**Mastery Loadouts**
Due to issues related to the release of Mastery Loadouts, the "free swap" period will be extended.
The new end date will be May 1st.
Due to issues related to the release of Mastery Loadouts, the "free swap" period will be extended.
The new end date will be May 1st.
Fix Battlegrounds in three easy steps (that we can argue about until the end of time)
This discussion has been closed.
Comments
Special treatment is just that. It’s special, and by definition not fair. I think most people would be willing to make the very reasonable compromise to allow this unfair special advantage under special circumstances, I’ve laid out my own version of this, but i also think anyone unwilling to even call it an unfair special advantage in the first place and acknowledge the concession as a concession is someone we will ultimately have to abandon in the conversation, because without an honest appraisal of what the real world trade offs are, there’s no way to consider reasonable alternatives or the cost benefit of any suggestion.
This bears staring directly. ELO matching does not allow matching players so unbalanced that one side has no chance of winning. The only time that happens is when the ELO rating itself is horribly wrong. ELO matches players that have a 50% chance to beat each other. That’s it’s mathematical foundation. If I wanted to be as callous as some others, I would say anyone opposed to ELO is probably someone currently beating up inferior competitors at high percentage and is afraid of facing equal competition.
Let’s first define what we mean by a fair match up. My definition of a fair match up will be this: a fair match up is a match up where both sides have a roughly equal chance of winning before the match starts. I think most people will agree that’s pretty reasonable.
But how do we find these matches? Well, now we have to make some assumptions. Let’s first assume that everyone has some actual intrinsic strength that represents who they will beat more often and who beats them more often. Higher strength means they beat more people, lower strength means more people beat them more of the time. That also seems reasonable (in strict mathematical terms this isn’t a given, but for our purposes it’s fine).
Let’s also make one other assumption before we proceed. Let’s assume the notion of a fair competition is even possible. For this to be true it must be the case that there exists a way to sort the players by strength, from the strongest to the weakest. Again, in mathematical terms this isn’t a given. For example, suppose we have only three competitors, A, B, and C. And A always beats B, B always beats C, and C always beats A. In this case, strength is not transitive. A beats B and B beats C but that doesn’t mean A beats C, and thus there is no fair way to sort the competitors in order by strength. So in a sense there’s no fair competition really possible here. Let’s ignore this for now.
Okay, so we have a bunch of competitors, all with different strength. They all compete against each other, and in theory there is an order that would place them from strongest to weakest. How do we get there?
Let’s pick a random player and arbitrarily assign him a numerical rating of 1000. It doesn’t matter who he is or how strong he is. Strength is relative anyway, we don’t care how strong anyone is absolutely, we only care how strong they are relative to everyone else. We could theoretically have this person play everyone else and compare strength that way. Everyone who wins gets a higher than 1000 rating, everyone who loses gets a lower than 1000 rating. To figure out how much higher or lower, we have them play each other. Everyone who wins gets 2000 rating, and then as they play each other they go up and down, over and over until everyone has a rating that tells us who they beat and who they lose to. If we do this enough times, everyone will have a rating that tells us everyone with higher rating beats them and everyone lower than then loses to them, and everyone with the same rating wins roughly half the time. Now we just match everyone with identical rating.
Except this is impractical, for obvious reasons. We can’t actually match everyone against everyone. But we don’t really have to. We could assign everyone 1000 rating, and then follow the rule: match everyone against equal rating, increase winner rating and decrease loser rating. Now what happens is everyone starts at 1000, but over time the players who actually play at a much higher strength win more often, which causes their rating to rise until they start winning only half the time, whereupon their rating starts bouncing up and down around some average value. This causes everyone to converge on a rating that actually represents their true strength.
But even this is problematic for a couple reasons. First, if we start everyone at the same rating most players ratings will initially be “wrong” because we’re making the assumption everyone is equally strong, which they are not. Gameplay over time will correct this, everyone’’s ratings will *eventually* be correct, but for a time they won’t be, and we will get uneven matches. We can mitigate that by instituting better initial estimates for rating, and for simply not trusting rating until the player has played enough games for us to believe the rating actually measures something.
We also have to deal with a further exploit. Someone could deliberately lose a lot on purpose in an attempt to temporarily lower rating and gain easier matches. To prevent this we can institute ratings floors which guarantee to player can manipulate their rating lower than some minimum value, which we will determine based on things like their intrinsic roster strength and their prior gameplay. Once they establish they are a certain strength, we will let them “get worse” to some extent as this can actually happen in reality, but we will throttle this to realistic levels.
One last thing. Up to this point I’ve placed no requirement on the ratings numbers other than they allow us to sort the players in order of strength. It would be nice if the numbers actually meant something. Since no one always loses or always wins, we usually have the case that when a stronger player faces a weaker player, they win some percent of the time, say 75% or 90%. It would be nice if the ratings reflected that, such that we could create a calculator whereby if we plug in the ratings of the two competitors A and B, what popped out of the calculator R(A,B) was not just a prediction of who would win, but also how often. But that’s a lot of math. Let’s hand that problem to the nerds to solve and get back to us.
So we have a rating that we use to decide who’s a fair match to who. This rating number is designed in such a way that if two players with equal rating face each other, the odds of winning are about 50/50. Higher rated players are intrinsically stronger than lower rated players and will win more often than 50%. A player is presumed to have some intrinsic strength we might not know, but the more they play, the more their in-game rating will move towards their intrinsic actual strength. We ensure that players ratings change over time to match their changing real world strength by adjusting rating upward for wins and downward for losses, and we ensure that ratings cannot be manipulated beyond a certain point with deliberate losses using ratings floors and collars. We ask the nerds to make the math work out so that not only can we predict who is likely to win, but how often, using mathy math. Then we just match everyone by this rating number, equal rating means equal match.
