A suggestion to how to fix PvP matchmaking

Marvel Puzzle Quest MPQ Suggestions and Feedback
1

In light of all the PvP changes that focuses on penalizing the player with higher ISO, HP and tournament point costs, I would like to propose a fix to inherent problems in the matchmaking system that would better balance out your competition. How about a system that utilizes the only consistent scale in this game: character levels!

What I’m suggesting is separating brackets into three listed difficulty options for one tournament that is based on the cummulative points of your team:

  • Easy - Cummulative team level points of 150 or less (no /* characters allowed)

  • Medium - Cummulative team level points of 151 to 255

  • Hard - Cummulative team level points of 256 or higher

So for example, a lv. 85 Thor, lv. 85 Wolverine, and a lv. 50 Storm would equal 220 (85+85+50=220), which means you can use this team in only the Medium bracket. These level thresholds are based on the experience of the player:

  • Easy would be for newcomers who can easily achieve three lv. 50 * characters through playing the Prologue. Even players that use a lv. 85 character can only use two other characters that add up to 65 together so it’s not terribly imbalanced.

  • Medium is for players that have gotten towards the end of the Prologue and started to get their ** characters. This bracket has a level cap set for three lv. 85 characters. You can use a lv. 115 character if you’d like but that just means your other two characters won’t be maxed out.

  • Hard has a baseline for any team that has two lv. 85 characters, and any character at 86 and up. You’ve played so long that you have a *** character that high, so you know what you’re doing at this point.

Here how the system could work:

  • When you enter one of these brackets and start picking out your team, the roster selection will dynamically “red block” characters that would exceed the max total lv. limit for that bracket.

  • The three difficulty brackets will have their own separate tournament point totals for each player, so getting points in an Easy bracket does not affect your points in the Hard bracket.

  • Each difficulty bracket will have their own commensurate rewards, i.e. the Easy bracket will have lower ISO/progression rewards and ** overall rewards, while Hard will have higher ISO/progression rewards and **** overall rewards. Having separate rewards gives newcomers a gradual climb to the harder difficulties while deterring higher players from trolling the easier difficulties because they don’t need those rewards.

  • Maybe consider making a player choose and lock themselves in one difficulty bracket to let them decide which rewards they want more. i.e. “Do I want to play with my B team in the Medium bracket to get this Punisher cover I’m missing, or fight tooth and nail with my A team in the Hard bracket to get the **** Wolverine cover?”

  • This system would eliminate the need for behind-the-scenes MMR and vicariously, tanking. You wouldn’t tank if you know your opponent difficulty is going to be of the same range if you want the better rewards.

  • Retaliations should always give you a consistent percentage of your loss back, like 65-70% back. This way a player is better able to keep his footing, while making losses bearable.

The emphasis of this system is that you’re playing against people in your skill level. This would reduce the current need to skip overpowered/underpowered teams, because each opponent is now within reason to fight. In fact, a system based on character levels almost means that there is no such thing as an overpowered team in the Easy and Medium brackets. In the Hard bracket, well that would require character rebalancing… evil glare at Spider-man Shields can go back to their original costs, because everyone being at the same level scale means they should be able to achieve that high score as well if they invest HP into shields.

I realize this overhaul might take a lot more programming, but it’s better than these quick fix-it solutions that pass the bill on to the players. What do you guys think? Any suggestions?

2

Well, just fyi, with the current matchmaking, I placed 2nd and 5th in the Rage and Ruin tourneys with a team of power lvl 160. I could easily downgrade 10 levels and wreck shop in the easy brackets.

The term used in many other games is “twinking”, which I think could apply here.

