Today there was a fight on Balak 4 between the alliances Best in the South and Regional Powers. During the fight, Golems stopped spawning on the attackers inhib (RP) making it impossible for the Defenders to advance the objective. Staff then terminated the fight because I and Crypt have informed them that if there is ever a major bug that affects one side fights should be terminated. Neither of us was around for this fight, but staff did everything correctly based on our policy regarding fight terminations.
However, due to the specific circumstances of this fight, me and crypt have since looked over the data and what happened to decide if we should amend the outcome. Regional Powers were clearly ahead in this fight and on course to win, so looking at the bigger picture, the fact that their golems were not spawning on the inhib made little to no difference to the overall result. This puts us in a difficult position when analyzing whether this fight should remain terminated or the win given to Regional Powers, as that would have been the outcome if the fight had not been terminated (irrelevant to the bug). If we're being completely honest we could make an argument, either way. We tend to always terminate fights when one side encounters a major bug like that, so in that sense, it is consistent and fair with what we normally do. However, there is also the common sense angle, and looking at the specific circumstances of this fight (with data showing there was only going to be one winner) we could say the bug did not have a real/large impact on the fight. While it would be easier for us to simply rule by the book and say there is a bug affecting just one side so termination is fair, it does not feel within the spirit of conquest. We, therefore, feel the most logical, correct, and common sense thing to do
is award the win to Regional Powers, as this would have been the result had we not terminated the fight. That puts
Regional Powers on 109 Strength and Best in the South on 97 (this will not show up on live immediately but we will re-run the calculations should any fights happen before the new numbers are pushed)
.
We appreciate that whatever way we rule on this, one side/alliance will feel aggrieved and hard done by. We apologize for that and are aware this issue stems from bugs in conquest. We believe it is one of the remaining bugs introduced by sharding the server (Putting each world on its own PC). With conquest being in a more polished and stable state these days, once these ones are ironed out, conquest bugs should become far more infrequent. Unfortunately, though, bugs are an inevitable part of coding/video games, every Minecraft update, paper update, or patch we make to Loka has the possibility of breaking something. Even major game releases are full of bugs and they have million-dollar budgets and thousands of people testing them. While many of you will not have much experience coding, those that do will understand the difficulties with it. The same goes for fixing bugs, as those that seem simple on the surface are sometimes far from it, and some fixes can even end up creating new bugs. What we can do though is try to fix bugs as soon as we are made aware of them and resolve any issues that those bugs caused. Obviously, we have very limited resources but I promise you we do try our best.
This opens up a conversation that we have had in the past with the community and LCRs, which is how we should handle bugged fights. We effectively have 3 options:
- Do not interfere with fights at all and let the RNG of bugs decide what happens. If you get unlucky and encounter a bugged conquest fight, too bad.
- Pro: No possible bias or human error.
- Con: The least fair and will result in the most 'wrong' decisions/outcomes. People may be able to abuse bugs to manipulate the results of conquest fights.
- Put in a strict set of rules that cover every eventuality we can think of and terminate fights that fall under those categories, regardless of context.
- Pro: Minimal chance of bias and human error but will still require interpretation of rules. Will result in the most extreme unfair fights being terminated.
- Con: No room for common sense to overrule so some decisions will appear dumb and illogical.
- Use common sense along with a guideline of rules that dictate what sort of things we are willing to terminate fights for.
- Pro: Should result in the most 'correct' and fair decision.
- Con: Maximum chance of subconscious bias and human error.
We are more than happy to talk about this again. The last time it was discussed the community landed on option 3. Back in the day we had no concept of terminating fights and option 1 was what always happened. There are of course pros and cons to each and everyone has different goals/motives so will likely gravitate to different options.
We want to terminate as few fights as possible, a balance between competitive fairness and fun. People put a lot of effort into preparing for fights and it sucks if it was all for nothing because the fight cannot go ahead. There are also plenty of peeps that show up for fights purely for fun and they would prefer if fights continued even if they are disadvantaged by bugs. This is why bugs that affect both sides we usually let slide. Obviously, if you are hyper-competitive you can argue these bugs could suit one sides fighting style more than another, but we feel that crosses the fun to competitive balance line. The same goes if a small handful of people are unable to warp from both sides. I suppose you could even make the argument that terminating a fight for one of these reasons would advantage one side over the other because they have fewer opportunities to gain strength if they are behind, it opened up their vuln, gave away a specific strat etc.
We also don't have every tool under the sun at our disposal. Things like 'simply' restarting a fight or refunding attacks may not be options we or staff have. As time has gone by we have added more and more tools to deal with potential issues that could arise in conquest to help smooth them out, but as with everything they take time and effort (and in this case for things that should not happen very often). Conquest is a complex feature so unless we have specifically coded a tool to do something we often can not do it at all (not even manually).
It is also worth noting that holding off fights until the end of the month runs a risk. There is always a chance a bug pops up or the server is ddossed etc. so that must-win fight may never get a chance to happen. Obviously, we want to avoid these issues cropping up but you can play it safer by getting a lead earlier in the month (if possible, we respect sometimes it is not).
In conclusion, we apologize for the bugs that have occurred in conquest recently as they are far from ideal from all party's perspectives. We will try our best to eradicate them and want to open up the discussion with the community on the best way we can deal with the incidents if they crop up. We'd love to get everyone's feedback on the topic.
