Changes To Scoring In Versus Events

Options
1…1213151718…27

Comments

  • Unknown
    Unknown
    Buret0 wrote:
    Yes, but doesn't this ignore shields?

    When you beat someone who is shielded, you are creating a ton of extra points into the slice because the gains are now bigger and the loss remains at 0.

    So if people are really interested in feeding the slice as a whole, they can climb up quickly, save a seed team, beat the seed team with a really weak party, and throw up a 24 hour shield.

    People will be able to climb off that person at a higher rate than before (up to 75 points per win) and no points will be removed from the economy of the slice because the shield will protect them.

    I don't know devs... sounds like you haven't really fixed the problem.

    1. This is a completely different old problem, and not really much of one.
    2. Doing so limits the choices the player has during their climb, making it slightly slower.
    3. After a person is shielded, they are no longer able to be queued. People who have this person already in the queue can attack this person, but that requires coordination and risks leaving yourself unshielded with a weak team for too long.
    4. Most of the retals will still be against the strong team.

    If you want to inject points, I think it's still faster to continually lose battles at 0, even if you are only putting in 13 additional points per loss.

    Any losses above 333 will take away points from the system.
  • Buret0
    Buret0 Posts: 1,591
    sc0ville wrote:
    And someone climbs very high, early, by what? hitting 3-4 point targets that are worth hundreds of points less than them while getting hit for 30-50 points a pop?

    Lol, it was very possible to get to 400 points with a 2-3* transition squad in the first hours of the event without getting hit.

    That 400 points will now be somewhat higher because of the increase in scoring. Which means that you will not be alone in having more points, which should give you more points to climb on as well.

    Obviously we will see what happens when the changes go live, but I don't foresee it being a challenge to get to 650 in the first four hours of the event.

    You could also stagger the climbers. Have someone get to 350 right away and drop a three hour shield with a weak team out. Hit that shielded target to help you climb, then have the next person drop their seed team and shield a weak team at 600. Then have a third team climb to 800+ in the next three hours using the shielded 600. Shield a weak team at the top and let the whole slice benefit.

    If it is really the intention of boosters to develop the number of points available in a slice, this is certainly a quick way to do that.
    daibar wrote:
    3. After a person is shielded, they are no longer able to be queued. People who have this person already in the queue can attack this person, but that requires coordination and risks leaving yourself unshielded with a weak team for too long.

    Considering how many times I get hit while shielded, I actually didn't know that.

    Must just be a ton of retaliations, which would not have my weak defensive team out there before shielding, it would have whatever team I used to climb.

    Hmmm... well, I guess there are holes in the plan, but hitting someone who is shielded, especially on retaliations, will still add a ton of points into the pool.
  • fmftint
    fmftint Posts: 3,653 Chairperson of the Boards
    One thing I haven't seen considered, what is this going to do to the 2* and transitioning player? Right now they are insulated from high level attacks, until they cross a certain point threshold. They too will be getting 50% more points, exposing them earlier and longer than before. A T100 finish may become impossible for them once again. But hey, they got 1-2 more progression rewards...
  • aesthetocyst
    aesthetocyst Posts: 538 Critical Contributor
    Buret0 wrote:
    aesthetocyst wrote:
    EDIT: Should have noted that pts being "minted" at the low end is nothing new, it's always been that way, and has to be to get the event off the ground and assist players in climbing. What's new is that the recent change (dividing loss pts by 1000 instead of 800) increased this "minting" ... and the following...

    Now perversely, beginning at 333 on up, if an attacker loses, pts will be destroyed! (Well, it's perverse IMO)

    If a player at 400 attacks a player at 400, and loses, attacker loses 15, defender gains 13, 2pts go *poof*.

    If a player at 1001 attacks a player at 1001, and loses, the attacker loses 38, the defender gets 13, the slice as a whole loses 25pts.

    Considering the 9-1 claimed ratio of wins to losses (which would have including all boosting that had been going on?), and the small portion of PVP action that occurs at high scores, the pts destroyed will not balance the pts created, not even close, thankfully. But this side effect seems really weird and undesirable to me.

    Yes, but doesn't this ignore shields?

    When you beat someone who is shielded, you are creating a ton of extra points into the slice because the gains are now bigger and the loss remains at 0.

    So if people are really interested in feeding the slice as a whole, they can climb up quickly, save a seed team, beat the seed team with a really weak party, and throw up a 24 hour shield.

    People will be able to climb off that person at a higher rate than before (up to 75 points per win) and no points will be removed from the economy of the slice because the shield will protect them.

    I don't know devs... sounds like you haven't really fixed the problem.

