The Halo 3 system for giving ranks in online ranked matches has dumbfounded me for quite some time now. If you are like most people (and me), you haven’t a clue how EXACTLY this ranking system works. And perhaps no one will ever know precisely how it works except for the bungie staff and a handful of computers. However, we can speculate and hypothesize exactly how it works.

Most of us have won three matches in a row and not gone up a rank, or lost 1 match for the first time in a winning streak of 5 and gone down. I don’t know about you, but I get rather upset when these things happen. After being utterly owned by the ranking system a few times, I decided to start collecting some data and doing a little bit of research.

Halo 3 uses a system which is commonly referred to as “TrueSkill”. This system is not linear as it was in Halo 2, which is the source of all this frustration. This system takes into account factors such as the skill level of your opponents, your total matches played in a certain ranked game type, and what kind of game you are playing.

Based on my own experience (In Lone Wolves), how drastically your level is effected by a win, loss, or otherwise is directly based on the skill level of your opponents. This means that if you are a level 10 and you beat a level 20, your skill will be effected drastically, in that you would go up more than if you beat another level 10 or a level 11. Likewise, if you were the level 20 that got beaten, you would go down more so than you would if you were playing others close to your level. Also, if you are playing with others that are a lower level than you, your skill is minimally effected as this means that you are correct for being at your current skill level/rank. This is not to say that beating people your level isn’t good, because it is very much so.

In free-for-all playlists, there is obviously more than just winning or losing. I’ve done some experimenting and therein drawn some conclusions.

• Getting first or second place is always good for your rank, no matter what. Likewise, getting fourth or fifth place is always bad for your rank.
• Getting third place, however, is slightly more difficult to understand. I personally have gone up a rank from getting in third place, and gone down a rank from getting in third place. This place is ENTIRELY dependent on the ranks of the other players in the match in my opinion. If you get in third place in a match where most players are a higher level than you, then third place is good, and the same goes for the corollary of this.
• Also, I am fairly certain that there is an outlier failsafe. This means simply that if one person in the match completely owns and the final score is something like 25-12, then, for one the person with 25 will get effected very well, and for two, the person in second place will be treated as if they got first place and everything will be calculated with the score of the person in first excluded (meaning second place is like first, third place is like second, and so on).

Team playlists add a slightly different dynamic to everything. They take into account how well you individually did in a match as compared to others and the levels (or possibly average of the levels) of the opposing team as it compares to yours. As there is no gray area between a win and a loss in team playlists, I think that the system takes into consideration how well your team did. This means that losing 49-50 would not reflect nearly as badly on your score as losing 20-50.

One other thing that I have noticed is the effect of “Novelty” gametypes on your rank. These are gametypes such as swords, snipers, king of the hill, oddball, shotty snipers, etc. Overall these seem to have less of an impact on your rank that the “normal” gametypes (Slayer, Slayer BRs, etc). (Obviously all gametypes in control and objective playlists are “novelty” and therefore this does not apply there)

Another factor that I know for a fact to be true is the percentage of matches won. If you have played team slayer 500 times and lost 400 games, it is much harder for you to go up ranks than someone that has played 500 times and lost 100 games. All gametypes abide by this rule. This is, without a doubt, true.

One last item to address. That is quitting. I did some experimenting on my own account with quitting. As far as I can tell, quitting a match effects your rank just as if you lost the match to players that were all your level. This means that I think that if you are at level 10 and quit a match, it will have the exact same impact as getting in last place in a match with all level 10s. This (and the -1exp penalty) are good reasons to not quit matches!

Hopefully my conclusions and research will help shed some light on the way Halo 3 ranking is done. If you have any questions, feel free to leave a comment and I will try to answer as best as I can.

This entry was posted on Saturday, April 12th, 2008 at 5:30 pm and is filed under Video Gaming. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.