Good idea! Another similar idea for merging rapid FFA/Solo is to morph from 3 1 -1 -3 to 4 -4/3 -4/3 -4/3 (instead of a discrete transition at 1850 rating).
Not sure about going back to pure 3 1 -1 -3 in blitz though, the 3 player stage would be very teamy again.
Not sure about going back to pure 3 1 -1 -3 in blitz though, the 3 player stage would be very teamy again.
i mean 3 1 -1 -3 for 1850- and 3 -1 -1 -1 for 1850+ in rapid like year ago, and 3 1 -1 -3 => 3 0 0 3 in blitz like now in FFA. I’ll delete the note so as not to cause misunderstanding
Why can't FFA and Solo just use the same ratings and leaderboards?
We will do this for antichess and variants in any case.
+ We add FFA rapid ratings back
You can still play Solo, but it just uses FFA rating.
You can still play Solo, but it just uses FFA rating.
then no one will play Solo for sure. in the Solo system it is much more difficult to gain a rating than in the FFA. 2400 is # 1 in Solo, but not even a top FFA. Also, there is one significant difference between FFA and Solo: in FFA there are three winners (two retain the rating, one receives bingo), in Solo there is only one winner (this gives the potential for game diversity at the 4-player stage, not only teaming)
i just wish there were one game, something between 3 0 0 -3 and 3 -1 -1 -1 that everyone were happy with. But probably there is no such game and the future is hopefully Hests +3 where we will likely not run into the problem of many players getting put off by their beloved game turning teams when they reach 2000.
we could add ffa rapid back, that does seem reasonable. solo bullet must remain as it is quite popular. as for how you are going to get people to play 1+15 Solo, I have no idea. The majority of players did not like being forced to play WTA above 1850, i think the 3 x -x -3 works much better in that regard.
1) I agree with Spacebar, we should re-merge FFA and Solo ratings and leaderboards,
2) as for timing, I believe we have definitely to revise the LIMITS between the Normal/Blitz/Bullet games (I presented already an idea according to which the admins should simply check the stats, what are the average DURATIONS of the different games per settings, so we shall fix then that Normal games are those with settings that generate games above 20 min. for example, blitz between 10 and 20 min. and bullet under 10 min. in average (I speak of course about the overall duration).
Typically, for me 3/1 and 4/0 are definitely blitz and not normal for example, as in most games the 3 players stage is reached in fact very fast.
You can still play Solo, but it just uses FFA rating.
then no one will play Solo for sure. in the Solo system it is much more difficult to gain a rating than in the FFA. 2400 is # 1 in Solo, but not even a top FFA. Also, there is one significant difference between FFA and Solo: in FFA there are three winners (two retain the rating, one receives bingo), in Solo there is only one winner (this gives the potential for game diversity at the 4-player stage, not only teaming)
That's not really a concern; it's clear that a coefficient should be applied, and precisely because of the reason you mention. We can easily fix the coef today empirically: it's enough to take the ratings of the top 50 players (OVERALL, not just those displayed who have played the required number of plays recently) from FFA and Solo boards and it will show which coef should be given to the Solo games ratings.
In addition, as for the transition, the resulting rating should be of course the (Solo*Coefsolo + FFA) / (Coefsolo+1).
The same for normal and blitz ratings.
(in this IMPORTANT thread 3 different issues are being discussed though, and I believe these should remain separate:
* calculation of the ratings (3 0 0 -3, 3 -1 -1 -1, etc.)
* leaderboards and rating systems
* timing of the games.
I do think it's a good idea to combinate the timing with modes like Arseny proposes. Different players like to be able to choose between different settings, timings, modes.