    Pts being created when shielded players are hit is nothing new, I assume this is why shielded players supposedly can't be qd, because if they stayed in the pool, the slice would overheat. (Would that really be a bad thing? Sounds more fun that a dry slice, and, having tried the misery of a dry slice, I say let's give overheating a shot icon_lol.gif )

    I say supposedly, because the same targets are appearing in qs, endlessly for hours. Enabling players to hit the same target multiple times, shielded or not. Why on earth should anyone ever see the same six targets fight after fight after fight? Being hit repeatedly by a player happens every tournament. Even multiple times long after shielding. I used to think they were jerks. Now I know they are most likely just stuck!

    This is also why alliances bounce off of each other. The PVP design seems to assume players don't known when others are shielding/unshielding. But so long as qs are so limited, and alliances able to target their members, they have to communicate to avoid friendly fire. (Ever been stuck being able to q only friendlies? I have. Awkward.)

    As for putting out a weak team and shielding, you have to chance hanging out unshielded for a few minutes to work into some qs. People who qd you during that fight would see your old team.

    As for climbing faster, everyone is hoping so. But see my posts above for my concerns. We don't play in a vacuum—one of you versus many hundreds of us—and you can only win one match at a time. Faster you climb, the faster and more often attacked. I've told my guys this means redoubling down on shield and hop discipline to expect the same results. And the 1-fight hop is going to gain the status of a religious ritual icon_lol.gif
  • aesthetocyst
    aesthetocyst Posts: 538 Critical Contributor
    sc0ville wrote:
    And someone climbs very high, early, by what? hitting 3-4 point targets that are worth hundreds of points less than them while getting hit for 30-50 points a pop?

    First, join literally when the event opens, and collect 150-185pts from seeds. Then keep on going, hitting your fellow early climbers. Use the skip button to find weaker rosters. Assuming you have a fairly strong roster (highly leveled feature, strong 3*s or better to team with), you won't take too many hits, and thanks to the discounting (pts lost x score/1000!), hits below 500 don't hurt much. Retaliate for profit. Some players will retaliate right back. Great! Hit'em right back. Hopefully they realize each cycle is a net gain.

    Continue on to 600. Shield, recharge, communicate with allies to coordinate hops and targets. Hop to 800, then 900, then 950, then 1000. 1-fight hops after that if fighting for placement or refreshing a shield.

    It's very possible to hit 1K in the first 12hrs if you're in an active, coordinated group, and in a slice with other competitive alliances.

    And, if you have a stronger roster (strongest toons in that event all maxed), you can cruise up to 800, 900, even 1000 or more before shielding. But no one does it alone. Takes friends.
  • Wonko33
    Wonko33 Posts: 985 Critical Contributor
    I think the rhetoric from top alliance people here is that they were forced to use boosting because of the changes to MMR (it was done before but not on such a scale I gather), but this caused an issue where the Devs did not see the bad effects of the MMR changes because they were masked by the artificially boosted scores so we all had to live with the bad changes for months. Now they are thinking what's up with all the boosting? But make changes that might affect the symptom more than the problem. We'll see in a few days I guess
  • SnowcaTT
    SnowcaTT Posts: 3,489 Chairperson of the Boards
    edited May 2015
    Wonko33 wrote:
    I think the rhetoric from top alliance people here is that they were forced to use boosting because of the changes to MMR (it was done before but not on such a scale I gather), but this caused an issue where the Devs did not see the bad effects of the MMR changes because they were masked by the artificially boosted scores so we all had to live with the bad changes for months. Now they are thinking what's up with all the boosting? But make changes that might affect the symptom more than the problem. We'll see in a few days I guess

    I assume everyone is boosting more. I basically never boosted for a year+, then they rapidly changed MMR, more boosted characters, and more health.

    Each of those made matches much longer, and hard to come out of without severely damaged characters. All together - the most "cost effective" way to keep playing anymore is boost heavily. I basically use up all the ISO I win in DDQ in boosts, LR/DDQ are about the only rounds I -don't- boost in anymore.

    Edit - Duh, Wonko is talking about teams boosting off of each other, and that opinion is unpopular. I'll back it - I'm not a fan of anything that needs out-of-game communication to really pull. I've said for as long as I've seen the idea: pull the name out of the matchup, and it eliminates the "megateams" missing each other.
  • Unknown
    Unknown
    fmftint wrote:
    One thing I haven't seen considered, what is this going to do to the 2* and transitioning player? Right now they are insulated from high level attacks, until they cross a certain point threshold. They too will be getting 50% more points, exposing them earlier and longer than before. A T100 finish may become impossible for them once again. But hey, they got 1-2 more progression rewards...

