344 replies to this topic

[Mother Tereza]

How do you rate people?

Needs to be fair, not just dm or assists. Because sometimes it's just about having the right pick at the right moment. How do you rate that?

 

Well our first step would possibly consider how each gains or loses leader board points Unlike my example, Starcraft 2, it will not be as easy for us to find a way to rank people. One thing we should do is even out how many classes gain how many LB (leaderboard) points per points scored ingame. lets say Class 1 scores an average 40 points per game, and Class 2 scores 80 points per game, Class 1 gets one LB point for every 20 points they score and Class 2 gets 1 LB point for every 40 points they score. This makes sure that no class is hoarded to gain points. we then just compare how many points said player has and raise or lower them accordingly, like a leaderboard.

 

Unfortunately, you have no idea how complicated this objective is. Fortunately, TF2 community has some math geniuses who had succeed in development of this system. Hopefully, we will integrate Jon's system into TF2C in a near-ish future.

[fraac]

Some mixed messages here. Jon is making his own ratings website because he lost faith in the ability of other sites to implement his ideas.

 

Some of you may know that I’ve worked on creating a rating system for TF2 balancing and ranking as detailed here http://www.vanillatf...ancing-a-lobby/, and despite my titanic efforts implementation in lobby services is basically nil and not coming any time soon

Edited by fraac, 02 January 2015 - 12:28 PM.

[CHERRY]

This is not a mixed message. Jon gave TF2Center textual description of the matchmaking system to be implemented.

The fact that he doesn't believe they will implement it doesn't mean that they won't.

Edited by CHERRY, 03 January 2015 - 03:12 PM.

[MR SLIN]

I've read the entire thread and very much dislike this idea. I know you've been toying with this idea for months and I love the positive changes being made to this site as new features are being rolled out. However, there doesn't seem to be much difference between an advanced lobby and a casual lobby other than the atmosphere and expectations of the players. Manual restrictions can already be set in casual lobbies (i.e. minimum 50 lobbies) and if you want to ban specific players as the lobby leader, it is very easy to do so. I think the only thing missing is the ability to halt game auto start until the lobby leader is ready to start the game (rather than the ready up auto start system currently in place). As a lobby leader myself, I typically just reserve my scout slot ahead of time and sit in spectator to prevent the auto start from happening so that I have time to kick the unwanted players. It's currently not a big issue as that is an easy workaround.

A couple other issues I'd like to address given your current ideas, and then I'll suggest a different system of my own for advanced lobbies:

1. Creating a separate group for elite or advanced players is no different than creating an in-house pug group or IRC pug channel. No value is being added here by creating an Advanced Lobbies group, and integrating this system into the site adds no value. Most in-house pug groups are run separately (see "Invite Pugs" steam group for ESEA-IM+ level players only here: http://steamcommunit...oups/InvitePugs ) and if you really wanted to do this in TF2Center you could do it right now by reserving all slots and creating a separate steam group.

2. Creating a set start time for pugs is ridiculous and would only serve to reduce the number of pugs run on a daily basis. It's better to do it the same way scrims are currently done -- people hop into a channel (or lobby) and advertise the game. First come, first served. Over time the players will begin to set the times for ALs on their own, naturally and organically. In practice, teams tend to show up to start scrims at X time and now all teams begin showing up at that time. This free system means that players can create ALs whenever they'd like and can encourage their friends to show up to get games started faster. They're not constricted by the system to show up at X time.

3. Limiting ability to play games by league or division excludes skilled players who aren't in a league, and this means that you'll have to manually manage the group (an issue for growth). Let the players decide who to include or not include through social pressure. Also known as "toxic communities", this social pressure discourages lower leveled players from playing. You see this in current IRC pug systems, other comp games, and in-house pug groups where new players are bullied out. You might argue that this sucks for new players, but that's why these circle jerks form. Everyone is technically welcome to show up to TF2mix or PUG.NA but they're scared to join for fear of being yelled at or being the worst player in the pug. This fear is natural and is a positive way to weed out players that you guys are talking about as "not welcome" for being unskilled. Captaining systems work for the same reasons. In the current casual lobby system, you can simply manually ban out these players that aren't welcome.

I really don't think that advanced lobbies in the way that was suggested earlier in this thread are adding value to the current casual lobby system. I instead would like to support the matchmaking system suggested earlier. Add up as a class and use a hidden ELO system to make games. Although this would require some extra dev work, it would have positive benefits:

1. Prevent team stacking. This in itself would start more games faster, as players tend to avoid joining a game when they see a stacked team on the other side. It would also provide for a more balanced gaming experience with less rolls.

2. It would sort those advanced players into more advanced games, giving you what you ultimately want which is a game against equally skilled players. This supports advanced players while avoiding the elitist mindset that scares away new players. It would also keep unskilled players out of the advanced games. Unskilled players would get matched up against other unskilled players and the system wouldn't label them outright as "casual" or "new". These labels only discourage new players.

3. It does not give players the idea that casual lobbies are for "dicking around" and off classing.

4. It fills games faster by automatically sorting players into lobbies, reducing search time.

5. No steam groups or extra management required. Bans and other punishments can be given automatically according to your predefined standards.

