Dota 2

Yet another look at fundamental mechanics and game design

Briefly about the writer (regarding MOBAs)

Before playing DOTA2 I had not played any games specifically of this type. Before getting this computer and starting to play DOTA2, I was mostly playing chess or GO or dwarf fortress, as well as mtg the card game, along with like idle browser games. So not a huge competitive e-sports background in that sense. Earlier in another lifetime, I played guild wars and shadowbane, and there was also like ultima online, which were fairly pvp oriented. Shadowbane is actually one of my favourite games, and I'll talk about it briefly because it sort of relates to this.

Briefly on shadowbane

Shadowbane is a highly complex open world pvp game with also farming content that was super boring, and in a sense that's a big similarity to DOTA2. In shadowbane you would want to farm npc monsters to make progress, get gold for items, and use those items to win other players. The items would then be used in things like sieges, which was basically about destroying or capturing a player owned city, as well as these so called mine fights. The playing field was never even, there was no set limits on how many people would participate on an encounter, nor there was even any clear sides to an encounter, necessarily. You might have a battle like 40v30v10 + some random folks just passing by. The build system was near perfect, the best I've seen in any RPG type game. You would have 580-640 points to distribute on skills, as well as these passive mastery lines, which would unlocks skills. And training a skill would have the liberty of placing anywhere between 1 to 40 points into it. Technically, there was a number of permutations for these skillpoint combinations that it's not possible to take into account each possibility nor have two characters randomly end up being the same. But practically, a lot of that complexity reduced to noise, as is often the case with such designs, but in this regard shadowbane managed to maintain the complexity due to various mechanics.

The character build process was combinatoric and sort of modular, you would choose a weapon, but maybe not train it, and depending on your choice, you would get a set of abilities, and new abilities would be unlocked depending on how much points you put into the weapon passiveskill. There was around 10 different types of weapons and for each of those there were a couple variants, and also the items themselves were using a combinatoric enchantment system so you had a like a prefix + suffix combination, and each type of item would have a limited list of enchantments you could select. There was about 15 races or so, and about 20 classes or something like that, with 4 base classes to choose from. The total number of baseclass+advancement class combinations wasn't that high though, it was in the order of like 50 or something like that, less than DOTA2 heroes in other words. But you also had the choice of race, which increased this count multiplicatively, and also you could choose a different weapon, which again did the same thing, and on top of that you could design freely how many points you're going to put in the weapon mastery. So this combinatoric system allowed for a very large number of different builds, practically the number was still lower as you would have certain logical restrictions on what would make sense. A lot of builds just wouldn't make sense, thus chopping the total count of permutations that were viable by a considerable degree. Regardles, basically you would have around 4-8 skills from your baseclass, around 10 skills from your advancement class, and about 8 weapon skills to use, as well as a combinatoric set of these like mini-classes called disciplines, you could select, which would give you access to another 10 skills. Basically you could have about 40 buttons on your interface, and you had a stance system, where you could between defensive, aggressive, normal, precise stance, and the details of these chances would vary depending your baseclass+advancement class combination. But basically you could choose this mode from 4 possibilities, that would give you background benefits such as accuracy, magical resistances, damage, defense, etc.

The point is in the context DOTA2 that it was a lot more complex than DOTA2 is, with this incredible number of abilities to use actively and even a larger number of builds. However, in that game basically every ability had a casting time, similar to channeling in DOTA2, and the fight-pace was a bit slower, so a very short fight would be like 10 seconds, where as in DOTA2 a very short fight is just like.. a single attack by phantom assassin so that the time is not even clear. If SB would have instant cast abilities, it would be completely unplayable as it would be come a total chaos and you wouldn't have any time to think. And on the flipside, if DOTA2 had abilities such that each ability had a casting time, it would get very slow-paced and boring. But the chief point here is that the absence of cast times makes playing with a limited list of abilities more complex and you don't have as much time to think. Personally I wish DOTA2 would have some more abilities and more choice, and more complexity, but it's difficult with the instant cast times and so forth. DOTA2 is already very complex overall.