    While these changes may have a different impact on the top10 players compared to the top50 players compared to the top150 players, it should have a relatively small impact on most peoples placement, as the people that have been placing around you should be affected in a similar manner to how you are affected.
  • Ruinate
    Ruinate Posts: 528 Critical Contributor
    Could someone explain to me the MMR change that everyone talks about? It must have happened while I quit the game and it's now old enough where it's buried in the forums. I tried the search function, but no luck.
  • SnowcaTT
    SnowcaTT Posts: 3,489 Chairperson of the Boards
    Ruinate wrote:
    Could someone explain to me the MMR change that everyone talks about? It must have happened while I quit the game and it's now old enough where it's buried in the forums. I tried the search function, but no luck.

    While devs refuse to give out much info, you used to see teams based on your placement, win/loss ratio, perhaps points in event. So you'd open up seeing 2* teams and wouldn't see max level teams until you broke 600 points or so.

    But since it used some win/loss and placement info, you could intentionally lose (tank) or skip events and you'd be in with 2*'s farther along (supposedly, I never tried it or saw it).

    Here is the announcement post:
    viewtopic.php?f=13&t=27106
  • CrookedKnight
    CrookedKnight Posts: 2,579 Chairperson of the Boards
    There was one (relatively) well-known case of a player who managed to manipulate his/her MMR to reach four-digit event scores while remaining effectively invisible to incoming attacks, which I would bet is what turned out to be the straw that broke the camel's back.
  • aesthetocyst
    aesthetocyst Posts: 538 Critical Contributor
    Ruinate wrote:
    Could someone explain to me the MMR change that everyone talks about? It must have happened while I quit the game and it's now old enough where it's buried in the forums. I tried the search function, but no luck.

    Hey Ruinate ... since early 2014(since shields were added?) PVP matchmaking had been very different below 600 vs. above 600.

    Below 600 you would typically only encounter 2* rosters and transitioners. Because of this, the first 600pts were supposedly "free" for 3* players.

    At 600, you would suddenly see anyone and everyone who was above 600 in your slice, and join the heavyweight fray.

    In late March, they announced some changes, which went live in early April with a season-opening, IM40 Heavy metal PVP. In this event, seed teams were much reduced, and from the git-go you were faced with your peers. It was quite a shock.

    The following PVP, it was switched back suddenly. Then, the following week, to the somewhere in between that we have now.

    See this poll from mid-April for a discussion/review of the whole sequence. You might try searchig the forum for matchmaking or Heavy Metal and look for late March/early April posts.

    viewtopic.php?f=7&t=27905&hilit=heavy+metal
  • Moon Roach
    Moon Roach Posts: 2,863 Chairperson of the Boards
    Demiurge_Will wrote:
    Q: Can you give us the complete algorithm for figuring out how many points you’ll lose or gain?
    A: The scoring system is based on the Elo rating system, but the formula for that is too complex to put here without mathematical notation. Check Wikipedia if you’re interested. But the algorithm goes something like this:
    Start with the points change given by the Elo rating system, with a k-value of 75.
    If the loser’s score is <=1000, multiply their loss by (their score / 1000).
    If the winner was the defender, divide their gain by 3.
    Round the result.

    Q: Yeah, that’s too much math. How about some more examples?
    A: OK! Here you go:

    A (500) attacks B (500): If A wins, A gains 38, B loses 19. If A loses, A loses 19, B gains 13.
    A (200) attacks B (500): If A wins, A gains 64, B loses 32. If A loses, A loses 2, B gains 4.
    A (500) attacks B (200): If A wins, A gains 11, B loses 2. If A loses, A loses 32, B gains 21.

    I looked up the Elo rating system and I've replicated the algorithm in an Excel spread sheet so that it reproduces the results shown. It works fine if you know the opponent's points.

    Stage 2 is to dust off my algebra and, from the points available and your own score (the only figures available), deduce the opponent's score and how many points you'd lose if you battled and lost.

    Why bother? Because it's interesting.
  • mohio
    mohio Posts: 1,690 Chairperson of the Boards
    So here's my take on what this will do to points...

    IF people do the same number of matches as before it should inject more points into the system since each match will add more points on average. However, if people play to the same level and start getting hit like usual and don't push through, then we'll be at the same point as we are now.

    The elephant in the room is the amount of points generated by rapid retreats early on in events. We need to make up for those points by playing more matches. Overall, it seems like we might be at a dead heat in the middle of the pack, but people who are able to rapidly climb from 6-900 should benefit since they may be able to get even higher, especially if the early hoppers are not sniped too much and get their added points (and are queued by others, etc).
  • Eddiemon
    Eddiemon Posts: 1,470 Chairperson of the Boards
    aesthetocyst wrote:
    Say, if you are at 700, and q'd a player at 1000, they are currently worth around 40pts to you, give or take. By attacking them you only risk losing 7. Ignoring possible losses by being attacked while you play this match, unless it's an impossible lineup for you, it's a no-brainer! And soon they will be worth about 60, and you'll only being risking loss of about 10 or 11. Sounds like an improvement, right?

