Let's Improve The Brawling Experience Together!

in #hive-13323last year

In recent months, there has been a lot of discontent about how Brawling works in Splinterlands. While there's a lot of discussion happening, nothing has been changed thus far and as several members of the team made it clear repeatedly, there are no major changes coming unless a DAO vote happens:

image.png

So with this post, I want to get into a discussion about what's working and what's not working in Brawls currently. This is not an official proposal, as I think it's important to first find some consensus and discuss different options available to us. That said, if we do come to a consensus, I'm more than happy to put up a formal proposals to get things changed.

Also, while I think that Brawls do need an overhaul, I don't want to waste too much of the teams time on this. So while I personally would love to see bigger changes, like changing the tier system to an actual league system with relegation in place instead of guilds choosing their own tier, I don't think that it's feasible to invest the effort to set this up. So instead, I'll try to provide ideas for (hopefully) small changes that should help address the major issues.

Lastly, at least to my understanding, the DAO has no say over how Brawls are designed, so any change proposed will still have to be accepted by the team which again makes it very important to find solutions that have a broad consensus behind them.

With these points out of the way, let's get into the actual details. I want to go over the issues I see with the way Brawls are set up right now, describe where I see problems and then provide possible solutions. Those won't be fully fleshed out, often missing exact numbers, but they could possibly serve as a basis for future discussion.


The main issues I see with Brawls is that there are not enough incentives for guilds to play in the highest tier possible to them. We've seen this every time the first guilds managed to upgrade their arena to a new level and we are seeing it very much now with tier 5. A lot of the top guilds had a level 9 arena for months now, but there's still no tier 5 Brawl happening at a regular basis. This issue is actually made up of two problems that both contribute to the overall effect:

  • If there are less than 10 Guilds in your Brawl, you will receive fewer Crowns and Merits
  • It's way more profitable to go second or third in tier 4 than it is to go 6th or 7th in tier 5

Let's take a look at the former problem first. Brawl rewards consist of 3 pieces: Staked SPS, Merits to buy Gladius packs, and Crowns to upgrade your guilds buildings. While SPS is taken from a fixed pool and distributed among all players in a given tier, the amount of Crowns and Merits is dependent on how many matches are played in the Brawl and thus on how many guilds are competing in your Brawl.


image.png

Because of that, in terms of Merits and Crowns, it is not worth it for the top guilds to move to tier 5 if nobody joins them there. Less guilds in the Brawl means less battles fought means less Merits and Crowns earned. So in effect, instead of having a somewhat fair Brawl between the few top guilds, it's more rewarding for them to stay in tier 4 and ideally dodge each other, providing each with a comfortable first place finish and thus a higher reward than they would have gotten in tier 5.

The other issue that comes with this is that the number of guilds in any tier directly effects the amount of Crowns and Merits everybody can earn. In the current cycle, there are 43 guilds in tier 4:


image.png

Because of that, there's a total of 5 brackets, made up of 7 or 8 guilds each. If there were only 40 or a total of 50 guilds, each guild would play in a bracket with 10 guilds, thus increasing the amount of Crowns and Merits earned quite dramatically. To me, this doesn't make sense.

So how would we fix this? I propose to change the Crown and Merits rewards to a fixed amount only depending on the tier you play in and not taking into account how many (or few) matches are actually played during the Brawl. The deciding factor thus wouldn't be how many matches you have won but what percentage of matches your guild has won. Let's provide a very simple example that I'm going to iterate on later:

Let's say the Crown rewards are set to 1000 Crowns in tier 5.
If you won all your matches, you'd receive 1000 Crowns as a reward.
If you only won 50% of your matches, you'd receive 500 Crowns.
If you didn't win a single match, you'd earn 0 Crowns.

So the highest amount available would always be fixed per tier and it wouldn't matter any longer if you only faced one guild or nine other guilds.

For this to work, we'd also have to change how frays that are not filled would be treated. Right now, they are simply ignored and not counted for the Brawl. Moving forward, they would have to be counted as a loss if your guild didn't fill that fray. I know that there are a lot of guilds struggling to fill all frays. This is mostly because there are a lot of gold foil frays, frays that only allow Alpha cards, and so on. I feel like this absolutely needs changing and is another major issue. The team already told us that they'll be looking at fray composition, though, so I don't want to touch on that part for now but am hopeful that this will be corrected anyway.


