Does ELO make Love Letter quite a different game?

Discussions about BGA (all languages)
Forum rules
Warning: challenging a moderation in Forum = 10 days ban
More info & details about how to challenge a moderation: viewtopic.php?p=119756
User avatar
pikiou
Posts: 389
Joined: 03 October 2011, 05:36

Does ELO make Love Letter quite a different game?

Post by pikiou »

I started playing 4-player Love Letter games here a few days ago and tried to win just like in the physical game.
But some people would attack me even though I was 3rd! I soon realized they were securing their 2nd place rather than trying to end up 1st, just so they get ELO points. Legit, I thought.
But now attacking the 1st player feels stupid cause in the meantime the others are attacking you to secure a 2nd place! So everyone ends up attacking the player ranked the same or right below. And there is no teaming up against the 1st.

It feels like a completely different game, which could be ok except it's nowhere as fun of the physical game.
Am I the only one with this feeling? Is my logic wrong? I'm definitely overthinking this...
latexrhombus
Posts: 65
Joined: 14 November 2011, 22:40

Re: Does ELO make Love Letter quite a different game?

Post by latexrhombus »

I don't think it makes it a different game, but in general I have found Love Letter on here pretty boring. The gameplay itself isn't especially compelling and with the exception of one of my games (in which the room host stated the table was a fast table for fast players) the game played incredibly slowly.

I would say the issue you mentioned should only present itself closer to the end because there is an ELO benefit to moving up in place. However, the whole system (not just for Love Letter) seems to overvalue non-1st place finishes in games featuring 3+ players. I think a more appropriate system would be to award big gains to someone who finishes first in a 3+ player game while only providing marginal benefits to 2nds, 3rds, etc.

The BGA system seems to be you get credit for beating each other player individually that you place above and losing to each individually that you place below. This system has many flaws as it inherently encourages kingmaking in many situations for essentially EVERY game (though the point about it being an aspect of Love Letter is definitely accurate).

A better ELO system for 3+ player games would probably be something similar to some heat scoring systems employed at the World Boardgaming Championships. For instance, in El Grande (which is typically played 5 player at the WBC) a 1st place finish scores 100 points, while a second scores only 20. 3rd and 4th place finishes yield essentially no benefit here. This forces people to always play to win which keeps the game fun.

Implementing a system here that takes into account ELOs but promotes "play to win" behavior in 3+ player games would be significantly more complicated than the static example provided above. I think the short-term workaround would be to just turn ELO off in those cases (which frankly, I don't find as fun) but I think what you have experienced is a much bigger issue on here and by no means unique to Love Letter.
demiurgsage
Posts: 229
Joined: 07 August 2015, 10:08

Re: Does ELO make Love Letter quite a different game?

Post by demiurgsage »

I agree, ELO doesn't suit to most of 3p+ games. Even if the system looks fair (e.g. 6nimmt), it provides too much score gains to winner and too much score loose to looser. You need only two lucky games to change your rank from Average to Expert and vice versa.
User avatar
Oso
Posts: 701
Joined: 29 September 2011, 04:04

Re: Does ELO make Love Letter quite a different game?

Post by Oso »

demiurgsage wrote:I agree, ELO doesn't suit to most of 3p+ games. Even if the system looks fair (e.g. 6nimmt), it provides too much score gains to winner and too much score loose to looser. You need only two lucky games to change your rank from Average to Expert and vice versa.
But well, being first in a 4p game is harder than a 2p game. It's because you get more ELO if there are lot of players on game.

I just say, don't play with people about -100 ELO than you.
latexrhombus
Posts: 65
Joined: 14 November 2011, 22:40

Re: Does ELO make Love Letter quite a different game?

Post by latexrhombus »

Tight ELO restrictions for larger games can be impractical. While winning a larger game shold be worth more points simply because you overcame more opponents, a system that awards each of those victories as individual victories is misrepresentative because they aren't individual victories.

As mentioned in previous posts, there are more factors involved in winning a larger game than simply out-playing each other player involved in a head-to-head basis. The ratings shold reflect that and encourage players to try to win rather than to hold back the weak for personal gain.

Along the same train of thought, you should naturally lose more for finishing last in a larger game but not as if you were individually beaten by each other player, because you likely weren't. There are too many factors in such a game to make that assumption.

Notwithstanding my viewpoints expressed above, I actually find the point gains for players in the middle to be the major issue with the current system. There is too much of an incentive to hold second place, for instance, because (in 4+ games) it is recorded as a win against multiple players and a loss against just one. This is typically a nice-sized ELO win, but in reality it should be a loss.

Placing above other players is being mistaken for winning in the current ELO system and this is the root cause of the issues experienced by the OP.

If you did not win the match, you should not be treated as if you did.
User avatar
pikiou
Posts: 389
Joined: 03 October 2011, 05:36

Re: Does ELO make Love Letter quite a different game?

Post by pikiou »

You all make very good points about the ELO system and its limits.

