P for Psychology: Heroes of the Storm, Matchmaking, and a Very Expensive Herring

Heroes of the Storm has been out for almost two months now, to moderate and respectable success. You should go play it if you enjoy games like League of Legends or DOTA2, but don’t enjoy slugging through libraries of guides on obtuse mechanics on last-hitting or jungling or whatever. However, this is not a plug. I condescend to you to discuss a problematic trend in the Heroes community right now. Despite the youth of the game, HotS continues to get a lot of flack for its “matchmaking problems.” This is the purported issue of how Heroes’ matchmaker prioritizes speed over relative skill as measured by MMR, leading to countless posts on how the matchmaker put someone with, say, 3000 MMR on the same team as someone with 2000 MMR. Moreover, since the rise of Hotslogs.com, MMR checking and other pernicious habits have started to infest the community’s mentality, despite Hotslogs.com being notoriously inaccurate. I do not dispute the existence of a matchmaking problem per se, since Blizzard has already admitted to certain issues with it, but I highly doubt just how widespread people think it is or if they even understand the supposed problem to begin with. From a higher perspective, my concern lies in people latching onto a convenient scapegoat instead of learning how to deal with the typical and inevitable variance they will encounter in an online multiplayer team game. As I said, Heroes is a young game; bad habits form easily in youth; bad habits die hard.


Including making bad memes.

Matchmaking in an online game has, is, and always will be an art more than a science, particularly in a team-based game. Often people offer blind praise for the matchmaking caliber of a game like Starcraft 2…a single-player RTS game. When you get right down to it, it’s very difficult to match five players against five players while trying to narrow any potential skill-gap as much as possible, let alone accounting for stuff like allowing for friends to queue with each other. Games are not at the point where their systems can make comprehensive value judgments on a human’s behavior, so any sort of matchmaking rating is a post-facto attempt to gauge pure game performance. As everyone knows, damage meters don’t tell you how good of a player someone is. Win ratios don’t reveal if a person yells at his teammates every second game or grasps strategic priorities on every map. At the end of the day, MMR only conveys how often a person has won or lost relative to the player base over a long period of time. Everything in between is variance.

But what is variance? Well, it covers a lot of ground. For instance, no one’s knowledge of the game will ever be complete. Heroes is not a solved game, like checkers. It is constantly shifting and being adjusted by its developers, who are in turn making decisions based on the collective decisions of a vast playerbase. Nobody knows everything about everything in Heroes at every point in time. Most people don’t learn even half the heroes of the game very well, even people in Master League. This is why you end up with players that only play assassins well or who can never play support. You will run into those kinds of people at every layer of the playerbase. Knowledge variance never disappears; it just stabilizes over time as you tend to encounter more people with wider game knowledge and wider competencies.

The same applies to performance variance. Robots are not playing this game: humans are. Humans are meat-bag primates whose nervous systems and physiques did not evolve to play highly complex computer games online. We have physical and mental limitations that affect our ability to perform at tasks. Sleep problems, a bad breakfast (or no breakfast), a miserable day at work, frustrating social events, etc., can all very negatively affect one’s ability to make judgments in a game like Heroes. Glaurung has bad days. Zp has bad days. Nick has bad days (and breaks the screen to cope). Everyone has bad days. Everyone makes a bad call on occasion, whiffs that key skillshot, gets tunnel vision, mixes up spawn timers, or even prioritizes playing a champion to have fun over winning. Just like knowledge variance, performance variance never goes away.


What? I had to get a Snape reference in here.

So what does this all have to do with matchmaking? Well, everything. Matchmaking is no god. It does not determine everything that happens to you. In fact, unless you’re at a very competitive level of the Heroes community, it probably doesn’t affect your games much at all. The other stuff I mentioned is so much more important. Skill gaps, even tiny ones, can have far more drastic influence on how often you win or lose. Your team knowing when to take mercs or to go for an objective, or knowing how certain champions scale and which ones counter which, can make or break a match before it even begins. The mere skill of knowing when to back off is something most people at the lower strata of play don’t understand, even though it’s vitally important. To chalk everything up to “I was matched with a n00b” is both lazy and self-defeating. There are very few matches where you played perfectly and everything was everyone else’s fault. Yes, they do happen, but only once in a blue moon. You have no control over whether someone’s Time Warner connection is shitting itself at that particular moment. You can only control what you’re doing, so you need to focus on helping both yourself and others instead of tossing all blame onto a convenient excuse you don’t really understand.