6. It is CLEARLY different than casual lobbies and players of ALL skill levels would want to use the system, especially after ranks are created. This rewards players with more prestigious ranks, encourages skilled gameplay and improvement, and keeps the community together as one whole piece rather than two separate pieces.

7. It doesn't fall prey to the problem discussed earlier which is that as the "elitist group" gets larger, it dilutes the skill level of the group. These "advanced groups" tend to grow over time as players do not want to wait to start games. Over time the skill level of these groups drops and the highest level players in the group become discouraged from playing in the pug. This creates a vicious cycle. A matchmaking system avoids this problem because as the best player in the matchmaking queue you may get matched up against lower level players from time to time but you'll ultimately get matched against the best player possible each time you queue up.

The TF2Center dev team has some amazing talent and has already created a fantastic product. Learn from the lessons of the past and do not recreate the same mistakes that have been made by the TF2 community in previous years by making yet another invite-only group. You guys have the talent to create something truly unique here.

Ultimately a matchmaking system is what the community needs. Look to CS:GO as a model. Eventually the serious players will immediately go to click on "play matchmaking" and the casual players or players who want to play with friends or goof off can create their casual lobbies.

Edited by MR SLIN, 04 January 2015 - 09:09 AM.

[b33p]

 

However, there doesn't seem to be much difference between an advanced lobby and a casual lobby other than the atmosphere and expectations of the players. Manual restrictions can already be set in casual lobbies (i.e. minimum 50 lobbies) and if you want to ban specific players as the lobby leader, it is very easy to do so. I think the only thing missing is the ability to halt game auto start until the lobby leader is ready to start the game (rather than the ready up auto start system currently in place). As a lobby leader myself, I typically just reserve my scout slot ahead of time and sit in spectator to prevent the auto start from happening so that I have time to kick the unwanted players. It's currently not a big issue as that is an easy workaround.

 

This isn't true, I disagree with this fundamental, so everything hereafter is based on flawed logic. The atmosphere might be a little bit different, but to be honest the expectation is the same. It's the ability of the players to live up to that expectation which is not only very different, but a large part of the rationale for creating advanced lobbies in my reckoning. I prefer to play with players who have a good idea of what to do and have DM to match. In the current system, there is a shortfall. In the new system, players will have some competitive experience are therefore in the new system the ability for players to meet expectations is greater than the old system. This is why ALs are needed, not to change the expections; to make them a reality.

 

 

 

 

1. Creating a separate group for elite or advanced players is no different than creating an in-house pug group or IRC pug channel. No value is being added here by creating an Advanced Lobbies group, and integrating this system into the site adds no value. Most in-house pug groups are run separately (see "Invite Pugs" steam group for ESEA-IM+ level players only here: http://steamcommunit...oups/InvitePugs ) and if you really wanted to do this in TF2Center you could do it right now by reserving all slots and creating a separate steam group.

 

This is absolutely not true. Improving the functionality of the site so people don't need to go to further lengths making steam groups, reserving all slots and inviting their supposed huge group of friends to play to meet the same functionality as ALs. At the very, very least, creating ALs improves functionality for those seeking the kind of thing you mention above, but taking all the leg work out of it and casting a wider net to gather players. And as you Americans would say, "and then some". There's a lot more to it than that, but arguing it adds no value is preposterous.

 

 

3. ... this social pressure discourages lower leveled players from playing. 

 

Is this a criticism of advanced lobbies? Doesn't sound like one but for the tone. The point of advanced lobbies is so that low leveled players who do not play competitive and/or those who are unfamiliar with highlander and sixes learn somewhere else. A bit like TF2Pickup. You don't learn to play the game there per se, you go there for practice. You learn to play in casual lobbies.

 

I like the idea of a matchmaking ELO system restricted to those with competitive profiles.

[MR SLIN]

Rather than use the quoting system I'll simply block my responses into three sections.

1. The only difference between casual lobbies and ALs seems to be a required minimum skill level to enter the group. My point was that you can easily set a minimum skill level in the current casual lobby system by setting a minimum number of hours played, lobbies played, or a combination thereof. You can also keep out unwanted players by setting a rage quit % or kicking players manually. I argued that it was the responsibility of the lobby leader to properly manage his own lobbies by kicking those players.

2. Earlier in the thread people said that they wanted to make a Steam group for advanced lobbies, inviting people to join this exclusive group and using the built in notice system to let people know when advanced lobbies started. I argued that doing so made it no different than an in-house lobby group and no value was being added by utilizing steam groups for ALs. If ALs were built into the site, what would it look like to you? How would you differentiate between worthy and unworthy players? How would you manage the system of admitting players to the group, and how would you go about removing unworthy players from the group?

3. It isn't a criticism but a positive. I argued that keeping these IRC pug systems public was not an issue. It might seem like toxic players or bad players can make it into the pug system because it's public, however the players can use social pressure to keep them out or to discourage them from playing. In other words, the public IRC pug systems don't necessarily need to be policed in the same way that a private Steam group or AL system would need to be constantly policed.