Then there's the issue of higher tier Brawls not being worth it for guilds that will go last in the higher tier but would fight for the top spots in the lower tier. As you can see in the reward page I've shown above, there's a dramatic difference in rewards between the top finishers and "the defeated". This issue wouldn't be addressed by the changes proposed above and indeed, that's why I would prefer to change Brawls to a relegation system where playing in a higher tier would always be better but you'd have to earn the right to move the next tier and had to fight to defend that right. As that's not going to happen, we will need to come up with a solution that still provides incentives to move up in tier, even if the guild is probably not going to compete for the top spots.

One way to make that happen would be to make the possible Crown and Merits rewards considerably higher for the higher tiers. Here's an example with totally made up numbers:

Tier 1 - 100 Crowns
Tier 2 - 300 Crowns
Tier 3 - 600 Crowns
Tier 4 - 1000 Crowns
Tier 5 - 1500 Crowns

These numbers would have to be tweaked obviously and there are people way better at this than me, but ultimately the idea would be to provide enough incentives to compete at the highest tier possible to you instead of staying lower just to have higher rewards. On top of that, we could provide a flat bonus for guilds that choose to compete in higher tiers. We could give 100 Crowns as a flat bonus to every guild that won at least 20% of their matches in tier 5 and maybe 50 Crowns in tier 4. This should be achievable for all guilds and would (hopefully) prevent any abuse of the bonus.

It would also address another issue that I'm seeing more and more - namely that the top guilds are growing stronger way faster than the rest of the pack can keep up. In parts, this is to be expected of course, but right now, it feels like the differences are too extreme. So by having that flat bonus for competing in higher tiers, we could reduce the difference between the top spots and the rest a bit.

Another option would be to change the Crown and merits rewards to logarithmic growth:


image.png

What that means is that the Crown reward would at first grow (almost) linear with an increasing win rate but it would flatten out the higher the win rate gets. Every additional win would still provide a bigger return, but it would stop guilds from getting ahead too quickly. Taking the example from above, again using made up numbers, things could look like that:

Let's say the Crown rewards are set to 1000 Crowns in tier 5.
If you won all your matches, you'd receive 1000 Crowns as a reward.
If you only won 50% of your matches, you'd receive 750 Crowns.
If you only won 25% of your matches, you'd receive 500 Crowns.
If you didn't win a single match, you'd still earn 0 Crowns.

Personally I feel like this would be the best solution to address all of the above issues and should really help to make playing in a higher tier more rewarding for all participants.


So, these are my ideas and some possible solutions we could try to implement. Personally, I feel like having a fixed amount of Crowns per tier, given out at a logarithmic scale, and maybe adding a (small!) flat bonus for competing at higher levels could really improve the experience for all participants, no matter which tier a guild is playing in right now. But as I said, these are just ideas and possible solutions I came up with. I'm sure there are more and possibly better ideas out there, so I just hope we can really start a discussion on the topic and come up with a solution that the majority of the stakeholders feels confident with. Feel free to leave comments with suggestions here or just message me in Discord. Ideally, though, we could get the discussion going in the Brawls channel itself so as many people as possible can take part in it.

Thanks to everybody who made it through this wall of text and thanks for any feedback, it would be much appreciated!

Sort:  

This post has been supported by @Splinterboost with a 15% upvote! Delagate HP to Splinterboost to Earn Daily HIVE rewards for supporting the @Splinterlands community!

Delegate HP | Join Discord

Nerd.

How many hours a week do you spend on SPL,, 10-20?

image.png

Playing I'd say 10-15 hours a week.
Reading Discord, watching Youtube videos... a lot more ;-)

Damn, that is a part time job worth of time. You must have one of those famous trees in your garden, eating Apples that create extra hours in a day.

image.png

I would love to see a relegation system. I do not see, this coming anytime soon.

Waiting for that, the proposals you made are all valid and I would support fixed crowns per Tier and like the way to distribute them.

