L’ultima patch di Heroes of the Storm ha portato non solo l’atteso rework di Tassadar, ma anche l’Anomalia che rende tutte le strutture più letali per gli Eroi. Attualmente diversi giocatori non gradiscono il potenziale allungamento delle partite, ma una discussione su reddit ha generato repliche interessanti.
L’utente _the_cheese ha espresso quanto sia frustrante perdere una partita dove si stava dominando per via di un singolo teamfight andato male post-20 (inversamente, una partita vinta allo stesso modo non è pienamente soddisfacente) e che l’Anomalia non stia facendo abbastanza per eliminare il problema. La sua proposta è quella di applicare a tutte le mappe il sistema di Passo d’Alterac: distruggere una fortezza rimuove Resistenza dal Nucleo, così facendo il team che è riuscito a conservarne di più sarebbe meno esposto a un push su una sola linea.
Lo sviluppatore Adam Jackson ha risposto con un lungo intervento (che riportiamo integralmente in fondo) dove chiarisce il punto di vista del team di sviluppo, affermando anche che prossimamente sono in arrivo delle modifiche per rendere nuovamente praticabili le strategie aggressive nell’early game.
Questi sono i punti principali del discorso di Jackson:
- Il team migliore deve essere in grado di vincere, ma al tempo stesso ogni partita non deve essere decisa fino a quando un Nucleo non viene distrutto.
- Le strategie aggressive dovrebbero essere incoraggiate, quelle “a tartaruga” scoraggiate.
- Secondo le statistiche interne di Blizzard, attualmente le partite ribaltate post-20 sono rare e il team non ritiene necessario un intervento mirato in proposito.
- L’idea della Resistenza del Nucleo è considerata valida, ma al tempo stesso è controintuitiva rispetto allo stile di gioco generale di Heroes. Richiederebbe sicuramente del lavoro aggiuntivo per essere adattata.
Oh boy, there’s a lot to unpack here. I’ll try to explain my thoughts on this, though it’s important to say that this is and has been a topic of debate within the team basically since our game started development.
So the topic of how much early game vs. late game matters is incredibly hard to parse because there are a lot of interconnected pieces that are balanced on their own spectrums. When we mess with any of these pieces (hero power scaling, death timers, structure power, minion/merc power, map objectives, etc), it will cascade down to other parts of the game, which is why we are careful before changing any of these components and why the chance of getting things right on the first try is basically 0%.
So before getting into the nitty gritty, I think it’s important to start with what our goals of a healthy game state would look like, because without a north star it’s extremely easy to get lost in the weeds. Keep in mind that these are ideal goals that are incredibly difficult to achieve, and that it’s not going to happen in every game that’s played because people are very complex beings with influences on their gameplay/psychology that we will never be able to account for (for example, having a bad day and not playing at your current MMR level, trying a new hero for the first time, orbeing on a loss streak and tilted before a game starts). Even so, in my vision, a healthy state for Heroes overall is:
- At any point in the game, the team that has been playing better has some kind of advantage
- There is not a point in the game where the losing team has no chance to come back (otherwise why keep playing it out?)
- Aggression is rewarded, and players are motivated to go out and make things happen (no infinite turtling)
So before getting into feelings, as far as the data is concerned, for as long as I’ve been on the team it’s been the case that the team that gets an early level lead has a much higher chance of winning the game. If one team is Level 10 vs. the other team being Level 8, their chances are much higher to win. This is also true of the team that destroys a Keep first – they have vastly increased their chances of winning. Of course, there’s a lot of lurking variables here, but the comeback games that you’re describing are much more rare, but they are also much more memorable because they had the unexpected outcome. Snowballing has been and always will be an issue that has no perfect solve – our game hinges on a power fantasy where you want to get stronger and dominate your opponent over time, yet we also want the other person to still feel like they have a chance to win while they’re essentially acting as your punching bag as your “reward” for playing well in the early game. Essentially what I’m saying here is that the notion that all games hinge on one late-game teamfight is not happening most of the time, though it does sometimes happen. The question here is: Is this OK, and if not, what would your expected outcome be, and where on the spectrum should a late-game teamfight exist when it comes to comebacks?