But what made that game interesting was the build design, and the how the npc content interacted with the pvp content. Like farming was super dull, but it was interesting, because someone could show up and interrupt your farming.. Additionally, items on backpack would drop to the ground, so you could basically try and farm the other farmers, a bit like this foodchain type of thing.

Read more:  Let's talk about mid Chen


What makes DOTA2 interesting to me, is that it has a similarish type of complexity about it. However the hero or character build doesn't have any complexity. You've a talent tree, and abilities, and when you're at max level, you can't even make a choice regarding which abilities you get, because there's only 1 way to spend all ability points excluding the talent tree, thus you always get 4-4-4-3 and there's just 1 way to distribute those points. However the early game is more interesting, because a choice is involved, do you want to get 2-3-2-2 or 3-1-4-2 and so on, this is partially restricted because ultimate abilities are only advanceable at specific levels and it's almost always the correct choice to put your point into it when possible, and you also can't max out any ability directly, but have to distribute your points by force.

DOTA2 item system is a character build extension/surrogate

In the absence of a complex character creation system DOTA2 is using a different type of system, where basically the items are your build and there's a lot of ways to choose them. This is in part because of the free item slot system, you can take 3 chainmails if you want, which isn't possible in your typical item slot system where you can only choose 1 chest piece etc. The fact that you've a numerical restriction to 6 items, also creates these opportunity cost issues and makes things more interesting that way.

DOTA2 comparison to shadowbane – live builds vs frozen builds

The real thing that makes DOTA2 more interesting than shadowbane is that you build your character during the game. In Shadowbane, for comparison, although on paper it sounds similar, you're similar things, farming gold, choosing items for your build, spending the gold on those items, and then trying to win pvp fights. But these two processes are more distinct, so you can like afk farm passively in shadowbane and forget about pvp, and then use the same set of items for a week in pvp. Basically you could have the same items in every fight all the time, if you wanted. Meanwhile in DOTA2, you kind of interact with your opposing team to make your decisions, you can look at what your opponents are building, and you can counter their item selections, you can pay attention to which hero is doing well and what position they're playing, and try to take that into account with your role and your itemization in mind.

So I would call the DOTA2 builds as live builds, compared to the SB builds which would be frozen.

Investment quality of item purchases

In DOTA2 items have an investment value, which makes things a lot more interesting. Take hand of midas for an obvious example. It's not a very cost-effective item for any kind of pvp, but it increases your rate of farming. Note though, that there's an opportunity cost, since you could get another item that increases your farming rate also, but if midas does it more effectively than your alternative, it would be appropriate to call it a farming item. Basically you give up some strength in your early game, to gain some strength at another point in the game. So this type of over the course of the game thinking and investment into items makes everything much more complex and more interesting. Additionally the game isn't about having an even playing field either. You don't have the same amount of gold to use for your items.

You could technically have a match of DOTA2 where each player gets 10000 gold, and they buy the best items they can think of, and it would be another type of match. However, in this type of game, DOTA2 would be much more boring, this is in part because DOTA2 isn't that complex. The number of reasonable item choices is fairly restricted. However, because of the livebuild system and these investment properties, everything becomes more complex – and as the item system interacts with this process, it makes using it more interesting as well.

Frozen builds as means of analysis

However you could try and understand the complexity -or rather the limitations of it – of the DOTA's item system by thinking in terms of these frozen builds. Let's say you get to choose two items with 5000 gold, and your hero is Y and your opponent's hero is Z. What kind of options do you actually have? There are not that many when you start to think about it.

And the point here is that if you can somehow get a grasp of the build strategy game on the frozen build level, you could potentially increase that complexity and find ways to make the live build game more interesting also. If the initial item choices were more significant and also more interesting, that would make the more interesting.

