Competitive Play and Luck (Big Post)

[BinAly]

Fellow forum-mates, I know that this is a very delicate subject, and by no means I want to start a 'flame-war' of sorts by trying to discuss it. Keep this topic clean and let's try to understand why Luck should be in, or out of this game.

Understand that by Luck, I don't mean things like using Devourer's Hook to catch someone inside the FoG, without really knowing where they were. Thats not the Luck I will be talking about. I will call this, healthy-luck, or situational-luck.

The Luck that I will be talking about is the one that is implemented into the game on purpose, written into the game-code. For example, Legionnaire's Whirling Blade, which is purposefully created to have a chance of working every time the Hero is attacked. I will call this by Luck, with capital L.


Comming from DotA, HoN is, I believe, aimed to be a very competitive game, with entire clans training to get better with every hero, studying opposing strategies and maximizing their own abilities, much like a soccer team, or a basketball team.
Players will play many hours just to reach that state fo "uncounscious though", were you play by instinct, reflexivelly knowing what to do in each moment of a match.

And here is were Luck, with capital L, may ruin the competitive aspect of the game.

You can't train your Luck. You can't control it, and neither your opponent's Luck. You just play and pray for it to work in your favor.
Sometimes, you get Lucky, and those % work brilliantly for you, allowing you for almost impossible Double Taps, or very, very close escapes. Right, those may result in "fun" moments, but and when the inverse happens? When you lose games because that 20% didn't really worked 1 time out of 4? When you trusted your skills and they failed you, not because you lacked training, but because "dices" just didn't went well for you on that match?
Things like these break the competitive aspect of the game, basically because your ability as player can't be trusted anymore, neither by yourself.

If we analise other competitive games (not really eletronic games, but games in general) searching for Luck Properties, we won't find any, really.
For example, imagine if the basket, on a basketball game, had a 20% chance of exploding the ball before it went through it, effectivelly denying a point that would be made, for sure, if the Luck wan't purposefully added to that basket.

Another example would be a soccer game were the ball had 25% chance of evasion. Even in very easy defenses, goal-keepers would miss it and a point would be made.

Ok, getting back to eletronic games. People might say that since we can't fully simulate real sittuations in eletronic games, we need to add Luck to it, for example, to simulate a hero that is very nimble and hard to hit.
That's not true. There are several sollutions that could be implemented that doesn't add artificial Luck to the game.

For example, for Starcraft 2, developers are removing the 30% miss chance for elevated terrain. Instead of it, they are simply letting units on higher ground fire without being revealed by the FoG.
See, we still have an advantage from being at higher ground, but without any artificially generated Luck involved. Players won't need to pray for their opponents to miss, or pray for a hit when in lower ground.

There are several other examples, even in HoN, of what could be made to avoid artificial Luck.

We all know a hero that's called Puppet Master. His third skill, called Whiplash, makes him release more damage in every fifth attack he makes. That's a skill that adds ZERO artificial Luck to the game. It could easilly be made as a chance of 20% to release a powerful attack, but it wasn't, and it is more reliable the way they made it.

Now, look at Scout, for example. He has a skill, Improve Dexterity, that gives him a chance (I don't really remember the exact numbers, but I think it is something like 15% or 20%) to deliver critical damage. This skill adds artificially generated Luck to the game, which is bad for competitive game-play. Why not make it like Puppet Master's Whiplash? Improved Dexterity could easilly be done as a skill that let's Scout deliver more damage every fifth attack he makes. More reliable, less frustrating.

Another example would be Legionnarie's Whirling Blade. If I'm not mistaken, it has a 20% chance of going of. Why not make it go of every fifth attack? It would make it more reliable, more predictable. You would be able to think about strategies that didn't count on a chance to really work.

I've already seen people state that abilities like Whirling Blade would be overpowered if they went of every fifth attack for sure. That's a wrong way of thinking.
Right, every fifth might be too strong, but then make it go every 6, or even every 8 attacks. It doesn't matter, it could be balanced without allowing Luck to interfere with competitive gameplay.

Even evasion can be balanced this way. Wingbow could make you avoid every third attack, instead of giving you a 30% chance.

Now I want to read what you guys think about it.
Also, I would really like, if possible, to read what S2 things about this subject. If they support artificial Luck, or if they would want to remove it.

Thanks for reading through all of this. :D

[Elerion]

I am always in favor of removing uncontrollable luck from competitive games. Games being decided because Lina/Pyro's ult did or did not get canceled by Void/Chrono's Backtrack is not fun, nor does it improve the competitive aspect of the game. Similarly, Ogre Mage/Blacksmith has infuriated players on both sides for years.

Note: An alternative to "Avoid every nth attack" and other "exploitable" variants is "Avoid x% of incoming damage".

[Lyte]

Making those skills predictable makes it far too powerful and they would have to re-balance everything downwards.

I agree with you, luck should be "removed" from competitive games, but all competitive games do have a bit of luck. For example in Warcraft III, there's luck in the items you get from creeps. Sometimes you'll get a good item, sometimes you won't. In Starcraft and other games that utilize "terrain ramps," shooting up ramps results in a 30% evasion. Sometimes you hit, sometimes you don't. Even in sports games, there are tons of "luck." It's just shown in a different way. Sometimes in basketball, you have a 3-point shooter with 25% chance to score, and he'll make 4 baskets in a row from 3-point range--that's statistics "luck." In soccer, sometimes you'll score from mid-field, even though it is a low%. That is statistics "luck" too. All games have this.

These types of things are OK in the competitive environment, because over time, the luck "balances out." If you have a competitive game where 1 lucky event completely changes the game and turns losers into winners, or winners into losers, then that's a problem.

[Fliesen]

luck's part of the game, every game mechanic is based around fuzzy values ... be it min -> max damage, be it missing shots from lower terrain, be it 20ms difference in ping

except for blacksmith (which you could just see as a gimmick class, because if he's not on a lucky streak, he's pretty sucky), "luck" doesn't create that many game-breaking "spikes" (be it damage, dodge, etc.)

as mentionned: "every 6th" attack basically means you can just hit the enemy hero with every 6th attack, like a well played puppetmaster does and that'll end up in endless "back and forth" thumbwrestling-like-gameplay

[Elerion]

Not quite, Lyte. The basket player missing the hoop is the equivalent of the PotM missing that crucial Arrow, or that Zeus failing to juke a deadly nuke - he failed a skill-based attempt. Void backtracking a Pyro is completely independent of skill.

For things to balance out, you also need to have a large enough number of observations for the sample mean to approach the expected value. There are a number of luck-based things in DotA that occur so few times that you can get very skewed results.

[BinAly]

I am always in favor of removing uncontrollable luck from competitive games. Games being decided because Lina/Pyro's ult did or did not get canceled by Void/Chrono's Backtrack is not fun, nor does it improve the competitive aspect of the game. Similarly, Ogre Mage/Blacksmith has infuriated players on both sides for years.

