Forged Alliance Forever Forged Alliance Forever Forums 2018-11-01T23:42:08+02:00 /feed.php?f=2&t=13920 2018-11-01T23:42:08+02:00 2018-11-01T23:42:08+02:00 /viewtopic.php?t=13920&p=169116#p169116 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]> Statistics: Posted by Evan_ — 01 Nov 2018, 23:42


]]>
2018-11-01T23:14:00+02:00 2018-11-01T23:14:00+02:00 /viewtopic.php?t=13920&p=169115#p169115 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
https://github.com/FAForever/server/com ... bf208b5f7a

Statistics: Posted by JaggedAppliance — 01 Nov 2018, 23:14


]]>
2018-11-01T23:08:53+02:00 2018-11-01T23:08:53+02:00 /viewtopic.php?t=13920&p=169114#p169114 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]> Statistics: Posted by FtXCommando — 01 Nov 2018, 23:08


]]>
2018-11-01T19:21:18+02:00 2018-11-01T19:21:18+02:00 /viewtopic.php?t=13920&p=169097#p169097 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
It is absolutely crucial that everyone in FAF has the option to play ladder, that the ability is at least available to them no matter who they are. It is the fastest way to get better and offers more variety than any other gamemode, especially when you have low rating and only Astro craters is open to you. I've tried to understand the arguments as to why new player's should still have a starting ladder of 1500, when it is in most cases slowing them down as well as a ton of other side effects.

So far, the only argument I've heard in response (and if there are any others I would gladly like to hear them) is that eventually it would shift the rating down back to even lower since people will be coming in with less rating on average:

... wrote:
by lowering the rating (and deviation) that users start with on faf we'd just end up lowering the overall rating and the relative difference between newbs and established users would not change


However, I've come to believe that this is not the case, at least, not for another 20 years at least. I also think that even if it were true, it wouldn't stop setting new player rating back from having any of the benefits listed below.

New players start with a rating of 1500. However they are not introducing 1500 rating into the system. after a couple of losses where their rating stabilizes to, lets say 300, they've dropped 1200 or more points, while their opponents gained perhaps 100 off said newb at most. Compare that to new players starting with an average of 700 rating, where after the same number of games they stabilize in around the same area. The new player has lost 400 or so points and his opponents gained perhaps 80. So there would be 20 less points for everyone in the world to share. Now this is extremely rough since it's been a long time since I took a look at player's first games. However I think the effect that a new player has on the global rating average is greatly overestimated.

Yes, perhaps the average rating will slowly decrease. However, that could take years, perhaps decades. And wasn't the original assumption for 1500 was that it was the AVERAGE rating? That is clearly not the case, given just how few games are played as ladder rating increases, and meanwhile 500-1200 is a party. Furthermore it's pretty clear to me that EVEN IF the rating average slid back it would still be easier for new players to get games.

See also: viewtopic.php?f=2&t=16357&start=10

Arguments FOR setting the rating back to a more suitable level (or some other change that pits new players against new players):

1: New players can start playing ladder (and no, I don't just mean "find their first game easier." For some people in certain timezones, there are NO 1500s to play against)
2: Decreased frustration for new players as 700s are way closer to their skill level and they don't get crushed in 4 minutes
3: 1500 Players are now less hesitant to hit the search button because getting 0 rated players is annoying and noone wants to ruin someone's first game.
4: Less incentives to smurf, less inflated rating (see viewtopic.php?f=2&t=16357)
5: Overall more active ladder as more people are now able to join and learn at their own pace
6: 700 reflects the average rating much better and is much closer to the starting skill level of most players

Of which I have not heard much in the way of counter-arguments.

Was thinking that now is a good time to bring this conversation back up again.

Statistics: Posted by Evan_ — 01 Nov 2018, 19:21


]]>
2018-10-29T20:35:55+02:00 2018-10-29T20:35:55+02:00 /viewtopic.php?t=13920&p=168944#p168944 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
prodromos wrote:
The average ladder rating is 200, + or - . I know, truth hurts, but as a general criterion one should reset ladder based on this.

It does not matter a little bit if the theoretical average SHOULD BE 1500, and no we don't all agree that it should be so.
Because it is an arbitrary number with no evidence whatsoever.Currently, 1500 equates to place 170 and 1000 to place 896(out of 7500 players). So, coming out and boasting shamelessly how good you are when you are 2000 ish and how all should at least be 1600ish, is elitist and lame. No-one is supposed to be top 100 to be considered "just average". When one starts planning and executing on theory , while reality goes out the window, you get situations like current status on ladder.

People who enter the game , are misguided into thinking that they should be 1000+ right from the beginning, while reality says they should be about -1000. Putting such expectations to community is a bad way encouraging new people trying this game and eventually improving and become decent players , or God forbid offer to the community.


P.S. And no supreme commander is not like chess. It is many orders of magnitude more difficult than chess. It just happens there are more professional players and more attention to chess(which is a very difficult game , just so my opinion is not misinterpreted). So, making a ranking by analogy is a secure way to fail.


This entire thread is kinda pointless.

All of this discussion has already been made:
viewtopic.php?f=2&t=16357#p165223

The average rating was about 850 in late 2017 based on our actual data. Probably didn't change much since then.
Sadly most people don't understand the rating system in detail enough to consider changing it.
Or they don't understand arguments of other people concerning the rating system, so its impossible to discuss properly.
Or nobody cares enough to do it.