The Heroes community needs to stop mistaking what has come to be called “matchmaking” for normal shit that every player goes through in every online game. You will be matched with feeders, AFK’ers, the first-pick Sonyas, the last pick Novas, and you will be matched against the first-pick Zeratuls that can Blink-dance with one hand tied behind their backs. It’s called “life”. Deal with it. Matchmaking shouldn’t even be on your mind until you’ve been Rank 1 for months and your win ratio is very stable. Only then can it be an intrusive element that you can legitimately complain about. In the meantime, if you really want to improve, watch replays, watch high-level players in tournaments, analyze what they do, analyze what you’ve done wrong, and treat your teammates with respect and decency. What’s more, part of this “matchmaking” issue arises from the way the competitive ranking system is structured at the moment. It’s much harder to rank yourself against other players and judge where you really are when there are only 50 ranks, only half of which matter, and there’s no Grandmaster League yet to stratify and discriminate between the people who’ve gotten to Rank 1. That will be fixed in time, though. It’s not something to get worked up about.

Heroes has the potential to be a very popular and excellent MOBA, but if we keep instilling this mentality that “matchmaking sucks” and teaching new players to blame their poor performance on a convenient scapegoat, we poison their experience from the very beginning. Even at the very worst, Heroes doesn’t fuck you over nearly as much as a game like LoL or DOTA, where you’re stuck with that feeder for 30+ minutes and just one ill-matched person on either side can sink the game for you and waste such a significant amount of time. If you get stomped in Heroes, the match is over in 10-15 minutes and you move on. We should be selling that as a big feature instead of running around with our heads cut off about that stupid Sonya pick.

Perspective is everything in life, and we shouldn’t lose it here just because it’s a game. Games really are serious business. We enjoy them and invest ourselves in them, which is the only thing that truly matters for us in an otherwise brief and largely futile existence. If you let something as specious as “matchmaking” control your thinking, you’re not going to have fun in Heroes, and that’s a crying shame, because that’s what games are all about.



And winning. Winning is more important.

Heroes of the Storm and the Deathly Release Date

SILLY.SNAPE-PREPAREDBlizzard announced the Heroes of the Storm release date today. June 2. Huh. Well, I’ve been playing HotS up to my eyeballs–enough to keep me from posting on this blog–and have a pretty high MMR, so I’m confident enough to speak on the game’s current status, particularly as to whether or not the game is actually ready for this sort of transition. Snape, loving curmudgeon that he is, has already given away my opinion.

As an aside, yes, I am alive. I do plan to post more, but there’s not too much to talk about as far as anime or whatever goes this season. I will do another Code Geass post soon, as that was a shitton of fun to write and read. Lelouch Lamperouge must be taken to task for all his “grand” schemes that make no bleeping sense whatsoever. I will need many bullets.

Get it? “Bullets”? They’re…eh. Philistines.

Oh, I was talking about Heroes of the Storm. Yeah, Blizzard, I don’t think you know what you’re doing. The game isn’t ready. It just isn’t. I play at high Diamond level and am pretty close to Master. From my perspective, the game has a lot of grime that needs to be rubbed off before you toss it out for public consumption. Most notably, we need more Heroes. Lots more Heroes. I was talking to a friend about it just now and we both agree the Hero pool isn’t sufficiently deep enough to create a healthy meta that supports flexible team composition selection. Sylvanas was a good step in the right direction, but since you’ve so thoughtfully set us down a collision course with destiny, we need about six-to-seven more Sylvanases to be in a decent state come June. Given that you’ve been releasing champions around every six weeks, that’s not enough time for more than two champions…unless you’ve got some surprise package of Heroes in the pipeline just waiting to go. To be honest, I was hoping today’s announcement was going to be a Diablo trio of Heroes or something thereof. Talk about disappointment in that category, but it just brings us back to the relative dearth of meta-viable Heroes to take into the Nexus. In an interview a few months ago, Blizzard stated clearly there were six Heroes in development at that time. Since then, both Lost Vikings and Sylvanas have been released. So that leaves us four. I’m going to assume that since then at least one other Hero has moved into the acute development stage or whatever you want to call it, so say five. Five isn’t enough, especially when there’s no guarantee all of them will have the same release-date quality as Sylvanas.

