You are here: CommFT

CommunityFortress

Understanding the XPL Ranking System

When looking at the XPL TF2 Standings page, a few different columns stand out, specifically Points and RPI.  What is unclear initially is which of those two factors is more important, and how each is calculated.  This article will aim to give players a greater understanding of exactly how teams are ranked in the XPL and how those rankings themselves are determined.

The points system in XPL is relatively simplistic.  Teams are awarded three points for each win, two points for a tie or a forfeit win, and one point for either a loss or a forfeit loss.  Remember that, since each map played technically counts as an individual match, teams can score a maximum of six points each week and a minimum of two.  However, the points system in XPL is largely irrelevant.  Teams are not ranked according to their points when it comes to playoff standings, and points appear to be used as little more than a modifier for a team's overall RPI.

Much more complicated is the Ratings Percentage Index, or RPI.  This is the number that every team's playoff aspirations will hinge on, as it is the number that decides where teams lie in the rankings.

Ratings Percentage Index is the system that is used to rank teams in men's NCAA basketball, specifically to aid in the selection of teams for the March Madness 65-team tournament.  The primary factor in RPI rankings is strength of schedule, and while this obviously gives potentially undue rewards to teams that play in tougher conferences or divisions, it also minimizes the ability of teams to run up match scores like they could and probably would if the primary ranking factor was rounds won and rounds lost.

Currently, the index is comprised of a team's winning percentage (25%), with seventy-five percent of the rating being based on the team's strength of schedule.  Of that seventy-five percent, two thirds are composed of the winning percentages of the team's opponents, with the final third allowing for the winning percentage of their opponents' opponents. 

For a more visual representation:
RPI = (WP * 0.25) + (OWP * 0.50) + (OOWP * 0.25)

Obviously, XPL's system isn't exactly the same as that system described above.  Rather, XPL uses its own proprietary formula that also takes into account game-specific factors such as rounds won and rounds lost divided by total possible rounds and point total divided by total matches played.  Strength of schedule plays a large part of the team's overall RPI in XPL, as does the team's winning percentage.  The other factors listed above are just further modifiers that make XPL's system a bit more unique.

As complicated as all of the above sounds, the most important factor is still winning games.  Amassing wins will go a long way toward raising your team's RPI and staying at the top of the leaderboards.  XPL's system merely gives teams a larger reward if they are forced to play tougher schedules while giving less to teams that are handed cream puff schedules.  The ultimate goal, of course, is to provide the fairest system possible for all teams involved, and from the information I've garnered, XPL is doing the job as well as or better than any other league presently.

Hopefully this article has helped anyone who was curious gain a greater understanding of exactly how the RPI system works in XPL.

Last Updated ( Wednesday, 30 June 2010 00:04 )  

Comments  

 
#3 Hawkeye 2010-06-30 04:19 Quoting brainpower4:
So to be clear, if your conference members are all either bad or evenly matched, so that most of them have a large number of ties and losses, even if you win every game, you aren't guaranteed to place at or near the top? Since their opponent's skill level is 75% of the overall ranking, even CoL would only seem decent playing against low teams compared to a low/mid team that managed to squeeze in a few wins against people better than them.


What your seeing in the RPI values right now is basically junk, For example the BCS (ranking and indexing of College Football) does not even post the numbers until 6 or 7 weeks into the season. So some teams currently are overly inflated, while others are under-inflated.

Your wins/losses does infact count for a large percentage, but the RF/RA is what divides the teams apart, and those are weighted based on your opponents skill levels, so at the end of the day, two 16-0 teams will be close, but one that survived the stronger teams will have a higher RPI value then one that went 16-0 against the lower tier teams.

FFW as noted, are weighted slightly different, as the modifier is 3 as opposed to 2 for an unchallenged Round, as can be seen by the numbers currently.

Also, we are taking 16 teams into the playoffs, so at the end of the day, 1 in 5 will be going. Plus the way we setup the schedule should allow all the teams to be weighted pretty well.
 
 
#2 improperdancing 2010-06-29 22:14 Quoting brainpower4:
So to be clear, if your conference members are all either bad or evenly matched, so that most of them have a large number of ties and losses, even if you win every game, you aren't guaranteed to place at or near the top? Since their opponent's skill level is 75% of the overall ranking, even CoL would only seem decent playing against low teams compared to a low/mid team that managed to squeeze in a few wins against people better than them.


Well, as I said, XPL uses a modified version of the college basketball RPI system, and an exact formula was not available to me, so I can't be certain if the records of your opponents take up 75% of your overall ranking.

However, one thing to consider is that the numbers have no idea if you're playing coL or some team that formed a week ago. All they care about is your opponent's win/loss record and the win/loss records of their opponents. Even if you're in a weak subdivision, you will inevitably play teams that have decent win/loss records, and thus they will appear to be better opponents than they may actually be.

Further, since teams are supposedly arranged according to skill (at least as much as possible given that there are a lot of new teams), if you're a playoff-caliber team you will likely be playing primarily teams that are of playoff-caliber skill.

And like I said, wins are still wins, and they make up a large chunk of your RPI on XPL. RPI was simply designed to make strength of schedule a significant factor in the rankings of teams. If you win all eight of your matches, I can't see any way you wouldn't make the playoffs unless sixteen teams also won all eight games and did it against tougher teams than yours, which is a very unlikely scenario.

And for the record, I have nothing to do with XPL, but am just relaying the information I have gathered on the subject. I'm sure Hawkeye can answer a lot of these questions better than I can, as he was my primary source when writing this article.
 
 
#1 brainpower4 2010-06-29 21:52 So to be clear, if your conference members are all either bad or evenly matched, so that most of them have a large number of ties and losses, even if you win every game, you aren't guaranteed to place at or near the top? Since their opponent's skill level is 75% of the overall ranking, even CoL would only seem decent playing against low teams compared to a low/mid team that managed to squeeze in a few wins against people better than them.
 

Game Server Provider

Sponsors



Sponsor Us!

Sponsor Us!

Yaooooooooooo

Top 10 Plays - December

Ads