3

Well one of the points was that the Easy bracket would have lesser scale rewards. So for example,

The Hard bracket rewards would be:
Placement 1 - Invisible Woman(*), one cover
Placement 2-5 - Patches Wolverine(), three covers
—>
Placement 46-75 - Classic Storm(**), three covers
—>
Progression reward 50 - 500 ISOs
Progression reward 200 - 250 ISOs

The Medium bracket rewards would be:
Placement 1 - Patches Wolverine(), one cover
Placement 2-5 - Classic Storm(), three covers
—>
Placement 46-75 - Modern Storm(), three covers
—>
Progression reward 50 - 250 ISOs
Progression reward 200 - 100 ISOs

The Easy bracket rewards would be:
Placement 1 - Classic Storm(**), one cover
Placement 2-5 - Modern Storm(), three covers
—>
Placement 46-75 - Juggernaut(), three covers
—>
Progression reward 50 - 100 ISOs
Progression reward 200 - 50 ISOs

The emphasis being that the rewards would really be the most beneficial to the players in that difficulty level. This would help prevent higher level players from stepping down a bracket because they would be rewarded more to play in their own bracket. The Juggernaut reward in the Easy bracket might seem trivial to you and me, but for a new player that would help them progress through the Prologue faster, thereby helping achieve their first ** characters and picking up more ISOs and standard recruit tokens along the way. Higher ISO rewards in the Hard bracket means leveling up your *** and **** characters faster.

4

Those rewards sound like a nerf imo. Most easy teams should already be capable of winning multiple 2-stars and most medium teams, multiple 3-stars.

So using my 160 power team again as an example, would have to compete against 256+ power teams for the same rewards I’m already getting.

5

Assuming the balance changes come in to curb the especially broken stuff, the levels do not mean nearly as much as they do. Most of the tournaments are setup to minimize the impact of the levels. A 100% boosted level 85 Thor or Ares is kind of like having a level 240 PvE boss on your team (their stats are very comparable), and in fact having the boost on an already ovepowered character (Thor) actually levels the playing field a lot, since most people do have Thor at a high level and he easily plows through other non boosted characters.

6

You do make a valid point; you were able to get a higher reward with a medium range team. So I have no immediate solution to that. Do you have any suggestions? Or leave the system the way it is?

7

so in your system lvl85 bagman equals 85 lvl thor/vol/mw, whataver, right?

8

Well twinking is pretty common in other games, and I still enjoy them. I’m not particularly against this system, just pointing out that such a thing can become prevalent. You could say that the tanking that currently exists is similar to twinking. Which would I’d prefer? I don’t know. I’m not sure either one is innately better than the other, just different.

They are introducing a new type of tourney. If they plan to introduce more in the future, I could see a tourney utilizing a system like this could work. Personally, I think the star system mentioned in other threads would be my ideal (similar concept, but limiting stars rather than levels). Reminds me of the team dynamic used in Capcom vs Snk.

9

I think some kind of meta user-specific leveling system (transparent to the user) would have worked better. Each time you win a match, your account accrues some exp that never goes away. People of similar levels (total play time) would find each other in matchmaking. Beating people of a higher levels will let you accrue more points per win. So, a new person could theoretically win top prizes, but they would have to put in way more work (and in doing so, increase their overall user level).

You could also cordon off the events (and grant prizes accordingly) based on this meta level. Advanced users could get higher ISO rewards (in order to level their level 4 characters quicker) and lower users can compete for covers that are more apt to help them where they currently are in the game (Thors, Wolvy’s, Doom’s, Ares, etc).

10

So would the amount of ISO spent work? The levels would be hidden from the user, but spending ISO could determine your level. It seems like a direct relationship to how much you play. I guess it could discourage leveling up a variety of characters or selling off characters you’ve leveled, unless that ISO doesn’t count. But if the levels are broad enough, I don’t think it would make that much of a difference. You also couldn’t tank or twink, which could put you in a frustrating spot when you first level up. However, the higher rewards should help you out and encourage you to keep going.

11

I think if you have a lv. 85 Bagman, you probably have bigger problems…

I’m not 100% sure, but this sounds a bit similar to what I understand is their MMR system. Doesn’t it give you a ranking based on how many matches you’ve won? Isn’t that how tanking works, by damaging your win-loss ratio in order to play lesser opponents? So are you suggesting then to remove the ability to tank by making losses not affect your ranking?

And to me, having rewards decided by a hidden rating system sounds like it could incite a lot of rage from players, i.e. “How come I didn’t get this or that reward??” Having systems more clear and upfront to the player seems more beneficial.