Counter-intuitiveness and complexity

As long as things make intuitive sense, complexity doesn't become an issue. Like a player doesn't need to be a master to understand that broadsword increases your damage and chainmail increases your defense. That makes intuitive sense. Things get complex when you've to choose between two different ways of defending, do you want a fluffy hat or ring of protection, etc. Agility gives heroes both damage and defense. Basically to refine the strength of your choices, you would need to increase your mastery over the game, while even a beginner could sort of intuit around what's the point of things.

Read more:  Markov generated patch notes

Lane mechanics are fundamentally counter-intuitive and somewhat anti-social

I think one of the challenges for new players is that the laning mechanics are counter-intuitive. A person would think okay these creeps are my allies, I need to help them survive and make progress forwards, and defeat the opposing creeps and so on. But that's not the case in DOTA2, instead you've to try and deny your own creeps in the early game, and avoid pushing the lane unnecessarily unless you want to get time to farm, basically you don't want to defeat the enemy creeps too fast. In fact you would rather often pull the minor camp of neutral creeps, to defeat your own allied creeps, so the opponents don't get to defeat them.

Faulty incentive crafting and bad design is at the core of this system – regardless it turned into an interesting feature by chance

The fundamental underlying problem basically has resulted from the following type of reasoning: The number of creeps spawning on both sides is constant. They're like two jets or streams of energy constantly clashing into each other, and they're at equilibrium. The number of creeps perishing over time remains constant, regardless of what the players do. They can cause a temporary shift in the number of perishing creeps, but as their spawn rate remains constant, this will inevitably be met with a later decrease in the number of creeps perishing, ( well almost inevitably ). So what incentive do the players have to defeat these creeps? What's the point of doing that? The same number of creeps are going to spawn anyway, and the neutral creeps or the towers are going to defeat the new creeps. There's like no reason to participate in this creep war. Ha! But you could incentivize it by making it so that the players get gold when they defeat the creeps. But only when the players defeat them. No gold when the lane creeps defeat each other, because if you would get gold in that situation, again players would have no real reason to do anything. So we can incentivize this by only giving gold from the last hits, this way, the players have a good reason to play forwards.

…But what resulted instead is this counter-intuitive system of last hits and denies. Which is fine when you get used to it – but it is counter-intuitive so new players might find it challenging. That's okay, as long as this counter-intuitiveness isn't overwhelmingly omnipresent, and confined into this small sector of the game, it's okay. The rest of everything still makes intuitive sense.

Last hitting is interesting because of the presence of the opposing team and the over time live build system with uneven playing field from uneven networth

What makes the last hitting and denying interesting is that you've to do it in the presence of your opponents. You are competing over getting gold faster than they're, you can have very many combinations of early laning heroes, two basically on each side, and the sidelanes are asymmetric. Many different strategies and things to take into account when you want to understand how to go about this last hitting and denying business. It has become an interesting sport in other words. But I wonder if there's a way to increase the intuitiveness of the game while preserving the last hitting and denying culture and sport?

The fact that getting gold faster than your opponents is interseting, is because of the live build system. You interact with the item purchases over the course of the game, thus this gold farming is also interesting. Obviously the items wouldn't be so significant, just the laning stage wouldn't be that significant or fun. The impact it has on the game later, is what makes it significant and interesting. Which by the way is a difficult challenge when you want to preserve chances for the losing team, yet also preserve the significance of getting advantage early on. How do you solve this equilibrium? Also is it even reasonable that the losing team has a chance if they're losing really bad? Etc.

Last hitting and denying system has also resulted to some anti-social properties. And a new player experience description

So when I was a new player, I played some DOTA2 with some public matches. One experience that happened to me a couple of times, was when I saw my team mate like farming the neutral creeps, and I thought they were being a little bit slow, and I wanted to help them farm, so I also attacked the creeps. They thought I was being a jerk and trying to steal their gold, and started pinging the X retreat symbol on the minimap. Understandably. However, this experience illustrates a certain anti-social property that is due to the mechanics. If you would find it reasonable to co-operatively defeat NPCs (think Roshan), it would be helpful for the team spirit and pro-social interactions. But because of this last hitting system (as well as having to use time efficiently in general), this kind of activity is against the grain of the game.