My biggest fear is that an AL system would require constant, manual moderation to operate. Skill levels of players would need to be constantly monitored, and problem players would have to be constantly removed, requiring moderator or admin support. This is an issue for scalability and long term growth for TF2C. Public IRC pugs and matchmaking systems do not require the same level of moderation.

[Humblebumbled.]

Rather than use the quoting system I'll simply block my responses into three sections.

1. The only difference between casual lobbies and ALs seems to be a required minimum skill level to enter the group. My point was that you can easily set a minimum skill level in the current casual lobby system by setting a minimum number of hours played, lobbies played, or a combination thereof. You can also keep out unwanted players by setting a rage quit % or kicking players manually. I argued that it was the responsibility of the lobby leader to properly manage his own lobbies by kicking those players.

2. Earlier in the thread people said that they wanted to make a Steam group for advanced lobbies, inviting people to join this exclusive group and using the built in notice system to let people know when advanced lobbies started. I argued that doing so made it no different than an in-house lobby group and no value was being added by utilizing steam groups for ALs. If ALs were built into the site, what would it look like to you? How would you differentiate between worthy and unworthy players? How would you manage the system of admitting players to the group, and how would you go about removing unworthy players from the group?

3. It isn't a criticism but a positive. I argued that keeping these IRC pug systems public was not an issue. It might seem like toxic players or bad players can make it into the pug system because it's public, however the players can use social pressure to keep them out or to discourage them from playing. In other words, the public IRC pug systems don't necessarily need to be policed in the same way that a private Steam group or AL system would need to be constantly policed.

My biggest fear is that an AL system would require constant, manual moderation to operate. Skill levels of players would need to be constantly monitored, and problem players would have to be constantly removed, requiring moderator or admin support. This is an issue for scalability and long term growth for TF2C. Public IRC pugs and matchmaking systems do not require the same level of moderation.

I see what you mean, but I think the best way to combat those problems would be to drop nearly all the specialty of casual lobbies. We should do what gentlemen Jon wanted with ranking systems and drop casual lobbies to the ground, we should refer to them as filthy casual lobbies and promote a ranked gaming system so that it's always competitive, since filthy casuals will probably not care and lose themselves points and go down the scoreboard and the real players would earn themselves points through determination and go up the scoreboards There's a clean way to let newbies float to the top into the real comp players.

 

I know a few programming languages but not the one you guys use, i think. 

[MR SLIN]

yeah humblebundled I totally agree with you. I think an advanced lobby system wouldn't work, and a matchmaking system / hidden ELO / ranking system would be the best thing for comp TF2.

[fraac]

Do we have an ETA for team balancing?

[MasterNoob]

Do we have an ETA for team balancing?

 

We never have an ETA for anything ;-)

[arnold.]

Is this happening? Scrapped? Being implemented?

 

???

[Feb]

As i see this, it would be great, but i am the kind of person who does'nt really want/can play at scheduled times, therefore liking lobbies. Will there be no way for me to get into advanced lobbies, as i dont play "real" competitive?

[K!NG]

A problem i can see with this is that restrictions will become null

[fraac]

Any progress on this subject that others solved ages ago and is still desperately needed?

[MasterNoob]

Nothing concrete so far.

[fraac]

What's the position on using tf2playerrankings.com's API to balance teams? Are you in favour but struggling to implement it?

[MasterNoob]

What's the position on using tf2playerrankings.com's API to balance teams? Are you in favour but struggling to implement it?

 

We're planning on use this for team balancing; at this point it's not so much "struggling" but more a lack of human resources to make this happen. And while I write this, I could have written another line of code, see my point? ^^

[fraac]

Haha yeah, I see your point, it's our fault you haven't added this simple feature after a year of asking and Jon building a whole site for it. Sorry.

[MasterNoob]

Haha yeah, I see your point, it's our fault you haven't added this simple feature after a year of asking and Jon building a whole site for it. Sorry.

 

1. This is not the only thing you have been asking for: http://forums.tf2cen...ge-9#entry18537
2. Jon's site has not been live for that long and it's currently in beta.
3. Where is your contribution in here: http://tf2center.com/donate
4. Where is your application in here: http://forums.tf2cen...g-for-new-devs/

[Kengur]

 

How do you rate people?

Needs to be fair, not just dm or assists. Because sometimes it's just about having the right pick at the right moment. How do you rate that?

 

Well our first step would possibly consider how each gains or loses leader board points Unlike my example, Starcraft 2, it will not be as easy for us to find a way to rank people. One thing we should do is even out how many classes gain how many LB (leaderboard) points per points scored ingame. lets say Class 1 scores an average 40 points per game, and Class 2 scores 80 points per game, Class 1 gets one LB point for every 20 points they score and Class 2 gets 1 LB point for every 40 points they score. This makes sure that no class is hoarded to gain points. we then just compare how many points said player has and raise or lower them accordingly, like a leaderboard.

 

Unfortunately, you have no idea how complicated this objective is. Fortunately, TF2 community has some math geniuses who had succeed in development of this system. Hopefully, we will integrate Jon's system into TF2C in a near-ish future.

 

please for the love of benji... you can't base a matchmaking algorithm on some obscure thing some guy did for himself and nobody else seen what's inside that black box.