12

I’m okay with a level-based system as long as it’s applied systematically. Don’t segregate me from a certain “rewards bracket” if i’m using my “B” team as opposed to my “A” team. Here’s how I see it. Make the matchmaking where my levels matter. If I come in with 3 level 140 covers, I’m supposed to be a good player. 3x140=420 so take that number +/- 10% and that’s the teams i’ll be paired against. That makes things fair. If I want a better chance to succeed and get more points, I’m going to have to sub in a level 40 cover so that I can compete or choose a different team altogether. This also eliminates all the “skipping” associated with finding an “ideal” match to gain my 50 points on. If I know i’m going to be matched up with equal players, what’s the point of skipping? This should be a game of skill, not just who can buy more Hero Points.

13

I think the idea was to have a total amount of experience earned from wins that can’t be reduced instead of a rating. With the current MMR system, you can tank. With this system you can’t because losing doesn’t change how much experience you have. I could be wrong on that. But that is why I suggested ISO spent. It is mostly proportional to how experienced your team is and already in place. It also allows people who pay for ISO to level up quickly, but not be suddenly overpowered in their rank. You are right though. You’d need to have a badge or rank or something so you know what rewards you are getting.

14

The point of skipping changes. It becomes an attempt to find matches where you will gain the most points, regardless of the opposing team. This is both to help you move up the ranks and to reduce the chances that you will be retaliated against. Having an equal playing field increases skipping because players have a great advantage over the AI. Players on the same level can almost always win when attacking. This would cause more skips because you’d only want to fight people worth high points.

15

The ability to skip makes it pretty much impossible to break away unless you’re completely off someone’s matchup list. You might think you’re gaining something when you skipped 20 guys to find a guy worth 25 points, but what you don’t know is that another 5 guys are also skipping 20 guys to see you show up and worth 40 points to them. It’s not really a secret that after a certain point all teams are roughly equally strong and your list of opponents becomes relatively small. There are plenty of times where I hit the same guy 5 times and I’m also hit 5 times by a different guy, because my target was the highest point on my list and I am the highest point on the list of whoever’s hitting me. Yes the retaliation from attacking a low point guy is going to be painful, but right now you’re never escaping from the guys that just skips 20 times to find you anyway. There’s also seems to be something weird with retaliation not always show up, so maybe you’ll actually escape them some of the time. You sure aren’t avoiding the guy who has you singled out because your point is the highest in the current system.

16

So to be clear, the reason why I chose levels over stars, win/loss ratio, or ISO accumulation because to me, it’s the most direct correlation to power. I haven’t sat there to analyze this thoroughly so I could be completely wrong, but it seems that two different characters of the same level should have somewhat similar basic tile attack strengths (disregarding special powers). Here’s a rough comparison of ** characters at level 85:

Black Widow (Original)
Lv. 85 tile damage: 10/11/45/50/9/39

Thor (Marvel NOW!)
Lv. 85 tile damage: 45/50/10/9/39/11

Wolverine (Astonishing X-Men)
Lv. 85 tile damage: 39/45/9/11/50/10

And just for good measure, Spider-Man (Bag-Man)
Lv. 85 tile damage: 39/11/45/50/10/9

As you can see, each character has a 50 damage, 45 damage, and 39 damage capability (yes, even Bagman). By limiting it to levels, it could be assumed that no matter the characters (again, disregarding special powers) the team’s damage output would be similar so long as they’re close in levels. When it gets to *** and **** characters, their damage outputs are a bit more scattered in range (seems like a high of 67/60/52 damage), but if you limit the Medium bracket to one max level *** character, the team damage output would still be in the same ballpark because your two other characters aren’t at max and therefore makes up for the damage difference.

Now the wrench in this is their special powers. Bag-Man’s basic tile damage is the same as the others, but people don’t use him because his abilities aren’t worth spit. Obviously Thor’s special powers will slam Bag-Man’s head into the ground, but these imbalanced skills are something the devs have to continuously work on in order to balance, for the life of this game. The reason I personally skip battles so much is because I know I can’t beat Spidey’s 2AP stun-lock, no matter if he’s lv. 50 or lv. 100. So fixing this skill would balance the playing field and therefore result in fewer skips.

Separating brackets by stars, a person with three lv. 141 *** characters could still be matched up with a player with three lv. 15 *** characters, and would obviously win. A team with two lv. 85 ** characters and a lv. 50 * character (team lv. total 220) could possibly take on a lv. 100 *** character, a lv. 65 ** character, and a 55 ** character (team lv. total 220), and have a fighting chance to win.

