Guild Wars 2

So I did alright in my placement matches and started near the upper tier of gold. I played relatively sparingly and made my way up to rank 181 (early January). Christmas time kept me from playing much.

I will confess that in most of my matches my specific play did not really make a significant difference. Having a bad player on my team at this level meant I lost. If I played really well or well, and there was a bad player, I lost. Sometimes if we were all good I actually felt I was the tipping point. but having a bad player far outweighed having someone I ’d noticed was a brilliant player.

On my way up I noticed that my matches were pretty much random win or loss. The random factors were things like the crud player, the oops we got 3 theives and when you play (affected who you were playing with). Both of these pretty much overrided the I played 10% better this game factor.

I did notice that in games were I was grabbing three or more top stats…. we typically lost. In the 250 and above area I basically won and lost about evenly and at least folks seemed to reasonably be counted on to do somewhat sane thing (though some would insist on far despite the rest of the groups class make up).

At rank 181 I had a choice. I knew things were very random and IF I wanted to stay there I could not trust my skill would make a significant difference. BUT I wanted the reward for finishing the reward track. SO I decided not to end this season in top 250 by continuing to play.

Sure enough I hit a bad game and it dropped me 34 points. I then proceeded to lose like 15 games pretty much strait and decended to lower half of gold. Litterally I was playing no differently, and by the time I was in lower gold I was consistently grabbing top rewards, winning 1 v 2’s and basically making clown shoes of most of my opponents, and it just didn’t matter.

Now I have since been on a relative winning streak except I’m not seeing any +34’s to my rating and the climb is quite slow, further troubled by the fact that I am litterally at the mercy of randomized match making to decide in all but 15% of the cases whether I win or lose (I estimate). And it puts some real oddballs out there.

It was actually harder and more frustrating for me in lower gold then in 250+ rank because of the randomness of my team. I anticipate that due to the randomness of match making I won’t make it back to 250.

IF you want to stay in the upper 250 take a bunch of alt accoutns, play your 10 matches, focus on the accounts that win and play the absolute minimum you can to stay in the upper ranks. DO NOT take your chances on the randomness for best results.

I believe generally because people at higher rank play less, you are on average (if High) likely to be playing against lower MMR folks and are risking more than you gain. Litterally the people you need to be beating aren’t playing.

The worst part of all this is I wanted to play ranked and when I was worried about my rating I actually played LESS PVP and was discouraged from wanting to do it because of the bad match making. Which is really twisted when you think about it. Maybe I need 2 accounts, one I play for rank on and one so I can actually… you know play the sheer randomness that assigning individuals a rating based on how their team does is.

Anyway my thoughts so far.

As I have been slowly crawling my way up in the gold division by maintaining a higher than 55% win rate, my experience has been very random, inconsistent, and very frustrating. I have gotten lots of points in matches that I had my dead corpse carried by team players, and have lost lots of points in matches that we’ve lost despite my exceptional performance. A player’s climb or descend in the ladder doesn’t seem to be based on your performance. One’s placement in the league doesn’t seem to make much sense either as there doesn’t seem to be a stark skill level difference between bronze and gold players. You can see equally bad and good players in every division.

This is so true and the exact problem with PvP right now.

I was maintaining my place in Gold T3 with a 60% win rate for days, never really moving. I was getting bored of DH meta so changed it up just to have a bit more fun. I played burn guard for the first time and wow it just clicked with me, I played so well with the spec. I was suddenly 1 vs 2 and 1 vs 3ing people and felt like I was making a huge difference. But as soon as I changed spec I was somehow losing games. I got onto a huge losing streak despite feeling like I was making a big difference. The craziest moment is when i killed 1 vs 3 at far in a match and was so proud then turned to mid to see my team had wiped 4 vs 2.

It doesn’t matter how well you play once you get past mid gold. Winning a match comes entirely down to who you get matched with.

It sounds like you guys are playing NA. On EU everybody is god-tier pro esports. Just ask any EU player.

Win or lose I always know I made a difference in somebody’s day

I’ve made a case of not looking at my rating or caring whether I win or lose. I just want to play better and be in a fun match.