My synthetic proposal would be:
- to have ONE individual rating (taking into account both FFA and Solo games, but with coef as explained above) per timing setting (Normal/Blitz)
- to keep bullet and hyperbullet outside of the system, with a separate rating
- to display for the players only their Individual and their Teams ratings for Normal and Blitz, i.e. maximum 4 ratings,
- to keep as many ratings and leaderboards as players want (as some players love niches), but basically these variants, antichess, bullet/hyperbullet etc. will remain SEPARATE and won't be taken into account for the Champions leaderboard,
- to let the person who launches the game free to chose either the game will be Solo or FFA (or something else as other possibilities have been proposed, cf. above), but all these calculation modes will be re-merged in the unified Individual mode,
- to fix very precisely the limits between Normal timing, Blitz timing and Bullet/Hyperbullet timing simply looking at the stats as for the average duration of the relevant games. It will fix the issue one for all, eliminating the debates whether 4/0 or 2/2D or 3/1I should be there or there. It seems cristal clear, I don't even understand why we improvise instead of adopting a clear classification that would take into account the observable REALITY. For me typically 4/0 cannot be compared to 15/1, it's 2 different game modes (that today contribute to the same rating though). Whereas 1/5D is in fact much closer to 4/1 than to 1/2I. Etc. If we analyse the stats over let's say 6 months, it would clarify everything quite easily.
as for how you are going to get people to play 1+15 Solo, I have no idea. The majority of players did not like being forced to play WTA above 1850, i think the 3 x -x -3 works much better in that regard
reading the comments of the players on the forum it seemed to me that people think differently
there was a solo of 1850+ 1|15D, there was ffa (3 1 -1 -3) 1850+ 1|15D, but ffa (3 0 0 -3) 2150+ (old 1850+ adjusted for inflation) 1|15D did not exist yet (almost all strong players have already switched to 4|0 by the time of the February update). we can try. but there will be either Solo 1|15D or FFA (3 0 0 -3) 1|15D - this will not coexist (the WTA rule in FFA for rapid is complete death for Solo). will need to choose one
FFA seems more fun and has a clearer game scenario. But in Solo there is more professionalism and the influence of teams is less, and most importantly, the rating system corresponds to the level of the game (many 1600 in Solo do not understand why helping opp, 1800 do not understand why helping rivals at the 3-player stage, 2000 do not understand why to play until the end, when there are far fewer pieces left; besides the strategic aspects of the game, tactical vision has an impact). But what some 2500+ FFA players do is simply awful: the game for blue and green did not become unbalanced in the 4-player stage, probably only due to a decrease in the level of the game for red and yellow. But this is not the worst, still ffa should be closer to the solo, and not to the team. But no. What the players do at the 3-players stage looks even worse - a complete lack of understanding of the basic solo things. maybe 1|15D will fix it. but it seems to me the main reason is the lack of punishment - can safely ruin the game and stay at your rating (on 2 or 3 place)
I think Solo Rapid makes more sense than FFA Rapid
I think I suggested 3 x -1 -2-x morph before, where x decreases from 1 to -1. so it's 3 1 -1 -3 --> 3 0 -1 -2 --> 3 -1 -1 -1.
idk if it makes sense, but it may change ffa rapid as a "softer" solution
it seems to me the main reason is the lack of punishment - can safely ruin the game and stay at your rating (on 2 or 3 place)
100%!!!
I think I suggested 3 x -1 -2-x morph before, where x decreases from 1 to -1. so it's 3 1 -1 -3 --> 3 0 -1 -2 --> 3 -1 -1 -1.
idk if it makes sense, but it may change ffa rapid as a "softer" solution
I would continue --> 3 -1.5 -1 -0.5.
but 3 0 -1 -2 scares me, as there will be players who are willing to take a chance and get 3 or -1 playing in solo and players who will play for 0 in the team until the end. the former will be at a loss, although we should, on the contrary, encourage them
What about +3 +0.5 -1 -2.5 for the unified Individual game (Solo+FFA remerged)?
In this case,
a) the 2nd place is not rewarding enough to encourage blatant teaming,
b) the difference between 2nd and 3rd on the one hand, and the one between 3rd and 4th on the other hand, are equal and are not that huge that it could generate perverse behaviours like "I betray in order to avoid being 4th at any cost",
c) to finish with, it's not the same neither being 2nd, 3rd and 4th, so people who are unlikely to win won't cooperate necessarily in some cases that today in Solo make it so difficult (lottery sometimes) to win.
I mean, the victory should depend only on one's skills, not on other players' blunders nor on their successful/failing cooperation against the leader. Our beautiful game is subtle and the points given should be weighed carefully.
Another idea would be to give the points as follows:
- all 4 players together still have a zero sum,
- points obtained in the game, weighted per rating before the game, make the points that you are given.