Yes, that's a good sintetization of how artificial Luck may break competitive games.


Note: An alternative to "Avoid every nth attack" and other "exploitable" variants is "Avoid x% of incoming damage".

For sure, my suggestions weren't written as the better ones. Agility already reduces damage instead of giving evasion chance, which is pretty good in my opinion.


Making those skills predictable makes it far too powerful and they would have to re-balance everything downwards.

So, in your oppinion it isn't worth the effort? S2 is already rebalancing and recreating the game from scratch, it might be easy for them to test new changes.


Even in sports games, there are tons of "luck." It's just shown in a different way. Sometimes in basketball, you have a 3-point shooter with 25% chance to score, and he'll make 4 baskets in a row from 3-point range--that's statistics "luck." In soccer, sometimes you'll score from mid-field, even though it is a low%. That is statistics "luck" too. All games have this.

That's what I called healthy-luck, or situational-luck. Notice that that kind of luck isn't artificially generated.
- You may train to hit the basket all the times. I already knew players that could hit for sure.
- You may train to hit the goal from mid field.
Your examples equate to my Devourer example: Throwing the Hook at the FoG without really knowing where your opponent is. That's fair-luck, healthy luck, and they involved conscious though of the players.
- Again, you may train to predict the way your opponents move, and hit them in the FoG with Mind-Games.

I ask you to try to separate healthy-luck from artificial-luck. Competitive playing is not about Gambling, it is about training and skill.



In Starcraft and other games that utilize "terrain ramps," shooting up ramps results in a 30% evasion.

Like I stated, that's already being changed for Starcraft 2, which is a game that is being made for high-competitive play, while Starcraft 1 wasn't really made with that in mind.

[BinAly]

l
as mentionned: "every 6th" attack basically means you can just hit the enemy hero with every 6th attack, like a well played puppetmaster does and that'll end up in endless "back and forth" thumbwrestling-like-gameplay

But that's about skill, and not about praying to hit. For competitive game, it might be better, don't you think? :(

[Bahamut]

nah sorry, i think this is pure crap

the game is how it is, DotA is a huge succes because THIS IS THE WAY IT IS, if you change it then it's not DotA and you dont know if it's gonna be a succes

Take LoL, HoN and DemiGod, which one will succed the most? HoN, why? because is doing things they way they are ment to be done

[Fliesen]

first: i think it'll end up in weird, ugly, UNFUNNY hit'n'run

also it against every roleplaying convention ... imagine yourself being a skilled ninja, sneaking up unto your enemy ... you stab him with your dagger multiple times. one of those strikes pierces his heart .... you're lucky, Hitler's dead!!

you can't just slash 5 nazi officers and then be guaranteed to stab the heart on Adolf :P

it's just unrealistic to be able to control "critical" hits like that ...
luck / chance is a part of life aswell as of games :(

[BinAly]

nah sorry, i think this is pure crap

the game is how it is, DotA is a huge succes because THIS IS THE WAY IT IS, if you change it then it's not DotA and you dont know if it's gonna be a succes

Take LoL, HoN and DemiGod, which one will succed the most? HoN, why? because is doing things they way they are ment to be done

I agree that DotA (HoN) is a huge success already, but we could improve it. I believe that it is worth the risk.

For example, even in DotA, IceFrog added (because competitive players were asking) a sure-hard-counter to evasion, which is now, Monkey King Bar. it gives users True-Strike, which totally negates evasion-chances, banishing luck from the game.

Just look at the heroes that are being added, less and less of them work on chance-skills. Artificial Luck kills competition, because you can't compete against what you can't control. :(

[BinAly]

first: i think it'll end up in weird, ugly, UNFUNNY hit'n'run

also it against every roleplaying convention ... imagine yourself being a skilled ninja, sneaking up unto your enemy ... you stab him with your dagger multiple times. one of those strikes pierces his heart .... you're lucky, Hitler's dead!!

you can't just slash 5 nazi officers and then be guaranteed to stab the heart on Adolf :P

it's just unrealistic to be able to control "critical" hits like that ...
luck / chance is a part of life aswell as of games :(

Being a Role-Play-Gamers (Storyteller) for 10 years, I know what you are talking about. It is pretty fun to let the dices decide the way the story will progress, but RPG isn't competition, it is a cooperation to build a great story.

We can't have Artificial Luck at competitive games, in the long run it ends up working against them.

Obs: Why exactly it would end in Hit-and-Running only?

[BLUEPOWERVAN]

many great competitive games involve luck. in that way they reflect life, where most challenges are of a stochastic nature

to clarify, poker is a wonderful game, that to an uninitiated outsider would seem dominated by luck. the same is true of backgammon. Even in "high-skill" competitive online gaming (say fps's), weapons have spread, and some spawn locations are random. In slow TBS or war games, battles results are heavily influenced by luck.

it's simply naieve to assert that competitive cannot have important random elements, since there are many examples of them doing so. A game is not interesting if a game is primarily determined by luck, but some smaller random elements are totally fine.

[akitoes]

If we analise other competitive games (not really eletronic games, but games in general) searching for Luck Properties, we won't find any, really.
Poker has been one of the most competitive games in the past years.
http://www.friday-night-poker.co.uk/mediac/400_0/media/72.jpg

But I halfly (wut) agree with you, despite the fact that "Luck" has been part of video games especially RTS for a long time (Warcraft/Starcraft also had crits, evasion and all that stuff buddy :>)

Related to picture, poker sharks know that luck is their worst enemy as they can lose big money on unlucky pots, but also their best friend as amateurs wouldn't play if there was no luck.
Think about it, why are there so many more poker players than chess players ? :>

Crits, helixes, and other randomness are a big part of what makes the game exciting, thus more accessible to players. Randomness made poker successful, and makes esports successful too.
Actually, losing a hand to misfortune in poker is far more devastating than getting multicast procced !

[Vadi]

I agree that extreme examples should be removed, but things like puppet masters whiplash are 'abusable' and it simply becomes another mechanic. I'd like a small amount of chance (like critical hits) to be gin.

[Lyte]

Not quite, Lyte. The basket player missing the hoop is the equivalent of the PotM missing that crucial Arrow, or that Zeus failing to juke a deadly nuke - he failed a skill-based attempt. Void backtracking a Pyro is completely independent of skill.

For things to balance out, you also need to have a large enough number of observations for the sample mean to approach the expected value. There are a number of luck-based things in DotA that occur so few times that you can get very skewed results.

Sorry, I meant "video game" sports :p Like if you were playing NBA Live at the World Championship Games, these innate stats generate "artificial" luck in the game.

[BinAly]

Even in "high-skill" competitive online gaming (say fps's), weapons have spread, and some spawn locations are random.