But that’s not all, Blizzard. There are other problems afoot at Hogwarts, since this school is run by morons who think putting soul-sucking, life-scarring Dementors a hundred yards away from school children is a perfectly sane idea.

Talents. Talents, talents, talents. Talents are a relatively new development in HotS, and it shows. Jaina and Sylvanas both demonstrate that Blizzard is finally getting the hang of what talents ought to look like, but so many other Heroes need an update in that department. There are only a few meta-viable talent builds for every Hero; many of those are sub-optimal. Some Heroes are stuck in the short end of the pool as far as that goes. Tassadar, Tyrael, Uther, etc., all have talents that suck really bad and no one will ever pick in any universe, prime or alternate. Furthermore, that’s not the kind of change you can realistically push out in the space of seven-ish weeks. Sure, surprise me by all means, but I remain skeptical, as I ought to.

Then there’s matchmaking, the canker afflicting HotS. People are not really matched well based on their actual MMR or some other solid performance indicator, but speed and general accuracy. This is less of an issue at the lower ends of the game where no one knows fuck-all and runs around thinking Anub’arak is OP or something, but it quickly becomes a serious issue the higher your skill level goes. At the very top end, it’s pretty annoying at best to be matched regularly with people at half your projected MMR, not that the other team doesn’t suffer from this just as much. It still takes the wind out of the sails of high-end competition. Few things in HotS are more enjoyable than a high-level match between equally skilled teams, win or lose. Trust me, I’ve done it. The problem is that it’s so rare that you have to slug through ten matches of Gazlowe-obsessed weirdos before you can face gather a proper team comp.

Suffice it to say, I’m not sure if Blizzard has listened to Illidan on this matter. We’re not ready for public release. The game systems need to be refined more, the talents need serious attention, we need more Heroes to shake things up and provide more viable picks and counterpicks, along with a simple team/guild/clan system and other similar features. HotS has the potential to be a great MOBA, but that only comes with time and patience. Am I wrong to suspect this release date is more internally dictated rather than developer-oriented? Maybe, but I wouldn’t bet one of my horcruxes on it.

I have no horcruxes, by the way. That was a joke.

Heroes of the Storm – Tassadar Feedback

I got into the cPRETTY.TASSADARlosed alpha for Heroes of the Storm about two weeks ago. This is the one Blizzard game I’ve been interested in for a while now. I bounce into WoW every now and again to hang out with guild mates, but the game lacks pull to keep me logged in every day. For the past two weeks I’ve been mostly maining Tassadar. I’m close to unlocking Level 10 on him. My plan as the game’s development goes forward is to main Tassadar and Nova and expand out to another support and assassin. In general, the game feels good. It’s quite distinct from LoL’s style. Most pleasant is how anti-snowbally HotS is compared to LoL. Very, very rarely do games feel hopeless. Even if you’re several levels behind, one good team fight can completely turn the game around. It makes playing the game decidedly less miserable than LoL, although not as engaging. I’d say that’s a fair exchange.

Anyway, onto the actual feedback:

Tassadar is designed to be a support/assassin. I put “support” first simply because all his core abilities fulfill that role. You’re not going to do the most damage on your team, but your clutch Q’s can save the day, a well-placed W’s can decimate an enemy team, and a Force Wall or Archon can turn a teamfight in your favor. Regardless, Tassadar always servers to augment his team’s abilities. To be honest, I’m not really sold on calling Tassadar an “assassin” either, unless Blizzard is just using “assassin” as a blank term for “damage dealer,” a nomenclature I deem problematic. Needless to say, there’s a very good reason Tassadar is marked as “Very Hard” on the champion select screen. His base kit is doesn’t seem terribly focused or intuitive, and I think this will make new players shy away from him. It takes time to really grasp what Tassadar can do. What’s worse, a lot of it doesn’t feel very interactive. You throw out shields, you throw out Storms and hope and your enemy stands in them, you cast Oracle at opportune times, and Archon is really just a temporary DPS buff. The most interactive and fun ability he has is Force Wall, which unfortunately feels weak compared to his assassin build.