Additionally when two players play on the lane, the players have come up with an optimization strategy on the grandscale, such that each participant of the game is classified or ordered based on their farming priority, since there's somewhat a limited count of gold and experience available, the players have come up with this culture that allows distributing that gold and experience according to their will and the plan is that it's going to increase chances of winning, when the most optimal player gets most of the gold. Which is a perfectly fine, after all the command of smaller and larger forces is just about dividing their numbers and deciding what each divisions is going to be doing, according to Sun Tzu if I remember correctly.

Read more:  Wraith King Rework

However it would be more pro-social if you could have your crystal maiden and your juggernaut on your safelane, both doing their best to defeat the enemy creeps as fast as possible, and get he last hits. And this way the two players would have a shared goal, and the interactions would be more pro-social. However, while this would be pro-social and intuitive, it would go against the last hitting and denying culture that the players have created as a strategy in the game. And it's an interesting culture, and once you get used to it, you can kind of start to like it.

Supports not participating in the game is slightly oppressive

I'm gonna start with a super-outlandish comparison that the supports are like the citizens of North-Korea. They're kind of oppressed. They'd want to get some gold for themselves and play the game, but they're met with this culture that your role is not to do any of that, but instead passively hold back without playing the game, and let your carry do it. Especially if someone insists that you must do that, it can be oppressive… But once you get used to your role, and your ego adapts to this, and you see this is appropriate and pro-social, you become to be in terms with it. In fact if there would be a support player, like a Crystal Maiden, who would go against this culture, you would find them anti-social. CM trying to take your last hits would be someone who is acting like a jerk towards their carry, or juts on purpose stealing their last hits knowing that it's going to be bad for the carry. Kind of like if you have a citizen in North-Korea starting a protest, they're just causing a fuss for no reason and being like a jerk. You've like accepted this social order, and your ego has accomodated to it, and that's the perspective from which you look at it now. It happened to me too. I thought this supports having to stay back and kind of not play the game, juts buy some wards and try to do something, was really rather oppressive, but once I got into the game, I thought it was natural and it makes a lot of sense, and there's no problem with it.

But then it's useful to understand that there sort of is a process like this that the players go through, that the expected way of playing for people is due to mechanics that do not promote pro-social behavior. The mechanics *could* incentivize the crystal maiden and the juggernaut both doing their best to get last hits and gold and rich asap. And they wouldn't need to be stealing from each other. If this was the case, and the mechanics were such, then it would be extremely weird if someone asked their support to stay back and not get any gold. Like who do you think you're are? Are you like a pos1 crystal maiden, huh?

Summa summarum

  1. I would like to see an increase in complexity of DOTA2 both in regards of hero building and itemization
  2. The gold incentive system has some problems of counter-inituitiveness, oppressiveness, and a lack of pro-social qualities
  3. The gold incentives and lane mechanics could be reviewed and you could try to improve on the situation.
  4. The way the game is is just fine, I like it as I'm used to it now, and I love playing supports, that's basically not where this coming from. But I think there's room to improve without jeopardizing the competitive sport of this last hitting and denying culture.

It's a long read, but hopefully it will offer some insights to our DOTA players, and hopefully it will attract the attention of the developers, and they gain some insight into what's going in the game they've created. Have a nice day and enjoy some DOTA2!


Similar Guides

More about Dota 2

Post: "Yet another look at fundamental mechanics and game design" specifically for the game Dota 2. Other useful information about this game:

Top 7 NEW Games of February 2021

Looking for something new to play on PC, PS5, PS4, Xbox, or Nintendo Switch in February 2021? Here are the notable video game releases.

Top 20 NEW Open World Games of 2021

2021 will bring us tons of open world games for PC, PS5, Xbox Series X, PS4, Switch, and beyond. Here's what we're looking forward to.

You Might Also Like

Leave a Reply

Your email address will not be published. Required fields are marked *