Forum

http://skyride.org/2013/01/the-problem-with-pickups/

Its probably because I think the world spins around me but still I must ask,
did you write this as a personal blog/message to me? :D

IRC has always been a Pickup/Gather platform, I personally know for CS1.6 in the early 2000’s, and I think for quake as well somewhere around the same time or even before.
Even though it is really appropriate, people won’t even bother with non-auth simple IRC because it is a hassle to start dealing with the unknown.
I think a non chat-based web pickup would be the best solution.

With regard to matchmaking/team balance, it is indeed no secret that nobody has ever perfected this.
That is because it is impossible. Even if you had every player ranked at his real skill, you are not guaranteed a super-close/tie game. There are too many variables that are out of your hand, regardless of how accurate the ranking is:

1. There is no guarantee that from the given players, you can form the same average ELO teams.
2. Every individual player might have a bad game, might not try his best, might lag and therefore play bad ETC

These are just 2 things off the top of my head as I am writing, I am sure there are more but they are not needed, I think I have shown my point.

With this taken into consideration, it does not mean there is no point in having a team balance feature. It will always be better than a random dividing of teams, and while unrealistic, if you give every player an initial ranking that is somewhere near his skill (as opposed to everyone starting the same), games have a good chance of not being a 5-0 roll.

#lz.pickup was a fail because it required a weird registration, being authed and hiding host name. I got it.
I stripped the bot down, no registration no auth. This means there is no way to identify people and therefore no team balance, but hopefully the easier simpler faster join-n-play style pickup channel will attract more people and be easier to use.
If this doesn’t get played on, then I give up really. (Except for a web-based pickup :D)

I just contacted the TF2Demos.com guy on email so that he can provide a way for auto-uploading of STV demo’s to the site, and hopefully along with logs, this might attract people to play more pugs as they might be able to actually learn a thing or two from the logs and also with STV’s people can watch the players that were against them and what went on, and also they can supply moviemakers with an STV easily.

I know you think all these features are nice but they are just features and more important is to get a simple winning formula (good user experience as you called it) but I think it has some good weight in attracting players.

If it wasn’t obvious I wrote this wall of text because I want your thoughts/response :)

Quoted from Trath

Its probably because I think the world spins around me but still I must ask,
did you write this as a personal blog/message to me? :D

Nope.

IRC has always been a Pickup/Gather platform, I personally know for CS1.6 in the early 2000’s, and I think for quake as well somewhere around the same time or even before. Even though it is really appropriate, people won’t even bother with non-auth simple IRC because it is a hassle to start dealing with the unknown.
I think a non chat-based web pickup would be the best solution.

I disagree. Web based pickups are definitely the future if implemented well, but I wouldn’t go as far as to say people won’t bother.

With this taken into consideration, it does not mean there is no point in having a team balance feature. It will always be better than a random dividing of teams, and while unrealistic, if you give every player an initial ranking that is somewhere near his skill (as opposed to everyone starting the same), games have a good chance of not being a 5-0 roll.

See this is where it becomes statistics. That isn’t true at all. If it’s completely random then the chance of a completely one-sided game is at a very low constant. Depending on the quality of your system and accuracy of your data, there is a real possibility for games to be imbalanced on a consistent basis.

I know you think all these features are nice but they are just features and more important is to get a simple winning formula (good user experience as you called it) but I think it has some good weight in attracting players.

See right here you’ve just missed the point. Attracting people is easy. Piss simple. It doesn’t keep people playing there for months on end. The basics I mentioned are what do that.