That’s basically how ELO works.
As for the rest, we can disagree until the cows come home, but War is now centered around one demographic. The Top. The recent changes are an example. The Rewards increase, the Matchmaking allowing absolutely no reason with the sizes, you name it. It's been structured to benefit the Top, and that's it. I'm not going to debate that. It's apparent in every change that comes.
You can't do that with a PVP mode that involves Players with such a range. They will stop playing it. They will stop caring.
I wasn't rebutting your suggestions. I was responding to the OP's suggestion to have random Matches from the beginning. I'm quite honestly, tired of people claiming others have an advantage just because there is protection from them TAKING advantage of them.
I’ve fought for years to nudge the mode in that direction, sometimes in ways other players liked and sometimes in ways where the majority of players thought I was insane. And I wasn’t the only one by far. But however it got there, it is there now. The same thing can happen to Battlegrounds. It won’t get there because I say so, and it won’t likely get there in exactly the way I say so, but the point is it can get there, because if there is at least one way to get there, it is by definition not impossible.
But we have to be realistic about how we get there, and what the trade offs are going to be to get there. We can’t keep backsliding into moot arguments over how things are unfair in general. It just puts us all back to square one every time the wind blows. To solve the progression specific issues with BG mode progression in the context of a progression game with progression rewards, the discussion itself has to ultimately make progress forward. Progress we don’t lose after every single engagement.
In any competition in which this option is practical, everyone faces everyone else, at least once. This is generally considered fair. It becomes impractical when the number of competitors gets too large, but I would love to see someone show up to a round robin competition and declare that they would not face certain competitors because it was unfair for them to do so. I’d like to see someone attempt to make that argument anywhere outside of an Internet forum.
We don’t do round robin because it is obviously impractical. But the fact that round robin competitions are intrinsically fair cuts the legs out from under the argument that allowing such match ups is unfair. We don’t suggest such things because they are impractical, because they hurt other interests, because they can’t be reasonably implemented in ways that eliminate troublesome side effects. But not because they are unfair.
If it was magically possible to conduct a full round robin tournament between everyone, I’d vote to do that. We’d be here until the sun burned out, so very strong magic would have to be involved, but I’d do it. And I would dare anyone make the argument that this was unfair in a setting where they couldn’t hide behind the limits of Internet forums.
It isn't just a full round. It's Fight after Fight because the people who are running it competitively are doing so from the onset, and by the time everyone pushes forward, there's about a week left to progress. That is, barring anyone waiting to start themselves. Then the process is pin ball.
People get too wrapped up in the word fair, but in actuality, it is wrong. It is wrong to include Players that are using Rosters that have no chance of winning, and expecting them to continually try and fail for the ease of Players with larger Rosters. Their gaming experience, their progress, matters more than that. They're not sacrificial lambs. They're human beings trying to play the game mode.
Again, there is a fine line between competitiveness and taking the desire to even play it because no matter what these Players do, they can't win out the gate. Numerically, they can't. Unless you have someone throwing a Match or absent-mindedly not paying attention, they're never going to do as much Damage as their Opponent. The differences in Champions make it so.
Just because it's a competition doesn't mean the gaming experience of Players who are using lower Accounts is not valid or significant. That's always been my point. You cause the majority to be disinterested, and let's be honest they're the majority because the minority is the Top, then you have a dwindling system.
However, it's not about "us vs. them", and it's not about favoring one demographic over the other. It's about having a game mode that's reasonably challenging, and at the same time enjoyable, for as many Players as possible. That's fairness.
My concerns are for the system as a whole, not for my own benefit.
Any form of matchmaking needs to achieve two objectives
1. Get a large number of people to play the mode. That requires incentives to make it worthwhile for them to play - rewards and potential to progress (within the mode as well as in the context of the larger game, i.e. resources to become stronger).
2. Avoid arbitrages which create perverse incentives. Such as sandbagging or encouraging weakening of rosters which hurts the game in the long run. This requires that the overall efforts of players is aligned with the longer term objectives of the overall game.
If you can suggest what you think works within these requirements, that would be helfpul than crying "unfair!", "wrong!" all the time.
I'm not just arbitrarily disputing ideas. I'm expressing my chief concers for whatever they implement.
VT: eliminate the coin loss for losing and slowing people down from climbing; but reduce the rewads to half. The other half can be given in a chain of x number of objectives that goes as follow...
Win 2 in a row, win 3 in a row (same as climbing the current ranks)
Since VT doesnt punish losing anymore...only give points toward Solo and Alliance events for WINNING, take away the playing and finishing points. Also include minimum participation ammounts in Alliance event to receive rewards.. same format as old gifting event.
I understand it seems cutthroat, but yeah I dont like leeches and people trying to just get free stuff for tapping a screen a few times.
People have voiced general concerns as well as voiced valid suggestions in this thread. While your suggestion is a viable option that could be considered, the tone and attitude behind it is far off.
You can keep the win-loss system from gold tier onwards, maybe with the revision to the number of token as per your suggestion.
I never said it was an issue. I asked why the amount of Trophies people are earning is an issue when the Store limits what they can buy.