One defense for rewarding 2nd and 3rd places in a 4-player game is that when you're sure you're not going to end 1st, you still have an incentive to play and maximize your score rather than ruin the current 1st player and make him/her end 2nd. But this is only valid in racing type games.

Because I do feel that some games are like racing where only the top players (say the top 20%) should get a reward for their mostly individual performances; and some games are like gladiator arenas where only the survivor should get a reward since you can attack specific opponents and teaming up against the strongest makes the essence of the game.
Shaq Jenkins
Posts: 159
Joined: 27 November 2013, 04:49

Re: Does ELO make Love Letter quite a different game?

Post by Shaq Jenkins »

I was just about to make a post complaining about this same issue. ;)

I've played almost 100 games of Love Letter at this point, and in about 75% of these games, the losing players refuse to go after the player in first place. A few nights ago, I played a game where the score was 2-1-0-0. Instead of trying to eliminate the player in first place, one player decided to eliminate another person who had 0 points. The first place person won the round, making it 3-1-0-0, and eventually won the game.

This is really bizarre to me - when I've played Love Letter in person, the goal was to eliminate the person who's currently winning so the other players have a shot at getting first place. I think ELO probably factors into why people play the game this way. Since Love Letter is a game that involves very, very little strategy and a lot of luck, I think it should be played with ELO off and no rankings, similar to how Hanabi is.

Love Letter is more fun and dynamic in person. I do enjoy playing it on BGA, but I'm sick of players who are more interested in protecting their precious ELO than eliminating the person in first place. That takes the fun out of it. I've seen way too much kingmaking and other forms of cheating going on. I played at a table with 3 other people who clearly knew each other, and as soon as someone played a Priest on me, the next player would use the Guard to correctly guess my card. It's easy to cheat online.
User avatar
pikiou
Posts: 389
Joined: 03 October 2011, 05:36

Re: Does ELO make Love Letter quite a different game?

Post by pikiou »

Ok so you're confirming that rewarding the 2nd place does make some tables of Love Letter very different from the physical game.
I'm not crazy, glad to hear ^^

About the players teaming up and cheating, this sadly is why red thumb exists (especially since now it "bans" these players from your future table). No ELO fixing will handle this situation :(
Liallan
Posts: 1221
Joined: 26 May 2014, 07:01

Re: Does ELO make Love Letter quite a different game?

Post by Liallan »

If I think I can't win, I'll play for 2nd, whether or not there's a scoring system. I play this way in person as well.

I think being able to come in 2nd has meaning. I know there's lots of people who think you either won or you lost, but I think coming in 2nd means you did better than whoever is behind you and I'll take pride in being 2nd in a 5 player game. I always play to win. But if I can't, then I'll go for however high I can, which to me says something about how well I can play. And yes I think I should be rewarded for that.

Furthermore, I have a right to shoot for 2nd if I want. Which is why the concept of discouraging people to play in any way except to win really seriously ticks me off. Because you don't have any right to decide for me that I can't shoot for 2nd (or 3rd) if I want. I always start a game trying to win, but there can come a point when I have to make the decision whether to let that go and just see how well I can play against the others. I know some people will take great risks to win, which can land them last if it fails. I'm just not that kind of player. Usually.

If it earns me more skill points, great. But I do this when no skill points are at stake because that's how I play.
Shaq Jenkins
Posts: 159
Joined: 27 November 2013, 04:49

Re: Does ELO make Love Letter quite a different game?

Post by Shaq Jenkins »

Liallan wrote:If I think I can't win, I'll play for 2nd, whether or not there's a scoring system. I play this way in person as well.

I think being able to come in 2nd has meaning. I know there's lots of people who think you either won or you lost, but I think coming in 2nd means you did better than whoever is behind you and I'll take pride in being 2nd in a 5 player game. I always play to win. But if I can't, then I'll go for however high I can, which to me says something about how well I can play. And yes I think I should be rewarded for that.

Furthermore, I have a right to shoot for 2nd if I want. Which is why the concept of discouraging people to play in any way except to win really seriously ticks me off. Because you don't have any right to decide for me that I can't shoot for 2nd (or 3rd) if I want. I always start a game trying to win, but there can come a point when I have to make the decision whether to let that go and just see how well I can play against the others. I know some people will take great risks to win, which can land them last if it fails. I'm just not that kind of player. Usually.

If it earns me more skill points, great. But I do this when no skill points are at stake because that's how I play.
We have completely different perspectives on this issue because I play to win, not to settle for 2nd or 3rd place. That's my philosophy in life - I'm not okay with being just good enough, as I want to be the best.

Furthermore, in Love Letter, by targeting the person who's leading, the other players now have a better chance at winning. I played a game recently where the score was 3-3-2-1, and the person with only 1 round win actually won the entire game because he didn't settle for gaining anything less than first place. Love Letter is such a unpredictable game - the person who's winning might have 2 or 3 rounds of terrible cards, so why concede and assume that they'll come in first place, anyway? Why make their win easier for them by not challenging them?
Locked

Return to “Discussions”