That's a good point. For example, Counter-Strike has 'spread' on most rifles. However, in my view, the 'spread' is a nerf to the weapon, and no a buff. It is there just to force players to use burst-fire instead of holding their finger on the mouse.
There aren't weapons that get better by Luck, they only get worse.


Poker has been one of the most competitive games in the past years.

Oh, man. It was very faulty from my part to forget about Poker. Face-Palm to me. :(

But, in my view, Poker isn't really about Luck, it is about bluffing. It is a psicological game, more than a card-game. Anyway, that's a good example of how luck may create good-gameplay.


I agree that extreme examples should be removed, but things like puppet masters whiplash are 'abusable' and it simply becomes another mechanic. I'd like a small amount of chance (like critical hits) to be gin.

By, 'abusable', you mean, you can get better at it by training? :D

I'm asking most of those questions because I'm working on an MMO, and we are in the middle of designing abilities and skills. I'm in favor of not allowing luck to interfere in competitive gaming, but I want to understand what exactly is good about it.

I don't think that there are problems on letting artificial Luck influence on game-play, as long as your training allows you to minimize it's influence.

[duglaw]

there is always the option of making the skill less luck based. 20% chance for +100% damage might be made to 50% chance for +40% damage. gives the same +dps in the long run, but less luck involved.

[BinAly]

there is always the option of making the skill less luck based. 20% chance for +100% damage might be made to 50% chance for +40% damage. gives the same +dps in the long run, but less luck involved.

That's a good aprocah, also, something in the lines of what Vadi would like. Not total removal, but less influence.

[mucergo]

I think luck adds a little flavor to competitive and non-competitive games (critical hits, etc.). But, luck can´t act like a decision point in the game. Players must have a chance to minimize the effect of luck. For example; Poker, Magic: The Gathering. These two games have luck involved, but the same players are world champions every year. Why?
Its simple, with strategy they can minimize the luck factor of the game.
Resuming: Luck is important for fun, but strategy and training must minimize the luck factor.

[Lyte]

I think luck adds a little flavor to competitive and non-competitive games (critical hits, etc.). But, luck can´t act like a decision point in the game. The players must have a chance to minimize the effect of luck. For example; Poker, Magic: The Gathering. These two games have luck involved, but the same players are world champions every year. Why?
Its simple, with strategy they can minimize the luck factor of the game.
Resuming: Luck is important for fun, but strategy and training must minimize the luck factor.

Which is really where DOTA is at right now. The same teams win consistently, so artificial luck isn't a big enough factor to allow a random team beat the top teams--which, I think is fine so far.

[ChimEuphoria]

Your examples equate to my Devourer example: Throwing the Hook at the FoG without really knowing where your opponent is. That's fair-luck, healthy luck, and they involved conscious though of the players.


"Fair-luck"? lol this is called skill and intuition, not luck. Get over yourself, Backtrack has been in dota for as long as i can remember, and I don't see it getting removed any time soon. You dont like facing him? Pick him for your team. The game is perfectly fine with %chances implemented.



I ask you to try to separate healthy-luck from artificial-luck. Competitive playing is not about Gambling, it is about training and skill.

This isnt a fighting game, training and skill arent worth **** if your team isnt coordinated effectively.

Skill takes a backseat to strategy in all aspects of the game aside from last hitting and denying.

[jamby]

Luck is probability taken personally.

[Lyte]

"Fair-luck"? lol this is called skill and intuition, not luck. Get over yourself, Backtrack has been in dota for as long as i can remember, and I don't see it getting removed any time soon. You dont like facing him? Pick him for your team. The game is perfectly fine with %chances implemented.

This isnt a fighting game, training and skill arent worth **** if your team isnt coordinated effectively.

Skill takes a backseat to strategy in all aspects of the game aside from last hitting and denying.

And what happens when two skilled teams go at each other? Individual skills begin to matter. If you don't have the individual skills, you have no chance of getting out of pub matches. Individual skill is a pre-req to even thinking about higher end play and playing with good teams that utilize strong strategies.

[antigrav]

I think that Luck is actually required in a game like this. I think that good players should and will adapt to situations that do not bend in their favor. This being said I am coming from the perspective of playing a card game professionally like Magic or Texas Hold'em. Yes, there is luck involved. But it's also about being aware of your chances, the odds, and then dealing with them with the skills involved in the respective games. I think that Luck is a fine aspect of HoN even though nobody likes getting permabashed or letting the enemy carry get away because Lady Luck blessed his Wingbow on the finishing hit. Good players will make themselves better and try to establish situations where they are better than Luck and will dominate it.

[jsat]

I hope you all read this:

I was JUST going to bring up mtg. Mtg is the first and certainly the deepest, furthest explored, and most interesting (as far as maximization of interactive capabilities) of any strategy game outside of classical games of a much more finite amount of interaction. To introduce this complexity in a way that leads to a playable game the rules of magic had to find ways to place a limit to the number of interactions in an actual game state. Unlike chess, where you have every piece available, magic has only a finite set of pieces available. But to add replayability to a game that is fundamentally a discrete set of interactive capabilities they had to add RANDOMNESS to the initial state of each game and the introduction of new resources.

What is the relevance here? In a game like dota or any sport we start like chess, with every player on the board and known. Where it quickly differs is in games like this there are for all real purposes infinite interactions (though many negligible). This by itself insures replayability where even named common motifs have insane variability and healthy variance of game state. It is unnecessary to introduce artificial variation in the form of die rolls for that purpose.

HOWEVER, randomness leads to a different set of plays. When you can not predict exactly what will happen (which you never can in hon anyways, though mtg is more explicit) you can not make exact plans. Pregame strats always resolve around reducing the infinite numbers of ways a game can play out and funneling them towards some game state like turtle or push or gank. But these states are still inherently unpredictable exactly anyways. So what i am getting at is although you can never dare claim evading laguna blade was skill, but playing in such a way as to maximize your chances is how you win a competitive game. This means letting the best three point shooters getting set up for the shot, putting in the right pitcher at the right time, going for the gank when you can see everyone on the map and plan it well, and orge magi trying for the blast every time he can 'just in case'. It is not NECCESARY for this game to have explicit randomness, this much is true, but it is not skillless. The same magic players have been winning upwards of 100k over the least decade, this is not luck; this is skill in a game of explicit chance. To say that randomness cheapens a victory is to discredit the accomplishments of many hard working intelligent people.

[akitoes]

Oh, man. It was very faulty from my part to forget about Poker. Face-Palm to me. :(

But, in my view, Poker isn't really about Luck, it is about bluffing. It is a psicological game, more than a card-game. Anyway, that's a good example of how luck may create good-gameplay.

Yup. If you want to learn more about poker, read on :>
The whole game of poker is centered about luck :)
Winning money in poker is very easily summarized : get the most of your money in the middle when your chances of winning are over 50%, get the least of your money in when your chances of winning are under 50%.
(for poker players I'm obviously talking about cash games and winnertakealls not tournaments :rolleyes:)

This is the n°1 principle of poker, and as you can see the element of randomness is omnipresent. When you go all-in preflop with two aces against 7-2, you will still lose 10% of the times.