Basically, today when you are 3rd with let's say 41 points whereas the 2nd has 42 is very discriminating and I feel it's not really fair.
The result 106 42 41 6 and the result 84 81 19 0 are pretty different. Etc. We could take into account not the place in itself, but the number of the points.
(That would also have for positive effect that people wouldn't abandon easily as they could still hope in some cases grasp some points and diminish the impact on their ratings; that would also reduce the teaming in fact as the queen put in risk would become much more likely to be eaten after the checkmate, etc.).
I like those 3 0 -1 -2 and 3 0.5 -1 -2.5 ideas.
This seems quite reasonable for me.
The Idea of rating by points can prolong games because before you take the mate, you will take every piece possible.
That reminds me, I have a ready made, ready to go rating system which is based on points (and used for play4mate).
It looks at the average points at the end of the game, and determines the number of wins or losses vs avg rating by looking at how far or above or below avg points each player was at the end. distribution is so that total wins=total losses=4
so for a 105-0-0-0 result you get something like solo, for 60-45-15-0 something like ffa
I always wondered how that would change the game. seems like mates should be worth less tho, +3 as hest likes it, or +5 +10 perhaps
I actually wanted to put that as an option a while back, like WTM was. What would you call it?
i just wish there were one game, something between 3 0 0 -3 and 3 -1 -1 -1 that everyone were happy with. But probably there is no such game and the future is hopefully Hests +3 where we will likely not run into the problem of many players getting put off by their beloved game turning teams when they reach 2000.
we could add ffa rapid back, that does seem reasonable. solo bullet must remain as it is quite popular. as for how you are going to get people to play 1+15 Solo, I have no idea. The majority of players did not like being forced to play WTA above 1850, i think the 3 x -x -3 works much better in that regard.
well I’ll try some more arguments in favor of removing the solo and embedding it in ffa. Of course, can count on +3 hest and even make a separate leaderboard for this game, but for a start it would be nice to optimize the leaderboards of FFA and Solo. Now there are 5 standard FFA and Solo leaderboards. I think this is a lot, ideally should be two. How to do it:
Remove the bullet solo. It is no different from ffa bullet. Here are some examples of rare high-rated games (a good rating for players who do not play hyperbullet or ½):
ffa bullet:
https://www.chess.com/4-player-chess?g=3605585
solo bullet:
https://www.chess.com/4-player-chess?g=3439424
As can see, this is pure solo in both examples (opening solo and there is no frank cooperation between opps). Oddly enough, Solo and FFA bullet is the only place where still play in real solo and where the last true solo players live, not touched by the destructive influence of the teams. I think the combined bullet will positively affect the pool of players, the number and quality of games. Now joining a solo bullet game, you are likely to meet 2 or 3 1500.
Remove solo blitz/rapid. Firstly, unlike FFA and Team blitz/rapid, there is no competition in it and its rating already means almost nothing, that's how player farm a place in the leaderboard:
https://www.chess.com/4-player-chess?g=3621000
Although good players earn a rating in the same way. i.e. there are no highly rated solo games (at rapid for sure). although, in my opinion, the solo is designed specifically for highly rated players, since in order to understand how to play for first place correctly (game strategy), you need to play a lot of games. Good games are sometimes found in blitz:
https://www.chess.com/4-player-chess?g=3612438
But as can see, this is ffa (team debuts and active cooperation at the beginning of the game), nothing would have changed if this game had happened in ffa.
But remain Solo in rapid (FFA/Solo). I personally liked the championship, and I hope to see another one. Watching and playing Solo games is more interesting than FFA games, it is more diverse. FFA is 100% Team in 4players stage (with a rather limited number openings compared to the Team variety, other debuts are not used due to the low skill of the teaming of players), in the solo their own debuts are more often used. In fact, this also affects the 3 players stage. The 3 players stage in the FFA is less balanced than in Solo, since the mate for flank player becomes more important than the points and position of own king.
Secondly, if still want to keep the championship solo, then need a decent selection for it. If we had one competitive leaderboard for rapid, then we could just take 16 places from this leaderboard (in essence, this is a leaderboard and performance together) and play more games in semi-finals.
