have 2 members, who have been on the site for 1 event total. Let's pretend that this event is like UFC 67, where there's lots of fights that seem easily predictable (if you don't believe me look at the Picks Overview on the event stats page). The two members of camp "B" both manage to get 7 out of 8 bouts predicted right and come to have an 87.5 win percentage. Granted, over the course of numerous events, highly debatable fights with close odds will happen and the high scores and low scores will separate themselves more and more as a season progresses.

If I was team on camp "A" with a collective team record of 375/500 accurate picks over the course of both predictable and unpredictable events and I looked at the camp ranking page and saw team "B" ranked above me I'd be like "man, we got hosed!".


However, on the other hand you have the problem that you also want newer members to also have a shot at doing something notable here on the site and not having only the first to stumble accross the site to be the only ones at the top of the pile for good.


The money and wagering system wasn't initially going to be something included into determining overall competition rankings. Rather, it was sort of a fun "bragging rights" type side-game. This was done for a few reasons. If someone happens to run out of money we do not want them to make a secondary profile to start over on the bankroll ladder. Having two profiles means being able to make 2 separate sets of picks for every event, which is of concern when everyone else plays fair (with 1 account) and it is for a desirable prize (like the $500 being given away for the first season). As such, the solution was to make it so that (up to one time per season) people can add some money to their fantasy bankroll if they really mess things up or they just would like a larger number to work with. It costs $19.95 which helps to pay for server and bandwidth costs and provides an answer to the question "what if I screw up royally and bet all my money on a fight that goes sour?". This may sound a bit unfair at first considering that everyone starts with only $500, but considering that you earn a bonus fantasy $25 for every exact fight prediction you get right (both winner and result) combined with the fact that some people will be wagering and raking in the fantasy dollars, people will naturally clear $2,000 on their bankroll in the course of the first season or two, and from there will only be having potentially exponentially larger and larger sums of fantasy dollars appearing beneath their username on the forum as time passes.


As such, here's what me and Jim Beam have come up with as a potential solution for a camp ranking system :

Each fight season is 10 events long. To get full credit for their collective record a fight camp has to be in existence for a full fight season. As time progresses, a camp will gradually get more and more credit for their collective record up until they are a camp for a full 10 events, at which poitnt they receive full scoring credit. For example, after 1 event a fight camp with a 50 and 25 record would have a win percentage of 67%, however because they've only been in existence for 10% of a season they'd have a ranking score of 6.7%. After 2 events and being around for 20% of a season (pretending they keep the same win ratio) they'd be at 13.4%, etc, up until 10 events when they are just scored as a whole along with every other camp that's been around for 10 events or more.

That still leaves open the fact though that it is potentially more difficult to have a larger camp be as accurate as a small camp consisting of 2 utterly psychic members. For this we could rely on the solution proposed by n0xin found here and divide the camps up into figurative camp "weight" divisions so that camps of a similar size would be competing with camps of a similar size.

So, in a nutshell, the formula for any camp that hasn't been around for 10 events would be:

((camp total wins) / (camp totals wins + camp total losses)) * (# of events / 10)

And any camp that has been around for a full season's worth of activity would simply be :

(camp total wins) / ((camp totals wins + camp total losses))


This would also make people consider their camp selection a little more seriously as well if they really want to attempt making a fight camp that can compete at the highest levels.


Even with this solution though I can think of loopholes. For example, what if a person made a 1-man camp, waited a while, then had all the players with the best records join their camp. Automatic top ranked camp via armbar manipulation.

_______________________________________

Post #3   1/19/07 6:34:10PM   

DoTheMMAth

MMA Sensei

DoTheMMAth Avatar
1
 
 
 


 
 
Posts:1,561
Career:1,439-884
Joined:Dec 2006
Chips:
355

Posted by titorican

It would also be cool if camps were able to challenge each other for specific events. For example- Team Gronud n Pound challenges Sick Camp for the UFC 67 event. Whoever's winning percentage is higher for that single event wins, and the camp as a whole gets a victory and is now (1-0) and the loser gets a loss. This would be an interesting way to possibly develop Camp 'Champions'.