So far, I’ve won more than I’ve lost and I’m not quite sure where I’m at on the Leaderboard atm….

Wait til you hear my story. Hit over 500 games this season now, and here’s what happened.

Placed in Gold 2, got to Plat 1, dropped to Silver 2, went to Plat 2, dropped to Gold 1, and now in Plat 1.

Lol

I’ve had a similar experience, although not as extreme as falling from top-250 into gold, or from plat to silver and back again…

I think one thing everyone needs to recognise, including the devs, is that even with a perfect matchmaker, when you’re reasonably close to where you should be the majority of games are going to come down to luck. You’re only one person out of ten, after all, and if you’re reasonably close to your proper division (as opposed to, say, a legendary player in bronze or silver) you can only carry so far.

In my experience, there’s maybe 20 or 30% of games where I felt like I actually made a difference to the outcome, for better or worse. The rest are either wins where my team as a whole outclassed theirs so much that all I needed to do was make a basic contribution and the team would likely win, or losses where my team was so outclassed that I couldn’t realistically see how I could have made a difference to the end outcome.

Adding to this, there are some maps which enhance the luck factor. Stealing a beast kill in Forest is worth 50 points to the final result (25 for the points you get, and the 25 you deny to the enemy) – however, while there are things you can do to maximise the chance that you last-hit the beast, in the end it’s often a matter of luck who actually gets the last hit in. Khylo games often become a question of who has someone who’s good at using the trebuchet (and knowing when not to) – a skillset that’s only useful on that particular map as far as sPvP is concerned. (For this reason, I actually preferred old Skyhammer to Khylo: anybody could use the cannon, and being centrally located it was often more viable to disrupt somebody using a cannon if they’re having a decisive effect).

By the law of averages, if you play enough games the luck-based games should cancel out and the fraction where you do make the difference should settle you to where you should be. But it certainly is frustrating to get a loss streak from a series of games where you had no realistic opportunity to change the outcome.

(It’s one of the reasons why I was sad to see full teams go. I’ve never had the opportunity to play league as a full team and possibly never will as a regular thing, but it’s one way to reduce the role luck plays.)

This very thing has been happening to me these past days… I was on rank 200ish on EU and then, suddenly, incredible lost matches streak. Out of sudden, my teams were filled with people flaming and raging BEFORE matches started. So now here I am, gold 3 and thinking seriously about stoping playing PvP for the first time in 4 years.

draxynnic.3719:

By the law of averages, if you play enough games the luck-based games should cancel out and the fraction where you do make the difference should settle you to where you should be. But it certainly is frustrating to get a loss streak from a series of games where you had no realistic opportunity to change the outcome.

I’m not sure that having an attempted 50% chance will guarantee over a non infinite number of games that they’ll cancel out. Go flip a coin twice and tell me if you got one head and one tail for instance.

But ignoring that it would only cancel out if your Rating changed the same for the wins and losses when adding the net effect of the changes.

Consider this… what if your a very high Rating, you are more likely to be in games where the average rating of the team is lower than you, correct? This means that a loss will hurt you more than a win will gain you.

I know for a fact that when I was 1630 ish rank, and I experiment duo’d with a 15XX player, that my losses were two points more than his, and my gains were less for matches we were both in.

So as you do better you risk more and gain less BECAUSE your MMR is higher than that of the team you are fighting it seems.

A rather significant flaw, my penalty or gain should be based on how my team was ranked against their team, why does my loss or gain amount differ from my teammates? And if it does, then why in gods green earth would I want to play games if I beat the odds and managed to climb.

Personal experience, as a complete newbie to PvP this season:

Absolutely bombed my placements 2-8 and ended up in Bronze 1 at 466.
Over the next day or two, climbed to around 670.
Hovered around 590-670 for a few more days.
Currently on an upward climb sitting at 893.

