21 posts in this topic
Your content will need to be approved by a moderator
Recently Browsing 0 members
No registered users viewing this page.
Ben Brode confirmed the 2 game win streak to be unintentional and explained why he's not satisfied with the Ranked Ladder System.
The developer team listened to the community and they will be making adjustments based on simulation systems that help predict changes that would do to the population curve. Here's what Brode had to say.
Seeing some comments here about how people are enjoying easier laddering due to this bug, and hoping we leave it unfixed. I thought I might chime in and talk about the ladder a bit, and hopefully get some feedback! We have been discussing the ladder system a lot recently - we're not 100% happy with it.
Here are some things we are currently discussing:
Rank 18 players are higher ranked than 50% of HS players. That number doesn't make you feel like you are in the top 50%, and that's a missed opportunity. We try and counter this by telling you all over the place what the mapping is to the rest of the population, but it'd be better if expectations and reality matched here.
We've received feedback that the last-minute jostling for high Legend ranks at the end of a season doesn't feel all that great.
We've received feedback that the ladder can feel like a grind.
We are reanalyzing the number of ranks, the number of stars per rank, the number of bonus stars given out at the start of the season, and other parts of the system.
We are developing simulation systems that let us predict what changes to the ladder would do to the population curve. If we inflate too many stars, the whole population ends up in the Legend bucket and while that might feel great for a single month, the entire system falls apart eventually. People who played waaaay back may remember when "3-star master" was the pinnacle of achievement, and it meant nothing because so many people ended up in that bucket. With better simulation tools, we are planning on trying a lot of crazy things. Iteration is important in design, and getting the tools to iterate quickly is very important.
Something I want to emphasize is that while I think we can improve the ladder, the metric for that improvement isn't necessarily any one player's individual rank increasing. Players want the better rewards (and prestige) associated with high ranks, or the Legend card back, so any change we make that increases the chances of those are likely to be perceived as "good", at least for the short term. But part of what makes the ranked ladder compelling is that exists to rank players. If you want to see how you stack up, ranked is the place to do it. So while some inflation might improve the experience, we need to be careful and make sure we end up with a system that makes people feel rewarded for increases in personal skill or for finding a new deck that breaks the meta.
Have you considered not making the fall in rank so severe in the start of the new season? When I was a new player at rank 21 it didn't feel good to lose against a wicked good deck all the time with my inferior cards.
Yes, that's what I meant by this: We are reanalyzing [...] the number of bonus stars given out at the start of the season.
We think it's a reasonable direction to explore in, but in general it's hard to predict the results on the population distribution among the ranks. Hoping to do some simulation there and make sure there aren't other, better options, too.
Is this really that bad considering legend has its own internal ranking system?
Yes, I think so. Imagine just one bucket and we sort by MMR. There aren't breakpoints with rewards as you increase in skill. There isn't an obvious way to communicate with others about how good you are. It's also difficult to know if you are bad or good. (Is Legend 135003 good? What about Legend 27809?)
The way we communicate our skill or progress is important. Ever heard a friend say "dude! I got to rank 5 this month!"? What is that achievement in a world where everyone is in the same bucket? Watching discreet rank buckets go by (and feeling that progression) and feeling the thrill of reaching a new rank that you've never hit before... those are pretty important, I think.