HotS is built off its talent system, which I won’t take too much time to explain. Any one with a passing familiarity in RPGs or gaming should understand the concept of a talent system. A full list of Tassadar’s talents can be found here. Overall, Tassadar’s talent choices after his latest rework don’t feel particularly compelling. As you’ll see below, most of his tiers have one or two good talents that crowd out the rest. Talent trees have always been haunted by the specter of opportunity cast. It’s just as important to consider what you’re going to give up as much as what you’re going to get from a talent. As such, on Tassadar you really know what you’re going to pick beforehand, so long as you’ve decided on which path to proceed down.

Tier 1:

  • Conjurer’s Pursuit – due a lack of options, this is mandatory for a support-build Tassadar. You’ll be spamming out spells as much as you can. Having a steady amount of regen is vital on a champion as mana-hungry as Tassadar is. You have to be pretty proactive about building up your regen, though, which I like. It doesn’t come necessarily for free.
  • Overload – useless talent due to Psi-Infusion.  Scrap it and provide something more for assassin-Tassdar, or change other talents to make it more attractive. A simple increase to Psionic Storm’s range serves no real purpose, as nine times out of ten your enemies will walk out of it, and that tenth time you need the range to snipe a target means you’ve done something wrong. Given that this talent is taken significantly more often than Psi-Infusion, yet gives the same win-rate, I rest my case.
  • Psi-Infusion – very good for assassin Tassadar, largely for the same reasons as Conjurer’s Pursuit. You need to hit 9 targets to really break even with it, but it saves you a ton of mana nonetheless and allows you to lane and team fight without worrying about going OOM quickly.
  • Minion Bulwark – absolutely useless. Scrap it and provide something more for support Tassadar. There are a handful of times where I’ve ever even thought about using my shield on a minion, and it’s almost always been a waste of mana.

Tier 2:

  • Mental Acuity – my default talent on any build. Frankly, this tier doesn’t have very many options. Oracle is a key component of Tassadar’s presence, and being able to cast it twice as often offers near unrivaled utility.
  • Leeching Plasma – this talent might be good if you’re paired with a tank or such, but it’s usefulness seems pretty situational. I don’t pick it too often.
  • Promote – Terrible. This game is not minion-focused whatsoever. Scrap it.
  • Healing Ward – can heal allies for 33% of their health if they stand in it for 10 seconds. I guess that’s not awful, but really, why would I take this?
  • Reinforce Structure – borderline useless. Borderline OP on Haunted Mines. This talent alone has won games for me on that map, particularly if the enemy team gets a stronger golem. However, this is entirely due to the golem’s behavior. The counterargument is that other talents in this tier can help you prevent that situation from even happening, so there’s not a very strong argument for it. Needs a buff or rework.

Tier 3:

  • Khala’s Embrace – Default talent on support Tassadin. An extra few hundred health everywhere you or your allies go is always useful. Synergizes well with other talents too, and makes sure shield absorbs do not get wasted ever. Very good talent.
  • Static Charge – default talent on assassin Tassadar. Greatly increases his DPS output and provides some much needed interactivity to his playstyle. Good talent.
  • Deep Shift – useless. Scrap it. You use your E to dodge abilities or break out of roots and such. Its duration is secondary. At no point can I think an extra 0.75 seconds would be worth it compared to the other talents in this tree.
  • Calldown: MULE. People were calling this talent OP somewhere, but I really, really don’t see it. You just snipe the thing down, and 100 HPS cannot counteract a full team push. Maybe I’m missing something.

Tier 4:

  • Archon – core talent for assassin Tassadar, but, frankly, a really boring one. It’s a DPS buff, as I said above. You get some juicy damage output for a few seconds, and data says it’s far more effective than Force Wall in winning games, yet that’s also a factor of sheer popularity.
  • Force Wall – core talent for support Tassadar. Can make or break teamfights. Has saved my ass countless times. Best ability Tassadar has in his arsenal as far as fun and feel goes. Effectiveness is suspect, though. I think it needs a buff.