I can tell you right now that even in the super late-game, if one team is ahead and loses a teamfight by only 1-2 people, those extra buildings that they have absolutely do act as a barrier/second life of sorts to keep them in the game when otherwise they could lose. The question here is where on the spectrum is it acceptable to lose the game when you were previously winning? Is it when you lose a fight by 1 Hero? 2? 4?
A comparison that I think is useful with these kinds of thought experiments is to compare MOBAS to traditional Sport games. In basketball for example, the winning team doesn’t get taller, or faster, or more accurate with their point shots. MOBAS have a growth and power fantasy that is a large selling point to players, and the drawback of that growth that we have to accept to a certain degree is that snowballing is much more of a thing, and that comebacks are more important to have be possible to counter-act that. This means that the swings in both directions (snowballing and comebacks) in a MOBA are going to happen more often than a traditional sport due to the decision of having power scaling based on performance. When we reduce this power growth too much, we get complaints like: “the early game doesn’t matter”, or “I can’t carry hard enough”. When we increase this too much, we get the complaints like: “all that matters is the first map objective” or “why bother even playing anymore, the game’s already over”. The weird thing to wrap your head around is that all of these are both true and false on an individual and larger scale due to how different each game and each person’s interpretation of that game can be. The part where we as designers come in is where on the spectrum of Snowballing vs. Comebacks do we want the game to be as a whole.
My personal view is that if I’m ahead in a game and I lose a full fight 0-5, I deserve to either lose the game or lose a fort/keep and have my core exposed and likely damaged. This is because I likely had lots of opportunities to bully the enemy team throughout the game while I was ahead, and if I won that last teamfight by an even less margain than a full wipe, my team would have won the game. The enemy team had to drastically outplay mine to win by enough to end the game, so they deserve the comeback win. This is actually the most rare of cases though, and more often what happens is my winning team loses a fight by a small-medium margain, the enemy team takes a keep and exposes the core, then the winning team makes a second mistake and the enemy team comes back to win the game. In this case I also think they deserve the win.
This is all theoretical where the game ought to be, but let’s talk about where we’re at right now:
Currently I think there is an argument that our tower changes have punished aggression a little too much with the current tuning on towers, particularly in the earlier portions of the game, and we have some changes coming that should help address this.
I don’t believe that every game is arbitrarily decided by one late-game team fight. The data does not support this, nor does my personal experience playing the game a ton and watching others play. It does happen, but I don’t think it happens too much (if the meta shifts to Nazeebo being first pick/ban, you’ll know we’re on our way there :P)
Core Armor is something we’ve discussed in the past. I think it’s a valid idea, but has some aspects that concern me. I’m partially dissatisfied with the current implementation on Alterac Pass, largely because it’s a hidden rule that it’s almost always a super bad idea to try to attack the enemy Core with only one Keep down. This took players longer to learn when it first came out that I was happy with, and to this day players try to take this largely incorrect fight more often than I would like, causing a winning team to make a huge misplay due to a mechanic that they don’t really understand. Armor in general is a powerful mechanic that is hard to sell to players, so if we were to do something like this we would probably need much better Art/UI to make it super clear what’s going on.
Thanks again for the post. I probably rambled a bit much and didn’t even cover everything that is going on with this subject due to how complex it is, but I hope it sheds some light on how we see things.
One thing sports nearly universally have: a clock. The advantage superior teams have is that the other team runs out of time. In hots, the only real clock is Hammers infinite bfg, or an open lane to the keep. I think perhaps the minion scaling should increase starting at level 16 to the point where winions are an extreme threat at level 21 and beyond.
Funny enough I wrote a paragraph about this difference but cut it in the name of brevity.
I’m not really a fan of minions scaling to be out of control mostly because it’s super unfun to be stuck in your base killing minions all day. It cuts down on aggressive play in a big way when they’re too strong.