In between, I duo’d with someone doing placement matches, and afaik, they’re high Gold or low Platinum rated on their main account. I was definitely getting some really high rated matches where losing was a -3 or -4 point loss, and both matches were very close losses (500-480, 500-450). There was also a win in between that earned me 26 points or so. My conclusion from that experience: my map awareness and rotations need a lot of work (D/P Thief), and while I can keep up with some guidance (my friend was giving me suggestions while playing and it helped immensely as they play Thief too) at Gold tier, I’d be a liability queuing by myself (at my current level of skill in Gold tier). With enough practice, I’m confident that I can improve to the point of belonging in Gold tier.

My goal for this season is to make it to Silver (at least 1050). If I really push myself to improve, I think I can get to Gold 1.

My goal for next season is to be fully prepared on 3 professions (D/P Thief, Warrior or DH, and support Ele) and place significantly higher than where I started Season 5. I want to be able to play at least one of each key role (roaming/+1, bunker, support), so that I can swap to (which happens very rarely at Bronze).

I’m maintaining an overall 53% win ratio, 60% not including placements.

Abelisk.4527:

Wait til you hear my story. Hit over 500 games this season now, and here’s what happened.

Placed in Gold 2, got to Plat 1, dropped to Silver 2, went to Plat 2, dropped to Gold 1, and now in Plat 1.

Lol

OMG this basically says no relevant carry. Randomness from teamcomposition …

..

Well fits. Had two heavy loss streaks most games totally uncarryable…

shion.2084:

draxynnic.3719:

By the law of averages, if you play enough games the luck-based games should cancel out and the fraction where you do make the difference should settle you to where you should be. But it certainly is frustrating to get a loss streak from a series of games where you had no realistic opportunity to change the outcome.

I’m not sure that having an attempted 50% chance will guarantee over a non infinite number of games that they’ll cancel out. Go flip a coin twice and tell me if you got one head and one tail for instance.

That’s pretty much my whole point. The more flips you get, the more likely you are to be somewhere near the ‘expected’ value.

Coin-flipping experiments are actually a good analogy because people have actually done experiments that showed that people are disinclined to believe that genuinely random strings of coin flips are actually genuinely random… because those flips have strings of the same result that people intuitively think are too long to be the result of random chance. However, the more games you play, the less likely that those strings of dumb luck are going to have made a big difference to your placing.

Now, let’s say, for argument’s sake, that 20% of matches you can make a difference on, 40% you’re going to win as long as you’re not trolling or AFKing, and 40% you’ll lose regardless of how well you do. Statistically speaking, across a large number of games, the 40%es should cancel out, leaving you with your own skill making the difference. However, there’s nothing stopping someone from getting a string of a dozen of those 40%es in a row. It’s unlikely, but hey, it’s only one in sixty thousand… with the number of people playing and the number of games they’re playing, these streaks are going to happen every so often.

The variable rating change is, I think, intended to smooth this out: the matchmaker can’t make teams it thinks are even every time, so it compensates for awarding less for games it thinks are easy wins than more difficult ones, and docking you less for games it was expecting you to lose. So, theoretically at least, the shift from getting a string of unwinnable or virtually unloseable games is less than it might otherwise be. The system is far from perfect of course – there are factors that the MM doesn’t take into account, and there are some that it simply can’t – but the intended effect is to reduce the influence of luck.

A rather significant flaw, my penalty or gain should be based on how my team was ranked against their team, why does my loss or gain amount differ from my teammates? And if it does, then why in gods green earth would I want to play games if I beat the odds and managed to climb.

This being one of the imperfections in particular.

Burjis.3087:

I have gotten lots of points in matches that I had my dead corpse carried by team players, and have lost lots of points in matches that we’ve lost despite my exceptional performance. A player’s climb or descend in the ladder doesn’t seem to be based on your performance.

Exactly. The current system is seriously kitten braindead.

I can’t believe no one at Anet realized the simple way to make this work and reward players for playing well: make loss/gain of points dependent on your own individual performance, NOT that of the team.

Otherwise, we get the kinds of experiences reported here and elsewhere. I just can’t believe Anet thought it would be a good idea to have your rating depend on a random assortment of teammates AND to not allow you to form full teams to counteract this.

It should be either full team formation capabilities and current ranking system. Or no teams and ranking dependent on individual performance only.

