🌿ELO System Updates

[orgle]

Power levelling: this is crazy. Cheaters can still cheat they just need 3 accounts not 2. Meanwhile if I have an evening playing BGA games with my wife all evening we get ELO punished for playing each other too often. Madness.

I completely agree. My Husband and I generally play Hanabi, but we'll branch out to Wingspan, Lost Cities, and Decrypto when the mood strikes. We enjoy playing together, but we're also super competitive. If this change is to support and foster community, and ELO is only meant to reflect your actual skill (and not a reward), then why are we not reflecting the increase to skill level when friends and family play together regularly?

[6element]

Played more with "power leveling fix". Essentially, it says "instead of playing against interesting opponent you know and value you should be playing against strangers / beginners". I was afraid it'd eventually become annoying, and within ~ten games it indeed became annoying

I don't agree that Elo earned when playing against only very small group of opponents is comparable with Elo earned against random players, so the problem this fix is trying to solve is real, but the way this fix is implemented feels totally wrong. Players who play a lot of games (i.e. ten a day ) in a few weeks are going to be discouraged from playing - all their opponents will be "marked by the fix".

Possibly better implementation would only apply the correction when playing against relatively frequent opponents (i.e. present in 10-20% of all the games), not "on the second game" against a particular opponent.

PS: As an experienced game developer myself (29+ years, 40+ published games) I wonder how this feature got passed the QA. Does BGA QA exist? Do they know / care about games? Are they allowed to express opinions and do developers listen to them?

[orgle]

In chatting with people on the Hanabi boards, I've run into a few who are happy with the ELO changes. They complain that they've had many run-ins with "toxic" players who cared too much about their ELO, and arguments over failed clues/plays were ruining their experience. It sounds like the Admin team is in agreement with those casual players, and that's why ELO was made invisible for Hanabi.

But there's already a system for players who just want to have a fun game, don't want to spend too much time learning about conventions and don't want to worry over bombing out -- it's called Friendly Mode. I guarantee that you won't run into "toxic" players in a Friendly match.

But as much as casual players complain about competitive player's obsession with ELO, and claim to be happy about the changes, they need to admit that they play for the fake internet points, too. Otherwise they'd already be playing games without ELO.

ELO worked.

Perhaps the competitive players were too vocal about their disappointment during their games, but the casual players failed first. They failed at warning players at their tables about their desired play style & known conventions in the Presentation field before the game started. It's easy enough to ward off competitive Hanabi players: tell them in advance that you're "casual," that you don't play with conventions (or list the handful you do use), or turn on Friendly Mode -- they'll never bother you again.

But rather than encourage casual players to avoid competitive players by declaring their table as "casual," the admin team is instead trying to cure competitive players' competitiveness & desire for perfection by making ELO invisible.

You're telling us that we're playing wrong. That we shouldn't be so competitive. That we shouldn't be expecting the players at our Master level tables to put some time & effort into learning and understanding a complex set of conventions. That we shouldn't be striving for ultimate efficiency and perfect games 100% of the time.

But who are you to tell us how to find enjoyment in our beloved game? Some people dabble with a Rubik's cube and others enter timed trials. Neither is a "wrong" way to play with the toy. So it is with Hanabi. You gave us a tool to approximate skill level and we used that tool to find players with similar aspirations for elevated Hanabi games. Now that we've built a community around those expectations, you're taking it away. Why? Not because it didn't accurately reflect players' skill level, but because casual players were playing in competitive games with ELO on and got miffed that competitive players were upset that they weren't trying harder to succeed in a competitive game.

The proper solution to appease casual players who say they don't care about ELO is to tell them to only play Friendly games. Or announce themselves as "Casual" in the Presentation field. We all need to warn our table what sort of game we're expecting. High Ranked Masters used to do that simply through their host's ELO or the "Layered" tag.

Give us back visible ELO.

Without visible ELO players can't feel good about any score less than perfect. Previously, I could feel alright about a game with low ranked Masters that ended with 29 because I could see that I still gained a few ELO. It was a morale boost at the end of a challenging deck -- it reassured us that even though we weren't perfect, the program thinks we did a good job considering our average skill level at the table.

But now I have no clue if 29 is "enough" to not lose ELO. People are now more frustrated with imperfect scores because we have no idea if that's considered a good result for our table. If we don't get perfect, we have to assume we lost ELO. That does not feel good. It's really, really dispiriting to spend 20 minutes struggling through a rough deck with inexperienced players, squeeeeeze out a 29, and still not know if you're now considered less skillful as a result -- if you're at risk of losing your current rank.