Players call this variance : the difference between what you are statistically supposed to win (in this example it's 190% of what you bet) and what you actually get (losing all your money 1 out of 10 times, doubling it 9 out of 10 times).

Variance is why poker has a different world champion every year (luck is a ***** when you have to beat thousands of other players).
But variance is what keeps fish playing, and so it is what feeds the sharks.


Luck is probability taken personally. ---

Let's get back on topic.
[+]If we want the game to be as fair as possible, removing luck is the wiser choice.
[-]Luck is minor in HoN/DotA:) It makes Blacksmith fun to play and doesn't prevent the same teams from winning most of the competitions.

The choice is actually hard.

[mucergo]

Which is really where DOTA is at right now. The same teams win consistently, so artificial luck isn't a big enough factor to allow a random team beat the top teams--which, I think is fine so far.

I agree, except by "backtrack" and "multicast". I like the 5.84c Ogre Magi, his multicast was 100% with huge mana cost of fireblast.

[rickster]

BinAly, do you play games with no rune spawns?

[IKHAN]

I think the game is fine. The only "gamebreaker" that involved luck was spectre's dispersion, and that was fixed. You can strategize to try and exploit luck, which isn't always the most effective way to build a hero, and you can strategize to minimize luck. At the end of the game the best players will win. Which ever team can minimize mistakes at crucial moments, rebound if they do mess up, and optimize every opportunity will end up winning every time.

[akitoes]

BinAly, do you play games with no rune spawns?
Well if someone is against luck in HoN, he would probably be against runes
But you're bringing up a great example here : I absolutely loved runes in my first dota games !

[antigrav]

I think the game is fine. The only "gamebreaker" that involved luck was spectre's dispersion, and that was fixed.

Bring's a tear to my eye. What amazing times those were.

[xahxah]

I agree that DotA (HoN) is a huge success already, but we could improve it. I believe that it is worth the risk.

For example, even in DotA, IceFrog added (because competitive players were asking) a sure-hard-counter to evasion, which is now, Monkey King Bar. it gives users True-Strike, which totally negates evasion-chances, banishing luck from the game.

Just look at the heroes that are being added, less and less of them work on chance-skills. Artificial Luck kills competition, because you can't compete against what you can't control. :(



Icefrog added that because......


A. Heroes with butterfly were near unkillable unless you purely focus nuking that one hero.

B. Evasion item was added

C. MKB was hardly used because there were better and cheaper alternatives.

[akitoes]

Icefrog added that because......


A. Heroes with butterfly were near unkillable unless you purely focus nuking that one hero.

B. Evasion item was added

C. MKB was hardly used because there were better and cheaper alternatives.
Doesn't invalidate the argument:)
Less luck = more competitive

but less fun :confused:

[Rhaegor]

I think that artificial variable luck is very trainable. You get a feel for how your character performs over a period of time and eventually you get a feel for how often those artificial luck type things happen and account for them with your gameplay style.

[mucergo]

The game is fine, but can be better. BinAly is doing his job as beta tester: Start a discussion to improve the game. IKHAN saw about spectre's dispersion ability. That was fixed 'cause some players discussed about in posts like this. I think we can do the same, and make HoN an exelent (not just fine) game.

I sugest to identify some specific balancing problems caused by LUCK and discuss about then.

[Alkara]

We see plenty of competitive games/sports with luck as a factor. Most MMO's and many RTS's have luck involved. HoN and DotA are basically MMORTS's and as such it makes sense to have luck involved. People like the aspect of luck, people cross their fingers hoping their skill will trigger or run in with Blacksmith and hope he gets a 5x Multicast.

We even see luck in professional sports, in hockey there are many times a lucky bounce will happen and the puck will go in the net or the puck will hit the crossbar. Skill often comes from minimizing luck, increasing accuracy and making sure those bad bounces don't happen as often. Thats why in hockey players block shots, less shots means less chance a flukey bounce will end in the net.

If no luck was involved, every fight would be the same everytime. The best players minimize luck and increase the probability of success. If there was no luck then every game would end the same. It would be like playing Tic-Tac-Toe.

[jsat]

Alkara, i would hesitate to equate no luck with simplicity. Chess is a game of very finite variability, but i doubt you call it a triviality. The question here is of exact versus random quantities and i'm hearing that the consensus is explicit random variables does not equate to a lack of skill inherent in the game IF the effect difference is not so large so as to have huge effects on gameplay. By this logic dodging a hit is fine with everyone, 25% miss chance up hill is interesting and strategic, dodging an ultimate is garbage. Ogre/blacksmith is at the edge, with his ult as is it seems to be a large variability in effect with dire consequences, but the way you play around it in most circumstances being the same in team fights as always, just with really large changes in smaller engagements.

[FiNGERS]

Poker has been one of the most competitive games in the past years.
http://www.friday-night-poker.co.uk/mediac/400_0/media/72.jpg

But I halfly (wut) agree with you, despite the fact that "Luck" has been part of video games especially RTS for a long time (Warcraft/Starcraft also had crits, evasion and all that stuff buddy :>)

Related to picture, poker sharks know that luck is their worst enemy as they can lose big money on unlucky pots, but also their best friend as amateurs wouldn't play if there was no luck.
Think about it, why are there so many more poker players than chess players ? :>

Crits, helixes, and other randomness are a big part of what makes the game exciting, thus more accessible to players. Randomness made poker successful, and makes esports successful too.
Actually, losing a hand to misfortune in poker is far more devastating than getting multicast procced !

I have to agree with this guy.

[xahxah]

If there were no variables to a game we might as well play one hero and duplicate heroes. That way things are fair. EVERYONE CAN ONLY PLAY THUNDERBRINGER D:

[BinAly]

If no luck was involved, every fight would be the same everytime. The best players minimize luck and increase the probability of success. If there was no luck then every game would end the same. It would be like playing Tic-Tac-Toe.

Lack of random factors doesn't mean simplicity. Super Smash Brothers Melee has very little random factors (Competitive play doesn't allow for itens), and is a game which is very deep and interesting.

On the other hand, his younger brother, Super Smash Brothers Brawl, isn't as great as Melee, because some random factors were added to the game. In fact, most competitive smashers doesn't like Brawl at all, and it got removed from several championships.
For example, in Brawl, every character has a very little chance (1%) of tripping. That's a very low chance, but it ultimatelly ruined the gameplay. You can't trust your characters anymore, because at any moment, you might trip and fall into the powerful smashes of your opponent.


The question here is of exact versus random quantities and i'm hearing that the consensus is explicit random variables does not equate to a lack of skill inherent in the game IF the effect difference is not so large so as to have huge effects on gameplay. By this logic dodging a hit is fine with everyone, 25% miss chance up hill is interesting and strategic, dodging an ultimate is garbage. Ogre/blacksmith is at the edge, with his ult as is it seems to be a large variability in effect with dire consequences, but the way you play around it in most circumstances being the same in team fights as always, just with really large changes in smaller engagements.

