AllegSkill

[Mastametz]

Agreed. That's why I did suggest in our discord conversation that newbie's displayed in-game rank be on a point based system until they achieve X rank and then they are assigned their base Elo value
their points will still be tracked past that point (as are everyone else's points) for other things but not to do with the displayed in-game rank

[Mastametz]

The points for objectives stuff are good for building player profiles (which is something I can work on as well).

Personally I'd love to be able to see what my accuracy % is with all the weapons I use most, that sort of thing. Like In Left 4 Dead and I'm sure many other games.
I don't know if this game engine/code allows for tracking of that sort of thing.

[Terran]

QUOTE (Xynth @ Mar 20 2018, 03:44 PM) BT wanted this ranking algorithm because he had hope for an influx of new players to allegiance that would enjoy a more modern progress based gaming experience (see Fortnight for example). We obviously did not achieve that influx and the old vets still rule the day. The ability to measure an allegiance e-peen is important to a large portion of the player base. I don't have the bandwidth to implement a new allegskill but if anyone has questions on how to wire it into the current system let me know. It should be fairly straight forward to get it plugged in and leverage steam stats to do it.

ah you mean one dev unilaterally decided to change an important part of the game and it turned out to be a complete $#@!up? you don't say...

[Terran]

QUOTE (Wasp @ Mar 21 2018, 01:05 PM) Laughable.

Sorting the leaderboard by kills on drones and k/e ratios provided better ranking than that "Allegskill" thing ever could. Seeing you here trying to sell it all over again, tells us that you're still in denial.

nah man. i recall all the good k/e players and the top drone killers also having the highest rank assigned.

[Wasp]

QUOTE (Terran @ Mar 21 2018, 07:18 PM) nah man. i recall all the good k/e players and the top drone killers also having the highest rank assigned.

If I could select by sorted stats,

I'd select the high drone killers first, the kills per hour second and the base killers third.

[zombywoof]

QUOTE (LANS @ Mar 21 2018, 06:50 AM) Don't lie. You are the stack.

I mean, I'm not here to claim that the majority of games I play don't become stacked, but again I'm not sure what you want. If there are two even teams with myself and, I dunno, tenforward in the lobby then the team I join becomes, by definition, the stacked team. If that makes me a stacker then ok you got me.

But when I get to choose where to fly, I prefer to grab sumv and join the underdog team and see if we can turn it around to joining the uberstacked team.

[TheAlaskan]

Where players topped out in AllegSkill was more or less a decent indicator of their team value. If you were a commander and knew all the individual players and their abilities, then that gave you a leg up on understanding your team or actually picking your team if given the opportunity.

Those who ended up in the 6-9 range were poor-to-potentially useful players. Some trainable, some not. I don't feel like pointing to anyone specifically.

Those who never made it past the 10-13 range were more or less decent voobs. Could do utility, but ultimately needed a tech advantage to do most of their whoring.

Players in the 14-16 range were usually good players, but not the Put The Team On My Back kind of guys. You could count on them to be a positive part of your team effort as a commander.

17+ were all upper echelon players... I'd say maybe 1 in 10 got to that level simply by playing a ton, and we all know who they were. I'm not sure TenF could have unintentionally abuse the AllegSkill ranking system as bad as it seems he is unintentionally abusing this one.

Perfect or not, players ended up falling into those buckets and the Majority Of The Community seemed fine with it.

While there was a separate commander rank, it's not like anyone ever saw that In The Game. It was a vanity for the people who actually had time to jerk off to the stats table.

[cashto]

QUOTE (pkk @ Mar 21 2018, 01:56 AM) Balanced teams:
TrueSkill considers teams balanced, if the average mu and sigma are equal on both teams, that doesn't mean if average ranks are equal that the game is balanced. This is $#@!ing misleading.

It's not that misleading. If both teams have similar distribution of sigmas, it balances out. If they don't -- do you really think a team full of complete newbies should have the same aggregate rank as a team full of known average players?

QUOTE Paps is a bad player, because he is loosing a lot. A newbie might be a better player, but because the system doesn't have enough data (high sigma), he has the same rank as Paps.[/quote]

My recollection was that if newbie was any good, they were probably a hider and went back to their original rank very quickly. Play more games and you rank up even faster, by mere virtue of sigma decreasing rapidly. Shouldn't we encourage that?

QUOTE (pkk @ Mar 21 2018, 07:58 AM) Purpose of TrueSkill: automatically create (balanced) games from a large queue of players, so there is an enjoyable environment for all skill levels

Result of AllegSkill: e-pen

Heh. That's true to a certain degree, but not completely true.

Creating balanced games is on the honor system around here, but believe it or not, some people DO want to help keep games balanced, and an accurate ranking system is a great tool for them to do that.

And don't knock the e-peen aspect completely. People like to see visible progress. Ranking up is a motivation to improve, and stat whoring is a motivation to play more. That's as true outside of Allegiance as within it.

I don't see what problem "remove stats" solves. You believe the ranking system is arbitrary and meaningless? Fine, don't look at it. There's no need to take it away from people who have a different opinion.

[Dome]

Ranks are great. Trying to climb a leaderboard can be fun. We are all so jaded now but as a new player, seeing your stats accumulate is fun.

Both hELO and Allegskill worked fine for me.

[zombywoof]

QUOTE (cashto @ Mar 21 2018, 08:09 PM) My recollection was that if newbie was any good, they were probably a hider and went back to their original rank very quickly. Play more games and you rank up even faster, by mere virtue of sigma decreasing rapidly. Shouldn't we encourage that?

P32 used to show up in a (0) like every third weekend and in the course of fifteen games or so rank up to 16. Then he'd show up with a new (0). But that was P32.

When Terran started playing a lot of pugs he couldn't even dump his rank by intentionally stacking the enemy team.

The fact is, AllegSkill worked like a charm and did what it said on the tin: the good players got higher rank, the bad players stayed the same.

QUOTE I don't see what problem "remove stats" solves. You believe the ranking system is arbitrary and meaningless? Fine, don't look at it. There's no need to take it away from people who have a different opinion.[/quote]
It solves the problem of people feeling bad because they're great but their rank is low. I guess.

QUOTE (Wasp @ Mar 21 2018, 05:20 PM) If I could select by sorted stats,

I'd select the high drone killers first, the kills per hour second and the base killers third.

Funny, this is almost *exactly* how the current rank system works. Sure there was the three week period with the probe silliness and opening scouting is counted, but TenForward didn't get to be top dog by nanning his way there, he got there because if there's anything he loves it's blowing up bases in TF bbrs.