Statistics: Posted by Katharsas — 29 Oct 2018, 20:35


]]>
2018-10-17T20:12:58+02:00 2018-10-17T20:12:58+02:00 /viewtopic.php?t=13920&p=168580#p168580 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
It does not matter a little bit if the theoretical average SHOULD BE 1500, and no we don't all agree that it should be so.
Because it is an arbitrary number with no evidence whatsoever.Currently, 1500 equates to place 170 and 1000 to place 896(out of 7500 players). So, coming out and boasting shamelessly how good you are when you are 2000 ish and how all should at least be 1600ish, is elitist and lame. No-one is supposed to be top 100 to be considered "just average". When one starts planning and executing on theory , while reality goes out the window, you get situations like current status on ladder.

People who enter the game , are misguided into thinking that they should be 1000+ right from the beginning, while reality says they should be about -1000. Putting such expectations to community is a bad way encouraging new people trying this game and eventually improving and become decent players , or God forbid offer to the community.


P.S. And no supreme commander is not like chess. It is many orders of magnitude more difficult than chess. It just happens there are more professional players and more attention to chess(which is a very difficult game , just so my opinion is not misinterpreted). So, making a ranking by analogy is a secure way to fail.

Statistics: Posted by prodromos — 17 Oct 2018, 20:12


]]>
2018-10-17T20:04:20+02:00 2018-10-17T20:04:20+02:00 /viewtopic.php?t=13920&p=168578#p168578 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]> Statistics: Posted by sasin — 17 Oct 2018, 20:04


]]>
2018-10-15T20:55:09+02:00 2018-10-15T20:55:09+02:00 /viewtopic.php?t=13920&p=168503#p168503 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]> Statistics: Posted by pedrovisk — 15 Oct 2018, 20:55


]]>
2017-02-13T09:57:03+02:00 2017-02-13T09:57:03+02:00 /viewtopic.php?t=13920&p=143452#p143452 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]> Statistics: Posted by Elvangreen — 13 Feb 2017, 09:57


]]>
2017-02-12T13:22:29+02:00 2017-02-12T13:22:29+02:00 /viewtopic.php?t=13920&p=143384#p143384 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]> Until they fix it, I will stick with modded gameplay with couple of friends.

Statistics: Posted by bratrstvoNODKane — 12 Feb 2017, 13:22


]]>
2017-02-07T08:15:24+02:00 2017-02-07T08:15:24+02:00 /viewtopic.php?t=13920&p=143097#p143097 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
The people sitting around talking might stay but the people affected by it will just leave, we need some solutions

Statistics: Posted by biass — 07 Feb 2017, 08:15


]]>
2017-02-04T16:35:19+02:00 2017-02-04T16:35:19+02:00 /viewtopic.php?t=13920&p=142990#p142990 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
Exotic_Retard wrote:
true, however that 0.08% is the highest change of your rating being there. i would think that a noob starting at supcom would have a higher chance being 300 rated than 1500 rated, which is not what this PDF shows.

so my point still stands, to get to a more accurate representation of your rating you need to lose 5-10 games first, in the majority of cases, since what the rating system assumes is not the distribution of actual new players.

what we could do is take all players under 50 games, and see their average rating, then use that as the starting mean + deviation. Im very sure thats its not 1500 +-1500.

mayyyyybe it would be more representative if there was a rating reset after all the rating changes & bugs

Statistics: Posted by Blodir — 04 Feb 2017, 16:35


]]>
2017-02-04T15:59:42+02:00 2017-02-04T15:59:42+02:00 /viewtopic.php?t=13920&p=142988#p142988 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
so my point still stands, to get to a more accurate representation of your rating you need to lose 5-10 games first, in the majority of cases, since what the rating system assumes is not the distribution of actual new players.

what we could do is take all players under 50 games, and see their average rating, then use that as the starting mean + deviation. Im very sure thats its not 1500 +-1500.

Statistics: Posted by Exotic_Retard — 04 Feb 2017, 15:59


]]>
2017-02-04T15:33:20+02:00 2017-02-04T15:33:20+02:00 /viewtopic.php?t=13920&p=142987#p142987 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]>
Exotic_Retard wrote:
i would expect the opposite effect - rating deflation - since whenever a new player joins theres this effect where they have a juicy 1500 rating and proceed to lose it all - injecting essentially free rating into the system. if we give them less to lose then we can expect deflation not inflation? also couple the fact that a bunch then leave the game and therefore never get it back off other noobs or their old rating farmers.

in any case the system will be affected by this change. im kinda tempted to say it would be for the better though.

Sheeo wrote:
Downlord wrote:For anyone interested, our wiki describes how TrueSkill works: http://wiki.faforever.com/index.php?tit ... kill_works

The relevant part is this: "By default, you have 1500 in mean, and 500 in deviation. 1500 is the average level."

So the system assumes that new players have a skill (NOT rating) of 1500 which, as we probably all agree, is very unlikely (btw. the average player's average skill at this point is 1348). This leads to new people facing ~1200 rated players.


That's not quite what an initial rating of 1500, 500 means, see the PDF (probability density function) of the normal distribution for N(1500, 500^2): https://www.wolframalpha.com/input/?i=N(1500,+500%5E2)

It means that the system assigns a probability of your skill being 1500: roughly 0.08%.

Retard there is no juicy 1500 rating to lose

Statistics: Posted by speed2 — 04 Feb 2017, 15:33


]]>
2017-02-04T08:02:27+02:00 2017-02-04T08:02:27+02:00 /viewtopic.php?t=13920&p=142979#p142979 <![CDATA[Re: Maybe its time to move the Ladder starting skill down.]]> years for new systems

Statistics: Posted by biass — 04 Feb 2017, 08:02


]]>