I am positive that you don't have a single person on the admin team who plays layered Hanabi and agrees with this decision. It's aggressively horrific for all players attempting to improve their skill in this delightful game.

[cwgordon7]

Just chiming in to say these changes are dumb as heck and poorly thought out. Please revert them and discuss with the community before rolling out such changes in the future.

[__X__]

Does this apply to cooperative games as well? We regularly run rematches after a successful Hanabi game. A good group that really meshes can be hard to find. I certainly hope that this is not the case -- right now we can't even see if we're gaining/losing ELO, so gaining nothing for a perfect game with friends would be an absolute travesty. Please advise as I'm unable to personally confirm due to the ridiculousness of invisible ELO.

EDIT: After further testing, it appears ELO is still being awarded with no noticeable decrease in back to back 2p games with the same teammate I retract the following, previously posted statement.

I'd say it absolutely does apply. My last 9 or 10 games of None Shall Pass have "earned" an ELO of +43, but since the NSP community is tiny, and I've played each of the 5 distinct opponents from those 9 or 10 games many, many times in the past 724 games, my elo as shown on the game accept screen, has grown by ZERO.

[orgle]

EDIT: After further testing, it appears ELO is still being awarded with no noticeable decrease in back to back 2p games with the same teammate I retract the following, previously posted statement.

That's a relief! Thanks for checking

[Grumbleguts]

The power levelling change is horrendous - such a broad rule will result in ELO being drained from the pool, in what is meant to be a close to zero-sum system.

I mostly play Arena and regularly get matched against the same opponents by mere chance, through the matchmaking system BGA administers. If am a highly ranked player and I regularly play against the same opponents (games with low player counts, niche time zones, or turn based), why would I and my fellow opponents who are similarly skilled be punished for winning a 'streak' of 2 or more times against the same opponent? This change beggars belief.

Please revert this change ASAP and find a better way to fix the underlying problem without affecting the entire community of paying customers who are doing the right thing.

[George Smiley]

Power Leveling:

I play mostly Agricola turn-based Arena. Win some, lose more. Rating hovers up and down in the 400s. Like everyone else in my games, we are typically strangers in real life. We are from all over the world. We know each other by our BGA names. I'd like to think that some type of affection develops through regular interaction over a game we all love. We also share a competitive streak. Hopefully we share also mutual respect. Probably all play to win, but accept losing fine (more or less). So, we have a community. But one that watches ELO as a reasonable measure of our ability against the community.

Power Leveling will not allow the balance of community, competition and gameplay in that Ecosystem to work properly, I fear. Seems like the wrong fix to the problem of a few people that play multiple seats in the same 4-player game under different accounts.

Please rethink this. It especially harms the truly exceptional players who win regularly ... and fairly ... players that are genuinely respected in the Agricola world.

[Wreckage]

None of the following relates to co-op games.

There is already a huge ELO advantage for creating the smallest table possible. This is why most games are dominated by players who only play 2p games. You may be great players, but you also have a few hundred ELO advantage because of the equations. There is a penalty for each additional player at your table.

If all the players on a table have exactly 200 ELO at the start of a game, the winner will gain about 10 ELO per opponent. However, each added player alters the percentage you gain.

So,
on a 2p table, the winner gains 100% of the allocated ELO. So you would gain 10 of the 10 ELO you earned.
on a 5p table, the winner gains 62.5% of the allocated ELO. So you would gain 25 of the 40 ELO you earned.
maybe this sounds fair so far…it isn’t…
on a 10p table, the winner gains 27.8% of the allocated ELO. So you would gain 25 of the 90 ELO you earned.
notice you gain 25 ELO whether you beat 4 opponents or 9 opponents, so nothing for the extra 5 opponents.

The ELO ceiling for 2p games is about 60%-70% higher than for 5p games, and about…I’m guessing…250% higher than the best player in the world who plays 10p games.

With the new penalty for playing the same opponent multiple times, the 2p advantage will be increased even more, since players at larger tables will meet the same opponents more.

[anykidrated]

With the new penalty for playing the same opponent multiple times, the 2p advantage will be increased even more, since players at larger tables will meet the same opponents more.

Might not be much of a 2p advantage at all. Since the winstreak penalty doesn't reset, your only recourse is to never play them ever again.

Speaking of which I will convert all my Homesteaders games to friendly soon, my winstreak penalty will be too big