I think this is a cool concept too. What sort of rules could we make up to have a camp vs camp "turf war" type competition system? Worth some thought... but I'm logging for a bit right now. BBL!

_______________________________________

Post #4   1/19/07 7:03:39PM   

calcifer

Learning to Sprawl

calcifer Avatar
 
 
 


 
 
Posts:20
Career:58-28
Joined:Jan 2007
Chips:
2

Posted by DoTheMMAth

For now the camp directory is showing all the camps, even if it only has 1 member in it, so that people who are telling and inviting friends to the site can hopefully connect and get camp'd up without having to search around too much. As things get going we'll change this so that camps won't be displayed unless they have at least 2 members. Sorry Matt Hughes and all Godsmack-filled government army commercials... army of one just ain't a fight team.



Sounds reasonable, maybe just make a page with all camps alphabetically for those trying to put camps together


Posted by DoTheMMAth
As such, here's what me and Jim Beam have come up with as a potential solution for a camp ranking system :

Each fight season is 10 events long. To get full credit for their collective record a fight camp has to be in existence for a full fight season. As time progresses, a camp will gradually get more and more credit for their collective record up until they are a camp for a full 10 events, at which poitnt they receive full scoring credit. For example, after 1 event a fight camp with a 50 and 25 record would have a win percentage of 67%, however because they've only been in existence for 10% of a season they'd have a ranking score of 6.7%. After 2 events and being around for 20% of a season (pretending they keep the same win ratio) they'd be at 13.4%, etc, up until 10 events when they are just scored as a whole along with every other camp that's been around for 10 events or more.


having a 7% ranking when everyone else has a 70% ranking is basically the same thing as not being ranked at all... i would suggest a modifier if a team is in a group with a small sample size (in other words, one with few events and team mates). for example, a reasonable modifier is a 10% deduction for each event less than 4 a team has. So a team with 60% win loss rate, with only 2 events, would end up with a ranking score of: (60*0.8) = 48%. still compettitive enough that new teams care about rankings, but it takes away the possibility of a one event team topping charts.
however, the number of players thing, i think the modifier should be different - i'd say that the percent should be modified down by 3% for each person less than 6 in the team. So, a 3 person team would suffer a 9% penalty, while any team with 6 people or more would be on even playing field. of course the numbers here would probably need to be tweaked, but i think you see what i'm getting at.



Posted by DoTheMMAth
Even with this solution though I can think of loopholes. For example, what if a person made a 1-man camp, waited a while, then had all the players with the best records join their camp. Automatic top ranked camp via armbar manipulation.



Well I think this should be the whole point of the fight camp system - eventually, 2 or 3 events down the line, some players will stand out as people that really know how to pick them. someone puts together an elite fight camp, gets the best to join, and is at the top of the ranks because they are the best at picking fights. why not? makes sense to me.

Post #5   1/19/07 7:27:23PM   

titorican

MMA Regular

titorican Avatar
 
 
 


 
 
Posts:50
Career:819-516
Joined:Jan 2007
Camp: Ground'N'Pound
Chips:
2
We could start with some sort of Tournament. Where all the fight camps randomly get paired up with each other. This way, say there are 32 camps involved, by the 5th event we will have one 5-0 team who will then be crowned heavyweight Champ. For members who come in late, we can then develop the LHW tournament next, Middleweight after that and so on and so forth, with the HW title being most prestigious. The champs can then defend their titles against the top contenders in each division (weight classes can be divided by Camps experience in the beginning but as teams get better they advance weight classes and say a HW team is performing poorly with the big boys they can be demoted to a lighter class). I don't know if what I wrote makes sense but I'm posting it anyway.

Post #6   1/20/07 5:13:37AM   

blakeon

MMA Regular

blakeon Avatar
5




 
 
 


 
 
Posts:200
Career:629-344
Joined:Jul 2007
Chips:
14
i agree

Post #7   7/23/07 4:01:23PM