That's what I gathered until now, too. I'm personally against any artificial luck on a game, but thats my opinion. People here showed that there are several situations where luck might help competitive gameplay.

The conscensus seems to be that: As long as Strategy/Training/Skill has the possibility to minimize random factors, they aren't a problem.


We even see luck in professional sports, in hockey there are many times a lucky bounce will happen and the puck will go in the net or the puck will hit the crossbar. Skill often comes from minimizing luck, increasing accuracy and making sure those bad bounces don't happen as often. Thats why in hockey players block shots, less shots means less chance a flukey bounce will end in the net.

This example equates to my example of Devourer throwing a Hook at the FoG, without knowing were the enemy is, and hitting him. That's not artificially generated Luck. This kind of luck, which in my opinion, is healthy, shouldn't be removed from the game.

The Luck I'm against is like this:

In Hockey, now we have a 20% that the puck will explode on impact. No ammount of training will stop that from happening.


Obs: People, let's keep a healthy discussion. My opinion won't change the game, and I want to know why you like the way random factors are applied to HoN. Like I said in my first post, no flame-wars.
I think I respected everyone until now.

[mucergo]

If there were no variables to a game we might as well play one hero and duplicate heroes. That way things are fair. EVERYONE CAN ONLY PLAY THUNDERBRINGER D:

We are not talking about remove luck from game, we are talking about balancing.

[BinAly]

If there were no variables to a game we might as well play one hero and duplicate heroes. That way things are fair. EVERYONE CAN ONLY PLAY THUNDERBRINGER D:

That's not what I'm talking about. Let me try to explain in a better way:

Variables doesn't equate to Random Factors.

Many Heroes: GOOD.
Many Abilities: GOOD.
Many Itens: GOOD.
Random Unbalanced Factors, which can't be controled: BAD.

That's what I'm trying to say. :)

[xahxah]

We are not talking about remove luck from game, we are talking about balancing.


Having only one hero for everyone means it's perfectly balanced.

[obstacle_1]

just remove Blacksmith. Thats all.

[BinAly]

Having only one hero for everyone means it's perfectly balanced.

Xahxah, that's not the point, my friend.

[xahxah]

That's not what I'm talking about. Let me try to explain in a better way:

Variables doesn't equate to Random Factors.

Many Heroes: GOOD.
Many Abilities: GOOD.
Many Itens: GOOD.
Random Unbalanced Factors, which can't be controled: BAD.

That's what I'm trying to say. :)



You cannot control who picks what heroes. Some heroes are inevitably more powerful than others. That is related to balance. By assigning only one hero for everyone, you remove this variable of balance, therefore making it balanced.


Since everyone has the same abilities that means there is no balance issues here either.


Oh, and remove items too, the variability on items can cause imbalances also.




This is exactly what your argument is.

[BinAly]

just remove Blacksmith. Thats all.

That's something that I'm not fully confortable with. We may tweak him, and make him less luck dependant, if it is a problem at all.

[xahxah]

Xahxah, that's not the point, my friend.


No, I'm using your own argument. Because each hero is unique and have different variations, that means there are variables that certainly cannot be controlled. You want to take away some randomness and variation of the game. I did the same thing also.

[Glorify1]

I've already seen people state that abilities like Whirling Blade would be overpowered if they went of every fifth attack for sure. That's a wrong way of thinking.
Right, every fifth might be too strong, but then make it go every 6, or even every 8 attacks. It doesn't matter, it could be balanced without allowing Luck to interfere with competitive gameplay.

Because this can be abused, let something hit you 4 times and then walk up to an enemy and get hit.

[BinAly]

You cannot control who picks what heroes. Some heroes are inevitably more powerful than others. That is related to balance. By assigning only one hero for everyone, you remove this variable of balance, therefore making it balanced.

Yes, you can control. That's why, in competitive play, teams can Ban Heroes.

[xahxah]

If I wanted to play a game that is cut and dry and requires pure skill, I would have chosen another game to play. However, I chose this game to play because...


A. It is exciting and fun
B. The game is predominantly skill and teambased
C. There is a small element of luck that creates variation.



If we wanted a game where luck didn't exist, we would all go play Chess or Kung Fu Fighter for the NES.

[mucergo]

Xahxah,

We have 2 options:

Start a flamewar or try to improve the game. Choose one.

[BinAly]

Because this can be abused, let something hit you 4 times and then walk up to an enemy and get hit.

Well, the 25% behaves the same, even better. I've seen Legionnaires get Bloodlust+Double-Tap early-game, just by being lucky.
Random Factors are, by no means, less abusable than fixed mechanics.

[xahxah]

Yes, you can control. That's why, in competitive play, teams can Ban Heroes.


You are only allowed a limited number of bans, you can only control up to an extent. That still does not remove the element of luck.

[BinAly]

If I wanted to play a game that is cut and dry and requires pure skill, I would have chosen another game to play. However, I chose this game to play because...


A. It is exciting and fun
B. The game is predominantly skill and teambased
C. There is a small element of luck that creates variation.



If we wanted a game where luck didn't exist, we would all go play Chess or Kung Fu Fighter for the NES.

Now that's constructive, thanks. You seem to agree with the majority. Small (emphasis on Small) amounts of luck doesn't break the competitive aspects of the game.

[xahxah]

Xahxah,

We have 2 options:

Star a flamewar or try to improve the game. Choose one.


You're trying to improve the game by taking some of the unique aspects of it out? That's a good one. I have never in my many years of playing DotA (since it was in it's RoC incarnation) have ever seen anyone complain about "random factors". Ever. The only exception to that was when Spectre was around, and that's because that skill was literally broken either way you put it.

[mucergo]

Now that's constructive, thanks. You seem to agree with the majority. Small (emphasis on Small) amounts of luck doesn't break the competitive aspects of the game.

Yes, and we don't want to remove the luck factor, just balance.

[BinAly]

You're trying to improve the game by taking some of the unique aspects of it out? That's a good one. I have never in my many years of playing DotA (since it was in it's RoC incarnation) have ever seen anyone complain about "random factors". Ever. The only exception to that was when Spectre was around, and that's because that skill was literally broken either way you put it.

I've seem several people complain about Ogre Mage/Blacksmith. I've seem several people complain about Void.

You don't need to be upset, man, just make your point.

[xahxah]

Now that's constructive, thanks. You seem to agree with the majority. Small (emphasis on Small) amounts of luck doesn't break the competitive aspects of the game.


Luck exists in every sport. In fact, even a small amount of luck exists in Chess, a game where luck is almost non-existant. Luck is there for a reason, and that's to add a small random factor. It allows for an exciting game. If it didn't exist, the better team/player always wins.