    I think the other factor is the limitation of Health Packs. So your gain is 60 but your loss is 7 points and 3 health packs. Unless you have one of those rosters with duplicates, losing your featured and your 2 best pairings with them forces you to break out the band aids.

    3 health packs is normaly worth 150-200 points in a rough PvP neighbourhood if you have a tank character you only need to heal every 2-3 fights. So that's the gamble they are taking.
  • MyNameIsRed
    MyNameIsRed Posts: 63
    Just out of curiosity, has anyone come up with a better, universally accepted solution?

    Not trying to troll, just wondering if anyone has any other ideas. I've given it some thought and every solution I think of is flawed.

    Instead of pointing out the problems (which seem well documented) how about some discussion on solutions?
  • ballingbees
    ballingbees Posts: 208 Tile Toppler
    ArkPrime wrote:
    Tredo wrote:
    Please just stop. Both of you. I'm begging. This is not productive. Your tête-à-tête is completely derailing this thread. What will it take for both of you to stop posting here until you're ready to have a discussion that actually focuses on the issue at hand?
    This is about the issue at hand. He's saying this needed to be emergency fixed because random people were boosting to 1k from nowhere at the start of the event. I know for a fact that's not the case. We're arguing about how bad the problem was and if it warranted the complete removal of defensive wins from the game as a "fix".

    Nothing personal here. I was in the same bracket in Divine as you, Ark, I saw with my own eyes. Without naming anyone, some guy shot from 750 to beyond 1k in the time I took to break shield and do two matches. Just two matches that each was worth about 20pts, and I was naturally sniped in the meantime. Not just one guy, I saw a few doing this, albeit at different times. And not random guys, the ones I saw were from the better known alliances, which is all the more disappointing.

    It is a real problem for the game that demanded an immediate fix. Whether the fix that the devs came up with is the right answer, that is another question. Probably not, and everyone will start ridiculing them for being clueless. Heck, we are already ridiculing them before we even get to try it out. At least they are trying to fix a real problem and acting fast about it. For me the Divine PVP has left a very very distasteful experience.
  • aesthetocyst
    aesthetocyst Posts: 538 Critical Contributor
    MyNameIsRed wrote:
    Just out of curiosity, has anyone come up with a better, universally accepted solution?

    Not trying to troll, just wondering if anyone has any other ideas. I've given it some thought and every solution I think of is flawed.

    Instead of pointing out the problems (which seem well documented) how about some discussion on solutions?


    Red, yes, ideas have been floated, in this thread and in other threads on the forum. This is my favorite suggestion, posted here earlier:
    viewtopic.php?f=7&t=30147&start=80#p362756

    If you can poke a hole in it or improve on it, great!

    What ideas did you have, and why did you decide they were flawed?
  • SnowcaTT
    SnowcaTT Posts: 3,489 Chairperson of the Boards
    One solution I often put out there (and others do as well) is make sure alliances can't hit each other (solves the alliance boost problem), and show only points - not names - on defenses. That would get rid of the mega-alliance boost problem.

    My biggest concern with the change isn't folks getting points, it is when they'll get them. If matches are worth more, I expect many will go strongly back to "steam-roll everyone in the last four hours" - which makes it hard to find anyone worth points for those without the best rosters that have to climb early, and makes them juicy targets as not ideal rosters with the highest points later on.
  • Moon Roach
    Moon Roach Posts: 2,863 Chairperson of the Boards
    Playing with my spread sheet, I found a few interesting things, but this one on retaliation I thought I'd share.

    Starting with A and B, both on 500.

    A attacks B and wins, gains 38, B loses 19, now we have 538 and 491.
    B retaliates and wins, gains 44, A loses 23, now we have 515 and 525.
    A retaliates and wins, gains 39, B loses 20, now we have 554 and 505.
    B retaliates and wins, gains 43, A loses 24, now we have 530 and 548.
    A retaliates and wins, gains 39, B loses 22, now we have 569 and 526.

    To cut a long story short, after 14 retaliations both are above 600, after 30 they're over 700, 53 gets them 800, 97 gets them 900, then I lost the will to live, or at least the will to copy and paste.

    Starting from 0, 18 mutual retaliations gets both A and B the token at 300.

    Completely academic and unrealistic, and of no real value, but I'm sharing it anyway.
This discussion has been closed.

Categories