Tier 5:

  • Distortion Beam – my default talent on either Tassadar build. A 25% slow is such good utility it’s almost impossible to resist.
  • Presence – mediocre talent. See Deep Shift. It’s some convenient grace, but that’s about it. The flipside is that Dimensional Shift effectively silences you, so it can trigger at inopportune times.
  • Scryer – mediocre talent. I don’t see what it really adds. Evasive Shielding fulfills this roll in the talent tree.
  • Spell Shield – no one takes this. There’s a good reason why.
  • Shrink Ray – the anti-assassin talent. It’s not bad. It’s not great, but it’s not bad.

Tier 6:

  • Evasive Shielding. Awesome talent for support Tassadin. Don’t change this.
  • Resonation – meh. A slight slow that doesn’t really do much. Scrap it.
  • Second Strike – a free Psychic Storm every cast greatly multiplies the effectiveness of the ability.
  • Dimensional Warp – why? Again, Dimensional Shift is something you want to be using as little as possible.

Tier 7:

  • Resurgence of the Storm – bullshit talent. I predict Blizzard will delete this talent soon. No MOBA has ever made a generally available instant-revive component work. It’s either useless or broken beyond belief. Right now it’s the latter.
  • Storm Shield – no one takes this.
  • Twilight Archon – everyone takes this.
  • Force Barrier – makes Force Barrier easy to use. I think the range should be baked into Force Wall and the talent given another component.

Current State and recommendations:

Tassadar’s support side needs some buffs. Right now everyone seems to go assassin Tassadar, and I think the reason is the same one behind his recent nerfs: he doesn’t give up much to do damage, so why would you go support unless you’re overloaded on damage already? Tassadar support needs to bring a little more active and reliable utility to the battlefield. Until Level 20, Force Wall is hard to use due to its short range, and Oracle is great but very, very passive and boring. On the other side of the spectrum, assassin Tassadar needs more flavor and some toning down to give support Tassadar room to shine, namely some tradeoffs for his damage.

Suggested talent changes. These are brainstorming ideas, nothing more. They are not meant to be taken too seriously:

Tier 1:

  • Minion Bulwark replaced with Khala’s Wisdom – If your shield is consumed with X seconds or expires without absorbing any damage, Y(%) of the mana cost is refunded.
    • Do you want to go for long-term mana regeneration for more general utility, or shield heavy for support Tassadar? It at least offers a potential decision, as opposed to now.

Tier 2:

  • Leeching Plasma (active talent) – Tassadar converts X of his remaining shield into Y stored damage that will be dealt by his next basic attack. This damage can stack up to Z, but will be lost if not used within A seconds.
    • Such a component would add some interactivity to his damage build and force you to make decisions on whether to risk your shield for burst. It would also tone down his innate utility by motivating him to shield himself over his allies.
  • Reinforce Structure – Reinforce Structure now reduces damage taken by the building by X% for Y seconds.
    • Gives the talent some added value. It might not save the talent, but it’s at least something.

Tier 3:

  • Deep Shift replaced with Dimensional Distortion – Deep Shift costs no mana and stuns enemies around you for Y seconds upon exiting it.
    • Allows Deep Shift to be used offensively, and makes the other Deep Shift-themed talents more appealing. Could also consider a cooldown reduction element instead of a mana cost one.

Tier 4:

  • Force Wall – make Force Barrier’s range increase baseline; nerf Force Wall’s baseline duration to compensate.

Tier 5:

  • Scryer – No longer gives increased duration. Casting Scryer grants X% movement speed to yourself and all allied champions in its initial cast range for Y seconds. Enemy champions revealed by Oracle are visible for Z seconds even if they leave Oracle’s range.
    • By making Scryer an active speed boost to your teammates, it makes the spell a little more interactive.

Tier 6:

  • Resonation – reduces Psionic Storm’s duration/damage by X%, but enemies struck by Psionic Storm are rooted for Y seconds.
    • Y would have to be brief, but making Psionic Storm a brief root or stun would add some reliability to spell and make Overload a much more attractive talent. As a tradeoff, its duration or damage would be reduced.

Tier 7:

  • Force Barrier – Force Wall’s cooldown is reduced by X seconds and its duration is increased by Y seconds.

Despite all of this feedback, Tassadar is a pretty fun champion, one I look forward to making one of my support mains as the game progresses towards release.