This would lead to Raptors joining T5 again.

The relegation system would be my dream solution but ya, I just don't think there's the time to code it right now.

Thanks a lot for your feedback, I'd love to face you guys in T5 again!

Those are very valid points. Thank you for going this step @khazrakh and making a post about this topic.

I really do hope the fray changes will look how most of us wish and won't just be adding "Rebellion Only" or similar.

I'd love to see 1 or 2 Regular Foil Bronze Frays in Tier 5 for example. In any case, a reduction of the Diamond Frays would be really welcome.

For the rest I totally agree with what @kheldar1982 mentioned and also hope we'll have T5 brawls on a regular basis soon!

That is so annoying that we can't play tier 5 brawls, because for many guilds more profitable is to play tier 4 ;(

Thanks for bringing this to the table! Brawls have been skewed for far too long. Glad to see this finally being addressed!!

Thanks for your feedback, I really hope we can find a solution together :-)

My general thoughts on this are that changing the economics/distribution of tokens for brawls wouldn't be a huge ask. If we're going to change anything beyond economics I'd say the priority is pretty low. There's just bigger fish to fry right now and not enough dev time. That said, if you want to get a discussion topic going in the DAOscord ( :D ) I can get one set up and we can all discuss how to adjust the distribution to give more incentive to move up in rank. If we get a good soft consensus then we can make a proposal and see how it goes.

Thanks a lot for your feedback! I agree that anything beyond just changing the formula would be a low priority for now. I'd love to see Brawl 2.0 some day but we can talk about that once we figured out the more pressing issues ;-)

Would be awesome if you could set up a discussion topic, I've been talking with a lot of people already and I think there's a lot of interest in changing things, so getting the discussion going there would be great.

I like the idea of non-linear crown distribution, especially the version suggested by @kalkulus. And tbh I think changing that one aspect could improve player experience a lot and encourage guilds to move to higher tiers. I don't think there is enough dev time available to make huge changes, so it would be nice to get the most out of this one change.
The only drawback I can see is that ... it would be difficult to explain to average player how that "hyperbolic tangent" distribution works (but hey, it's not easy to explain current way either!).

Also I am not a fan of relegation system, the problem I can see that e.g. guild who has exactly X players to fill X frays will be forced to fill X + n frays and different levels of them. Sometimes guild is just not ready to advance even though that guild wins brawls in current tier. But maybe guilds would just adapt.

Great post overall, I hope we can find some consensus. I will definitely support the proposal.

Thanks for your feedback!
After I got a lot of positive responses from different stakeholders I now contacted Yabapmatt directly to see where he stands on the matter and if he'd be willing to spare some dev time to make changes. If so, we can start working on the exact formula.
I do agree that it won't be super easy to explain how it exactly works but as you already said, most people won't understand how it's working right now either, so that shouldn't be an issue.

Regarding the relegation system - it certainly would need a lot of work and there are many questions that would have to be answered first. I still hope we'll see Brawl 2.0 eventually with a new system... that might be something for 2025 or later, though ;-)

Yeah, I agree that hyberbolic tangent doesn't sound very nice :)
But I think the explanation problem can be solved by just providing a table with example payouts at different win percentages. This seemed to work for the sps multiplier, and that equation is way more complicated!

Splinterlands has a very diverse crowd of players. Brawls have always been a unique opportunity to let low cap players have additional achievements.

I'd simplify and streamline the process.

  1. Auto-matchmaking
  2. Ghost Summoner on Gold level for everyone (you can use higher if you own them)
  3. Daily rewards for activity

Now you have a Guild Brawl system that engages everyone, people will rent Monsters for specific matches, small players grind daily, big players can't dodge anymore.

Thank you for this initiative @khazrakh!

You've done a good job at describing the problem, and your suggested solution looks good to me. 👍

Since I'm a math nerd, I have to poke a little bit into the crown reward structure.