However, in the real world where luck exists, the better team doesn't always win because of this, and that is why watching sports is fun. Because sometimes even the best of teams/players can be beaten.

[xahxah]

I've seem several people complain about Ogre Mage/Blacksmith. I've seem several people complain about Void.

You don't need to be upset, man, just make your point.


Where? Do feel free to show me a link. Last time I checked, most people were complaining that Ogre Mage's fireblast was proccing too often, not the other way around.

[rhodric1]

While you are at it, remove min/max damage and just make heroes do the average damage everytime....

I really hope the developers don't take this thread into consideration.

[ChimEuphoria]

And what happens when two skilled teams go at each other? Individual skills begin to matter. If you don't have the individual skills, you have no chance of getting out of pub matches. Individual skill is a pre-req to even thinking about higher end play and playing with good teams that utilize strong strategies.

Again, i would love to reiterate that HoN is not a fighting game, nor is it a first person shooter.

Games in both of these genres continue to evolve competitively to keep players on their toes, skill levels keep raising ad inifinitum.

HoN on the other hand has a very steep learning curve, but once you get past the curve you plateau, no longer building true "skill".

A good competitive team is one whos players are entirely synergized.

Real skill is not memorizing item builds, or knowing what hero counters what.

Real skill is parrying a full super from chun-li while you have a chip left of health. http://www.youtube.com/watch?v=XeM0rH_4ung

Real skill is bunnyhopping and chainrocket jumping across levels in Quake 2 3 and 4, all the while decimating everything in your path.

The "skill" you mentioned in your post is about as real as skill in WoW.


Skill in HoN can be directly translated to strategy for those past the learning curve. It takes synergy, and creative strategies to be a winning team.



Nice little personal jab you threw in there too, Immature.

[Pzzz10uS]

Here's my thoughts on the matters.

Luck has a peculiar habit of favoring those who don't depend on it
I consider that in most cases 'luck' is just a byproduct of good planning and execution. There are very few truly 'random' things that happen in this game, and their individual contribution will rarely change the course of a game. How often does a single lucky crit change the flow of a game, turning a lose into a win? I've never seen it.

[livmew]

This'll be well and proper long it seems, TLDR is last paragraph.

The only real downside I've seen to luck is the blacksmith, who is either useless, or amazing. A total inconsistent hero that can at any time be turned into a complete zero. A true roll-the-dice, see-what-happens kind of hero. While he's admittably fun, I think he does need some sort of balancing towards a middle ground (IceFrog did this in dota by adding % for 2x/3x/4x fireblast aswell). Being either shite or amazing is rarely fun for the player, nor the enemy.

Other than that, I think luck actually increases the skill of players when used in moderation. The more unrandom a game becomes, the more static the gameplay becomes. You use A, I use B. Luck gets in the way of that by forcing you to immediately re-think your situation. Let's say you're slugging it out with another hero, who lands a lucky crit.

This forces you to immediately work the maths of whether or not you'll live, and take appropiate defensive measures if you won't. I think one of the best ways to measure an individual's skill is how fast he or she can respond to a rapidly changing situation. While you may lose even though you were the better player because of his crit, you also have to exercise your skill in another way than just being quick at maths in order to get away or try to stack the odds via clever spell use and juking.

When this gets out of hand, and you are forced to rely more on luck than your innate abilities, I think the Blacksmith is what is produced. But as far as runes, crits, dodges etc are concerned, it makes the game that much more fun. The reason you may not want to ulti chronos is because, yes, he may dodge it. It enhances the "reactive skill" in forcing people to quickly adapt to situations and also to take risks, which spices up the gameplay and makes it a lot more exciting to watch.

So I guess what I am trying to say, I do not believe the competitive enviroment suffers all that much from luck or chance, and the way dota/HoN has balanced this out is the fact that no single kill will finish the game. If you're ever at a point where one single hero (or one single lucky 5xfireblast proc) determines victory or defeat, chances are you've already screwed up many times before in overfeeding or mismanaging your team, and luck isn't the problem. Luck very very very very rarely if ever wins games, and certainly not tournaments.

[Lyte]

Again, i would love to reiterate that HoN is not a fighting game, nor is it a first person shooter.

Games in both of these genres continue to evolve competitively to keep players on their toes, skill levels keep raising ad inifinitum.

HoN on the other hand has a very steep learning curve, but once you get past the curve you plateau, no longer building true "skill".

A good competitive team is one whos players are entirely synergized.

Real skill is not memorizing item builds, or knowing what hero counters what.

Real skill is parrying a full super from chun-li while you have a chip left of health. http://www.youtube.com/watch?v=XeM0rH_4ung

Real skill is bunnyhopping and chainrocket jumping across levels in Quake 2 3 and 4, all the while decimating everything in your path.

The "skill" you mentioned in your post is about as real as skill in WoW.

Skill in HoN can be directly translated to strategy for those past the learning curve. It takes synergy, and creative strategies to be a winning team.

Nice little personal jab you threw in there too, Immature.

When I was saying "you" I wasn't referring specifically to you.

However, if you think it's so easy to "plateau" in skill in DOTA, you still have a lot to learn, so maybe you are included in the comments broad sweep afterall.

Do you think you could beat any individual player, on any top team, in a single lane with identical heroes? I guarantee you would lose in denies, last-hits, and kills/deaths. I wonder what's happening here, if skill is so easily plateau'd in this game?

[Whitebushid1]

I agree, it's ridiculous that my base damage can range from 49-57. What if I get Luck'd and so and so hero gets away with 7 life!

[ChimEuphoria]

However, if you think it's so easy to "plateau" in skill in DOTA, you still have a lot to learn, so maybe you are included in the comments broad sweep afterall

Do you think you could beat any individual player, on any top team, in a single lane with identical heroes? I guarantee you would lose in denies, last-hits, and kills/deaths. I wonder what's happening here, if skill is so easily plateau'd in this game?

Again with the personal attacks, sheesh. Cool down brah.

I never said stated once that the plateau was easy to reach, I in fact stated that the learning curve was quite steep. Try reading before you post, and refer to my earlier post, where i stated that the only thing requiring real skill in DoTA and HoN is last hitting and denying.

So are you supporting my argument?

20 bucks says I could play a game with somebody of your choosing and not die once.

Would I be skilled then? Would they be bad for not killing me?

[Orcheon]

Same thing as with any other somewhat luck-based game.

In the short run, luck will play a factor. In the long run, luck will even itself out.

And luck should play a factor in the short run of a game. The game(ANY game) would be far less interesting or exciting if the game was completely predictable.

[Lyte]

Again with the personal attacks, sheesh. Cool down brah.

I never said stated once that the plateau was easy to reach, I in fact stated that the learning curve was quite steep. Try reading before you post, and refer to my earlier post, where i stated that the only thing requiring real skill in DoTA and HoN is last hitting and denying.