At this stage I don’t care about my rank because of this. I just want enough ascended glory shards to get those ascended gear. Any idea of a way to compare myself in PvP to others is irrelevant now to me.

alccode.1297:

Burjis.3087:

I have gotten lots of points in matches that I had my dead corpse carried by team players, and have lost lots of points in matches that we’ve lost despite my exceptional performance. A player’s climb or descend in the ladder doesn’t seem to be based on your performance.

Exactly. The current system is seriously kitten braindead.

I can’t believe no one at Anet realized the simple way to make this work and reward players for playing well: make loss/gain of points dependent on your own individual performance, NOT that of the team.

Otherwise, we get the kinds of experiences reported here and elsewhere. I just can’t believe Anet thought it would be a good idea to have your rating depend on a random assortment of teammates AND to not allow you to form full teams to counteract this.

It should be either full team formation capabilities and current ranking system. Or no teams and ranking dependent on individual performance only.

At this stage I don’t care about my rank because of this. I just want enough ascended glory shards to get those ascended gear. Any idea of a way to compare myself in PvP to others is irrelevant now to me.

The current system to rate personal performance is not enough to accurately rank people. Most flagrant is the +1 role. You can make decisive actions but still have no top stats. top kill being the easiest to achieve. Beeing there at the right time just gives you average score.

The problem is that there’s really no algorithm to judge how much of an influence somebody’s had on the game.

For instance, one problem with the old system of individual points was that players didn’t get points for bunkering a point. Should they? How effective bunkering a point is is highly dependent on circumstances in the game. I’ve seen games that have been won because of someone sitting on a point and holding it against a stream of attempted decaps, and I’ve also seen games that have been lost because someone was sitting on a point that was rarely if ever contested by the other team while their team was 5v4’d on the rest of the map. You can’t write an algorithm that can determine whether sitting on point is a good idea or a bad one.

Any algorithm that attempts to determine a player’s individual contribution is going to fall foul of this problem. Once players figure out what metrics the system is using to gauge individual contributions, people are going to start competing with their own teams to game the system rather than focusing on winning the game. (You see this with missions in World of Tanks that reward a player for personal performance: every so often you see player behaviour that you can tell is aimed at completing a mission more than it is to help the team win.)

The bigger problem is that apparently rating gain/loss is determined by how your individual MMR rates against the average of the enemy team MMR, not on how your team’s average compares to the enemy team’s average. This means that if you’re the highest-ranked player who needs to carry a team of lame ducks, you have the potential to lose a lot and little to gain if you do pull off a win, even if the odds are stacked against you. Meanwhile, if you’re the lowest-ranked player in a match and you get carried by a good team, you have the potential to gain a lot of points and not many to lose, even if the odds are stacked firmly in your team’s favour and you’re not contributing any more than a warm body.

The system really should make it so that the potential loss or gain is determined by the team MMR and not the individual MMR, meaning everyone in the team gains or loses the same amount.

You know, if the top of NA wasn’t full of smurf accounts those people you’re beating in gold 1 easily would be gold 3 most likely. This means you would have more consistent games in your rating bracket as the entire system is distributing the rating between players properly.

What happens when you have a bunch of alt accounts taking up 4 or 5 spots on the higher end is they’re effectively taking rating off other players to stay there. It comes down to the way the system takes more rating away for a loss than you get for a win unless you play really well. The more smurfs the less places and people there are playing at the same time at the higher end for you to gain rating off and the high the chance of going down.

A good example would be the top ESL players, each one is sitting on a high win ratio so chances are if you ever fought them you would lose and thus lose rating. We’re going to use just 5 and assume they have 5 smurfs each, this means when a smurf is getting to the top you will fight it and you will lose. This pushes you down and them up, they do this 5 times which means there are 25 accounts you will always lose to and because they can’t play at the same time they act as a rating storage.

Take away the smurfs and suddenly there’s 20 more people playing at the same times for those 5 ESL players to take rating off and for you to do the same as people will move to fill in the ratings gaps.

I just used ESL players as an example of people who are most likely the top or near the top of player skill and people you’re not likely to beat often. In no way is it an accusation.