I suggest a change from logarithmic growth to a hyperbolic tangent structure. The reason for this is that unless you scale the crowns given out in a brawl tier depending on the results of all the guilds, its hard to make the logarithmic growth agree with the fixed amount of crowns awarded per tier. If we use a variant of a hyperbolic tangent instead, then that becomes much easier



The key is that this function is symmetric around 50%. If the crowns are based on any function like that, then the total number of crowns given out in a brawl group is simply the value at 50% multiplied by the number of guilds in the group (since the average winrate necessarily becomes 50% when you look at all the guilds). Of course, a flat bonus can be added to that, and the parameters in the function can be adjusted (apart from the number 50, which ensures the symmetry). This gives Splinterlands direct control over the amount of crowns they want to give out per tier.

Edit: I had written merits instead of crowns several places. Also, the claim above that the total number of crowns given out is the value at 50% times the number of guilds, is only correct when there are 2 guilds in the group. For a larger group, the result is not exactly that, but its close because the function is approximately linear around the center.

Love it, thank you so much for the reply!
It's been a while since I had to deal with any "real math", so I'm super grateful for the input! If we do go to the proposal stage, we will have to make sure that the new formula is as exact as possible so that people know exactly what they are voting for and that's already way better than what I could have come up with!

Ok, I have more questions than solutions really, my thoughts about this are quite simple as math has never been my strongest point to be honest:

  1. How does exactly this encourage to raise to higher tiers?, you would have a fixed pool that I'm guessing it's always higher in the top right?. But, would be profitable for guilds winning 50% of the matches?, instead of those winning 80% in a lower tier?, also, if I'm not wrong this would nullify the utility of the current Arena, as now has influence in the amount of crowns that are in play in a certain group?. Or it's added up on top of the arena bonuses?.

  2. I Like the option of a formula with a logarithmic grow, Kalkulus's suggestion seem also quite nice, maybe this would be fairer but I'm guessing that we could extrapolate this examples to a current brawls to see their results that would have with these formulas.

Because for not math nerds as me it's quite difficult to see how this would play out, and I'm guessing that there is people not commenting just because it's hard to understand for them.

So an example extrapolating this to one real brawl could convince people to vote a potential proposal.

Thank you very much for feedback!
I'm not exactly a math guru either and I agree, it's hard to grasp without seeing the numbers, that's why I tried to give some simple examples.
I hope to collect more feedback for another day or two and then try to come up with an actual formula with the help of @kalkulus and then pitch it to Cryptomancer and Yabapmatt. Best case they'd be willing to work out the exact numbers together.

You are correct about the Arena, though, we will have to figure out how that's going to work with the potential new system. Maybe it could be linked to the flat bonus for playing in higher tiers.

Frankly, I think it won't be easy to come up with an ideal solution, but I feel like we need to figure things out or this is going to haunt us for years to come.

Lets take this brawl as an example to calculate the new rewards:
https://splex.gg/guilds/f1840e0d297c7a6aaf4fbb9dd68dd88d2b60af2c/brawl/GUILD-BC184-BL78-BRAWL2

To set the scale of the new function I took the total crown payout with the old system and divided by number of guilds (that gives roughly 100 crowns). Then the new function is crowns(x) = 200*(1+tanh((x-50)/25.))/2.). So with this function, the maximum payout for this Tier 4 brawl would be 200 crowns, and the payout at 50% winrate is 100 crowns.

Comparing the old and new crown payouts:

GuildWin PercentageOld CrownsNew Crowns
The Guild of Neoxian82207186
PeakMonsters Pirates57142126
YGG Brawlers (Ásgard)55121120
✨[RKO] Outta Nowhere 🧨55104120
Team Rocket448377
Roaring Twenties387155
Praetorian Pandas315836
第一公會 #1 GUILD305734
Immortal Phoenix10198

I hope that makes it a little easier to grasp.

It does!, ty very much.

Congratulations @khazrakh! You have completed the following achievement on the Hive blockchain And have been rewarded with New badge(s)

You published more than 250 posts.
Your next target is to reach 300 posts.

You can view your badges on your board and compare yourself to others in the Ranking
If you no longer want to receive notifications, reply to this comment with the word STOP

To support your work, I also upvoted your post!

Check out our last posts:

Be ready for the November edition of the Hive Power Up Month!
Hive Power Up Day - November 1st 2023