So are you supporting my argument?

20 bucks says I could play a game with somebody of your choosing and not die once.

Would I be skilled then? Would they be bad for not killing me?

There'd be a lot more good players if last-hitting and denying were the only skills in HoN/DOTA. I'm not trying to add in personal attacks, but your comments are simply based on inexperience and low level of play, so I have no choice but to point out your errors and misguided ideas about the metagame. It's like a brand new basketball player saying, "Well, the only skill in basketball is shooting the ball." Comments like that irritate the players that actually understand the depth to the game.

I guess the only skills in Starcraft and Warcraft III is controlling units right??

[antigrav]

There's an old saying in that the more prepared you are (in this case, the more skilled/practiced/experienced) the luckier you will appear to be.

Orcheon is completely right. Luck evens itself out over time. A single lucky victory or escape will not outweigh the consistent times that someone wins or gets away cleanly because of their skill in the game.

[ChimEuphoria]

There'd be a lot more good players if last-hitting and denying were the only skills in HoN/DOTA. I'm not trying to add in personal attacks, but your comments are simply based on inexperience and low level of play, so I have no choice but to point out your errors and misguided ideas about the metagame. It's like a brand new basketball player saying, "Well, the only skill in basketball is shooting the ball." Comments like that irritate the players that actually understand the depth to the game.

I guess the only skills in Starcraft and Warcraft III is controlling units right??

Im gonna refer to what you said earlier.


Do you think you could beat any individual player, on any top team, in a single lane with identical heroes? I guarantee you would lose in denies, last-hits, and kills/deaths.

According to you, the only thing that would be separating me from the individual player is my number of creep kills and denies. Meaning that the only thing that makes his skill superior to mine is the fact that he can last hit better than I.

Again, supporting what I said about last hitting and denying being the only real skill based element of HoN and dota.

[jsat]

What needs to be again emphasized is HON is a game of finite number of relevant interactions, thus it is not always the case that there are enough encounters to average out. Take for instance an old SK replay when they were just emerging with the global strat with specter, furion, zues, and loda on SF doing his thing. During a game with MYM (i believe) they were being pushed hard with a great series of ganks against a team using lina, sand king, ES, sven, something, very aggressive heroes. In this game, spectre was not doing great with furion having a hell of a time trying to protect her in lane that suffered from ganks and double stun. In the early mid game she dispersed to separate laguna blades that killed sven letting her live, get a double and single kill, and have time to farm up. Sure she probably missed dodging relevent spells later; but this was all she needed to get into the game and finish radiance.

Now of course this is a ridiculous anecdote, but the point still stands that DOTA/Hon has only a limited number of relevant engages before one window or another closes. Finishing radiance against a team like that is the end of the window if they don't already have you really out farmed. This was a HUGE random effect that cannot be accounted for or played around. You must gank specter, you must kill her quickly, because the other team is never far from there carry especially with a global strat, and to do so sven has to get in her face and lina ult. Situations like this are what people see as an issue, where a game is hugely effected by true luck

[Lyte]

Im gonna refer to what you said earlier.

According to you, the only thing that would be separating me from the individual player is my number of creep kills and denies. Meaning that the only thing that makes his skill superior to mine is the fact that he can last hit better than I.

Again, supporting what I said about last hitting and denying being the only real skill based element of HoN and dota.

Actually, that challenge was just to show what is considered DOTA basics, and how difficult it is to really see the depth of the game beyond that. I can guarantee top players will beat you in a single lane, identical heroes, on every aspect of the lane game including "hero damage dealt/taken."

Once you add in the other elements, the game becomes much more complicated and in-depth, and require many more skills to excel at. Most players can't even master the basics--last hitting, denying and harassing.

[GaIactic]

Guaranteed multi cast ogre magi?

No thanks.

^_-;

[ChimEuphoria]

Once you add in the other elements, the game becomes much more complicated and in-depth, and require many more skills to excel at.

Name these "skills".

I attribute DoTA to being a lot like a game of Go. Simple to learn the basics of play and the intricacies of a few old strategies, but to master the game requires intesive thought on the psyche of your enemies.

"Skill" in dota and HoN comes from your ability to out strategize the enemy.

[Lyte]

Name these "skills".

I attribute DoTA to being a lot like a game of Go. Simple to learn the basics of play and the intricacies of a few old strategies, but to master the game requires intesive thought on the psyche of your enemies.

"Skill" in dota and HoN comes from your ability to out strategize the enemy.

Which is still skill. You could argue that the definition of skill varies from player to player, but anything that differentiates two players in the game can be considered a "skill." Whether that's anticipating your opponent's maneuvers to understanding metagame, these are all different skills.

[icevalk]

Ideas like this make me want to cry into my cereal. Knowing when your going to get a bash/crit/whatever will make the game far less exciting.

[ChimEuphoria]

Which is still skill. You could argue that the definition of skill varies from player to player, but anything that differentiates two players in the game can be considered a "skill." Whether that's anticipating your opponent's maneuvers to understanding metagame, these are all different skills.

I now realize that we are wholly agreeing, just on different levels of application.

I do not consider technical (things you do almost mechanically/with no thought) skills to be real skills, you do.

I consider thinking critically, reevaluating the situation and developing on the fly strategies to be skills.

You also consider these as skills, you just pair them up with other mechanical ones. Making your definition of "skill" seem iffy to me.

Nevertheless, we agree more on "skills" than I had initially thought.

[Soulmate]

Luck exists in every sport. In fact, even a small amount of luck exists in Chess, a game where luck is almost non-existant. Luck is there for a reason, and that's to add a small random factor. It allows for an exciting game. If it didn't exist, the better team/player always wins.

And that's not what we want right ?
Why should the stronger team win ?
You can have an exciting game without randomness.
Exemple : 2 top-teams fighting in a close match.



I do not consider technical (things you do almost mechanically/with no thought) skills to be real skills, you do.

&


Real skill is parrying a full super from chun-li while you have a chip left of health. http://www.youtube.com/watch?v=XeM0rH_4ung

I didn't watch the video but from your description I guess It's Daigo versus Wong @ evo on SF.
So you said what daigo did was skilled.
And now you say that mechanical/technical skills aren't.

Do you really think that he didn't do that mechanically ?
Of course he did


And someone said that training was ****, only team-synchronisation was important : Teach me how to be synchronized without practice.

[ChimEuphoria]

I didn't watch the video but from your description I guess It's Daigo versus Wong @ evo on SF.
So you said what daigo did was skilled.
And now you say that mechanical/technical skills aren't.

Do you really think that he didn't do that mechanically ?
Of course he did


During my post about real skill in games i was comparing HoN "skill" to true skill portrayed by gamers in their respective games, hence the Daigo reference.

In my later posts i was adressing ways that "Skill" in HoN could be translated.