Thirdly, FFA (300-3) solved the problem with teaming only at the 3-player stage. It seems to me that the players are more outraged by the frank teaming at the 4 players stage than the mandatory transition to solo. Now the border is somewhere around 2150+, here is the FFA (solo) 2100+:
https://www.chess.com/4-player-chess?g=3599076
here is the FFA (team) 2100+:
https://www.chess.com/4-player-chess?g=3621604
But these were examples of players ready for this, and here is an example of a transition that someone is not happy with:
https://www.chess.com/4-player-chess?g=3585682
I can’t even imagine what horrors are happening in this rating range and what verbal battles are taking place here, but it seems to me that not everyone enjoys it, someone quits the game.
Fourth, the FFA rating system poorly represents the level of players. The player rating must change after each game, this updates the player rating. Also, between the ratings of players in brackets there is often a large spread, which is also not very good.
Fifth, for 300-3 there is a strategy that is often used for blue or green in high-rated games:
https://www.chess.com/4-player-chess?g=3642439
player just wait until him more active opp dies (in the example, blue waits until green dies). This is actually not a small problem, I think it makes the game for the green and blue unbalanced. Here is an even more revealing example:
https://www.chess.com/4-player-chess?g=3656633
blue and green just wait for their opp to be killed, as see, green is unlucky (luck decides who loses in 2500, this is nonsense).
Sixth, there are true solo players who have not changed their style, have not quit playing, and even remain on the leaderboard, for example, Spaksi:
https://www.chess.com/4-player-chess?g=3615349
I can’t imagine how many of those who quit. I would like to play with him as in the good old days:
https://www.chess.com/4-player-chess?g=1236191
Seventh, ffa is still a symbiosis of solo and team. See how many team players are on the ffa leaderboard. But should have seen how they play in solo - this is horror, but they are saved by the absence of 4 places, sometimes a good position after the stage of 4 players. FFA is for those who are a little good both in team and in solo. But we have a separate leaderboards for team, need to make a separate leaderboard for players in solo, but already more competently. I understand that combining Solo and FFA is a duct tape, but it is better than a non-competitive Solo and unpunished for a bad game FFA.
if a gamemode needed a rapid leaderboard, that's definitely Solo. in order for "good solo" to be played the game needs to have some increment/delay, check the top solo player's archives all they do is play no rating range 4|0 solo and beat 1500s with 0 effort obviously and because of rating system the still get some points for that, that is not good solo. Rapid Solo leaderboard would fix some problems and would get more people playing rapid/ somewhat good solo, well, just to get a place on a leaderboard. But we got a FFA Rapid leaderboard for which I will not complain too much ))
a discussion of this topic (https://www.chess.com/clubs/forum/view/separate-rapid-and-blitz) gave me the idea: to separate FFA blitz and rapid, but at the same time to make the rapid according to the FFA/Solo system (border 1850+), but leave blitz as FFA (like now 3 1 -1 -3=> 3 0 0 -3) and remove Solo blitz/rapid as a separate leaderboard
)
I think this is a brilliant idea (as in the Russian proverb: you will not praise yourself - no one will praise you
advantages: preservation of the popular 4|0 FFA format, exclusion of farmers in 4|0 Solo, preservation of the number of leaderboards (although I would delete all Solo leaderboards (bullet, variants, anti) and simply add the WTA rule to FFA (as it was before), then the number of leaderboards will become smaller, which is even better) and most importantly it will revitalize Solo
this solves two more important problems:
- Solo without FFA is dead (a scientific fact obtained as a result of a year of experiment). For beginners, an incentive for 2nd place is needed - this is a stimulus for further play and a training court
- FFA 2200+ 1|15D (combined with 4|0) is boring. playing 1|15D with those who got their rating at 4|0 is not interesting, but playing with them at 4|0 is fun, because you yourself start to play poorly (if evaluate in terms of rapid). But I miss those times when I actively played in FFA 1|15D, and even more when I actively played in Solo 1|15D, and I think I'm not alone
implementation: FFA blitz players get their FFA blitz/rapid rating at the beginning, FFA/Solo rapid players get their Solo blitz/rapid rating. then the change will not offend anyone and will make the transition smoother and more acceptable for the start