A system that’s based on ISO accumulation sounds too chaotic since ISOs can be gained anywhere in various amounts. Playing 5-10 minutes in a tourney nets you a quick 500 ISO at 50 progression at the moment, as well as an option to purchase ISOs, so it’s not really a good judge as to how much time a player has spent in the game. I could also invest a lot of ISOs into a character and then not use that character anymore because of nerfs (i.e. Ragnarok) but my ISO accumulation would put me into higher matches even though I might be using a weaker team.

Again, I don’t have all the right answers and I don’t want to pretend to have them. I honestly think you guys are bringing up excellent counterpoints, and I just hope that this discussion will lead to a helpful improvement to the game.

17

Yes, I agree. Repeatedly skipping opponents until you hit one with an optimal amount of reward points seems like a part of an inherent flaw in the current system which is not easy to solve. I know that I personally have been blindly skipping a lot of 30 point battles (which used to be the optimal amount) lately in search of a new and shiny 50 point golden goose egg. But this is just how humans work; we want to maximize our rewards for the least amount of time spent. We also want to maximize our points to compensate for the bulk of points possibly lost in the same amount of time. Whatever the case may be, adding the proposed skipping tax probably won’t stop many players from looking for their golden egg, but it will sure frustrate the players (easy business solution) instead of fixing the flaws in the system (difficult solution).

18

This is a stupid idea I have no character over 69(lol) and I still fight for top prizes fighting I could use characters under 150 lvls and I would be in the easy bracket and have an easy ride for first place.
I would prefer to fight people with 20-30 points this would be fair for everyone, obviously this would change when you reach higher points.

19

I agree, if everyone does it, it’s not an advantage. But if you don’t do it, it’s a disadvantage. The skipping problem needs a different solution, and I don’t think charging for it will work. The problem is that the point value you gain from defeating an opponent is not really related to how strong that opponent is. While there should be a correlation since better teams should win more, it just doesn’t seem to be the reality of the situation. Part of this is because of the ability of the players to beat the AI and the boosts they can use. The other part is that while incremental changes in damage ability does help, it takes a good amount of levels to provide a decent advantage and even more to be feared by other players. This leaves us with a player pool that attacks the point value and not the team.

Now I’m sure the data does show that better teams win, so the system is not completely broken. It functions fairly well. It just results in tournaments where a player has almost no control over how well he or she does at the end. Shields help, but feel more like a band-aid. So players do what they can to game the system. They tank and skip. There isn’t a problem with one thing. It’s just the way the systems work together that gets us to this situation.

Segregating the player pool is one way to ease the tension. This is what the MMR is for. However, the idea of better rewards at higher levels and/or a rating based on character level would encourage less tanking. I do think that if you want to fight up a class, you should be able to. But if you are always going to be fighting the AI, I think points should be more in line with the strength of the opponent, like they are in PvE. This would be another way to combat tanking. The AI could also use an advantage to compensate for its inability to play at the same level as the players. Maybe characters have a defensive mode that increases their stats when attacked. Also, retaliations should yield at least as many points as you lost, and it should end after one retaliation. As for skipping, every time you win a fight, you get a more difficult opponent. When you lose a fight, you stay at the same level. When you skip, you get easier opponents. That way you can control how many points you are going for and if you are in over your head. Retaliations would not effect this. They are just a way to get the back the points you lost.

I don’t know if the combination of those things would be any good. There would also be a lot of balancing that would go into it. It also might not make the end of tournaments less unpredictable. But it seems like people would be fighting the difficulty they want to fight for the points they want to earn with a way to get back points taken from them.

20

Even without factoring advantages of boosts and whatnot, at around the 1000 rating (probably earlier now), most teams just look more or less the same, and certainly no team is going to be twice as tough to beat as another, so you’d never fight a team that’s worth 15 over one that’s worth 30 points. Not to mention attacking a team with 15 points carries seriously retaliation risks while attacking the 30 point team is considerably safer on retaliation. Currently with no limit to skip, you’ll simply always find the team that’s worth the most, and if you’re the team that’s worth the most to someone else, they will also always find you.

I think we need a system like if you fight the first available opponent without skipping there’s a 50% chance you won’t score a retaliation.