Discussion about RI Arena format

[starbar102]

To all RI Arena players,

I want to open up a discussion on the current RI arena format.

Over the past few arena seasons, it seems like RI is getting more and more players (which is great). But it’s the type of game where it’s not difficult to become decent in a short amount of time. This is leading to a large amount of Elite players (30% of all players) and some mis-matched games, in my opinion.

For example, I have my settings to look for games with players only at my level, because with the large amount of luck involved in the game, I don’t like playing a lower tier player and risking 35+ negative points to win 5. Losing one of these games takes 5-6 wins to recover from, and even the best players can lose to unlucky rolls. Because of that, I try to avoid matchmaking with anyone outside of Elite. But with the “same level” setting enabled, I’m getting matched with Elite players at 1,500 or 1,400 points, where the point differential is huge, very similar to playing a Platinum or Gold player.

All that to say that the Elite tier is no longer just Elite players, and with so many people in Elite (the people who play the most), I don’t think the skill levels are being accurately identified.

I think RI Arena (and probably others) could benefit from more tiers and the addition of rank demotion. What I’m suggesting below is taken straight from the Apex Legends ranked format. This format runs 100% on points, no more “10 point levels” like on Platinum and lower.

Entry Level
Bronze (0 - 300pts)
Silver (301 - 600pts)
Gold (601 - 1,000pts)
Platinum (1,001 - 1,300pts)
Diamond (1,301 - 1,600pts)
Master (1,601+ pts)
Elite (top 50)

(the points I'm suggesting are arbitrary, exact numbers would require a look at the data)

All new players are put in Entry Level and only play against Entry Level players. After three wins, they are put in Bronze with 150 points. You cannot get demoted from Bronze once you’re in it.

The Elite tier is just the top 50 players of the Master rank. So this will continually be changing as the top 50 players gain/lose points.

To help reduce constantly going up and down tiers, whenever you successfully rank up (except for becoming Elite), you not only keep all the points gained from the match, but you also get a 50pt bonus. For example, if you’re at 990 points and win 25 from a game, you would end up at 1,065 points. This gives you a 2-3 game buffer before being demoted.

Default matchmaking should be: 1) Your current tier only, 2) One tier above your current tier, 3) One tier below your current tier. After that, you can offer settings for the players to allow for matches against further away tiers.

Potential Upsides:
- More tiers and demotion will spread out the players, giving a more accurate idea of skill level, while making the Elite tier truly the elite group of players.
- You always win (or lose) the full amount of points from a match…not like the current point system in the ranks under Elite **
- Better matchmaking with your skill level so you can avoid some of the high-risk, low-reward games.
- Simpler to manage because it's all using points (maybe?)

Potential Downsides:
- Could lead to further wait times because you’re lowering your pool of matchable players.
- Any others?


I understand this would be a huge change for BGA to make, so right now, it’s all hypothetical. But I’m curious to know how other RI arena players feel about the overall matchmaking, the current tier system, and if anyone thinks changes (these or otherwise) would improve the system.


** If you don’t know what I’m referring to with this. In the Bronze-Platinum levels, there are 10 badges you need to get to level up. A win usually gets you 3-4 badges. If you need one badge to get into Gold tier, and would win 3, you only win one badge and level up to Gold, the other two you would have won are lost.

[Mauritz86]

I agree. And the idea would be worth to consider.

[Richi5000]

Seems you've put a lot of thought into this idea. I do agree and like the concept.

I think this is issue is not exclusively related to Railroad Ink, bit also to other games... I think this might be good on BGA arena in general.

[dschingis27]

I agree that Arena mode needs some serious tweaking. The following was posted by me already in the general discussions forum but I want to share it here again with the competetive Railroad Ink folks.

Here are some plots that prove how ridiculously sensitive the Arena score is on the most recent results.

These are based on a simulation. I compare two players, one starts with Arena score 1700, the other with 1800. (It doesn't matter where exactly I set the scores, what matters is that the difference is 100 at the beginning.) I simulate 100 games of these players against random opponents with Arena scores in 1500 to 2000. In the simulation, both players play the exact same sequence of opponents and get the exact same sequence of match results.

The plots show the Arena score difference between the two players after X games. It shows how quickly the original Elo difference is "forgotton". At the beginning, the difference is 100. With each equal match result, both Arena scores converge closer together, the difference gets smaller.

K=40 (current Arena system) https://imgur.com/a/CRxDagJ
K=20 (current Elo system) https://imgur.com/a/elo-difference-y-ax ... 20-Kxcf2ke
K=10 https://imgur.com/a/elo-difference-y-ax ... 10-Sz8DiwA
K=5 https://imgur.com/a/cRe1nWg

In general, this is expected/wanted behavior. If two players get the exact same results against the exact same opponents, then their Elo/Arena score should converge to the exact same value. The problem is how quickly this occurs. With a high K factor, the convergence is super quick and the original Arena score difference of 100 is nullified quickly. As pointed out, this is stupid for games with a strong luck component.

With K=40, the original Arena score difference of 100 is reduced by half already after only about 15 games. After 30 games, the difference is only at 20, meaning that 80% of historic performance difference is already forgotten after 30 games. (Note that it doesn't matter much what initial Elo difference I start the simulation with. Had I set a difference of 200 in the beginning, still about 50% of this difference is gone after 15 games, about 80% of the difference is gone after 30 games.)
There you have it, with the current Arena system, your score is mostly dependent on your last ~30 games, with stronger weight for the more recent games.

With K=20, it looks already much more reasonable, but the convergence is still way too quick for luck-heavy games. If there is a lot of luck, you need a longer series of games actually matter for the Elo/Arena score to get a somewhat meaningful performance score.

K=10 is something I could live with for games like 7Wonders, Railroad Ink, Harmonies. For games like Can't Stop or Lucky Numbers, you could even choose K=5. The effect is that less recent games are not downweighted so quickly for the Elo/Arena score, less sensitivity on the latest result.