And to adress your question, yes i think he did it mechanically he's the beast, but this isnt what is so impressive. I think he baited the super, knowing full well it would leave Wong open to a combod super if fully parried.

Daigo is the true embodiment of videogame "skill".


"lolol we remoev min max dmg okay"
"lol only wan hiro lololol u r too funny"

This thread is getting retarded don't you guys think

This guy is getting trolled dont you guys think?

[Czech0]

Now I don't follow the Dota leagues. But I'm willing to guessing that there are a handful of teams/players that always finish at the top. Heck, there might even be one team/player that is almost always at the top.

This isn't because of Luck.

They are not consistently doing this because they land more whirling blades than everyone else or what ever. Its actually kinds of ridiculous to say that someone is the best at dota because they land the most whirling blades. lol.

With a little research -

"It is no surprise that Vigoss has been called the best DotA player of all time. He has been invaluable in winning Virtus.Pro back-to-back MYM Pride tournament victories and much more. He has the uncanny ability to roam around the map killing everything in sight and yet end with the highest level. Even when he has a rough start, it's no surprise to see him at the top of the scoreboard. In nearly every game, Vigoss is seen leading the ganks which completely shut down the other team. Another amazing aspect is that he can play any hero flawlessly. Whether it's hitting near perfect arrows on PotM or his reckless Bounty Hunter, or even his Witch Doctor, one thing is clear: Vigoss is here to stay and to continue dominating the competition."

[_Archangel_]

Right, those may result in "fun" moments, but and when the inverse happens? When you lose games because that 20% didn't really worked 1 time out of 4?

Math fail detected.

On a serious note, I don't believe that the competitive scene is affected negatively by this significantly. I strongly believe that if one team is better than the other, they will win regardless of luck unless one team gets really lucky with Blacksmith (Who is flawed in this respect).

[evotech]

Meh, situations where you know exactly what will happen is boring, a little luck adds spice.

Luck isnt to get away with 20 HP, its just good. Luck is to get 4X multicast 20 times in a row which should be removed

Crit chance is a chance, you have to calculate luck / bad luck into your strategies.

[Goon]

I am in favor of removing luck from the game.

entire games have died because of luck.

[Infas]

/facepalm
This thread fail. Seriously.
Crits, Evade, Miss, Bash...
The entire game is based on %.

[Glorify1]

Well, the 25% behaves the same, even better. I've seen Legionnaires get Bloodlust+Double-Tap early-game, just by being lucky.
Random Factors are, by no means, less abusable than fixed mechanics.

You have no control when random factors go off, changing it to go off when you want it to is definately not a turn for the best.

[StRiDaH]

Similarly, Ogre Mage/Blacksmith has infuriated players on both sides for years.

Oh god, such truth, such truth.

[zzSleeper]

Consistency is certainly the primary factor in a successful competitive game, but regular small amounts of uniform randomness are not a threat to this (in fact, it adds variety). The problem is when single random events are allowed to be so strong that they alone can decides the outcome of the game. Off the top of my head, multicast is really the only thing that could potentially qualify under this category.

In WC3, Blizzard reduces randomness for probabilistic abilities (crits/bash/evasion) by artificially localizing the distribution graph based on the percentage. What this means is that abilities must trigger after a calculated number of failures-to-trigger based on the trigger percentage. For example, in WC3, if you have a 50% chance to crit you will never see 6 consecutive hits without critting (the exact formula was discussed on the dota mechanics forum some time ago). The overall probability of triggering remains the same, but there is less variance. For the sake of the competitive play and general player perception, I would strongly support such a feature be implemented in HoN for all probabilistic abilities.

Overall though, it takes enough attacks to kill a hero and attacks happen frequently enough that high-frequency/low-effect abilities like critical hits and evasion do not cause sufficient statistical abnormalities to justify their removal. The same goes for the base damage range on heroes. What constitutes high-frequency is up for debate of course, and stuff like Mortred's crit may be near the border in terms of frequency to effect magnitude.

The point is that low variance randomness does not preclude skill - and I would argue that the game is more interesting because of it.

[Kry1]

DotA/HoN doesn't have a luck problem.

[Glorify1]

Multicast does not decide a game.

[BinAly]

In WC3, Blizzard reduces randomness for probabilistic abilities (crits/bash/evasion) by artificially localizing the distribution graph based on the percentage. What this means is that abilities must trigger after a calculated number of failures-to-trigger based on the trigger percentage. For example, in WC3, if you have a 50% chance to crit you will never see 6 consecutive hits without critting (the exact formula was discussed on the dota mechanics forum some time ago). The overall probability of triggering remains the same, but there is less variance. For the sake of the competitive play and general player perception, I would strongly support such a feature be implemented in HoN for all probabilistic abilities.

This is certainly a very elegant sollution that won't even be noticed by the players. In fact, that's not entirelly luck anymore, since the code works to ensure that the probability is respected.

[SolidStroke]

Consistency is certainly the primary factor in a successful competitive game, but regular small amounts of uniform randomness are not a threat to this (in fact, it adds variety). The problem is when single random events are allowed to be so strong that they alone can decides the outcome of the game. Off the top of my head, multicast is really the only thing that could potentially qualify under this category.

In WC3, Blizzard reduces randomness for probabilistic abilities (crits/bash/evasion) by artificially localizing the distribution graph based on the percentage. What this means is that abilities must trigger after a calculated number of failures-to-trigger based on the trigger percentage. For example, in WC3, if you have a 50% chance to crit you will never see 6 consecutive hits without critting (the exact formula was discussed on the dota mechanics forum some time ago). The overall probability of triggering remains the same, but there is less variance. For the sake of the competitive play and general player perception, I would strongly support such a feature be implemented in HoN for all probabilistic abilities.

Overall though, it takes enough attacks to kill a hero and attacks happen frequently enough that high-frequency/low-effect abilities like critical hits and evasion do not cause sufficient statistical abnormalities to justify their removal. The same goes for the base damage range on heroes. What constitutes high-frequency is up for debate of course, and stuff like Mortred's crit may be near the border in terms of frequency to effect magnitude.

The point is that low variance randomness does not preclude skill - and I would argue that the game is more interesting because of it.

For what it's worth, I fully endorse this solution.

[_Archangel_]

Multicast does not decide a game.

It does when you get twenty of them in a row. If you are lucky enough to pretty much one-shot their support hero every teamfight your team will likely win

[Lyte]

It does when you get twenty of them in a row. If you are lucky enough to pretty much one-shot their support hero every teamfight your team will likely win

Please, show me the # of games where a string of 20 multicasts has singlehandedly decided a game.

0 exist in high-end play, that's for sure. But I'll let you search pub replays.

[Avan1]

What needs to be again emphasized is HON is a game of finite number of relevant interactions, thus it is not always the case that there are enough encounters to average out. Take for instance an old SK replay when they were just emerging with the global strat with specter, furion, zues, and loda on SF doing his thing. During a game with MYM (i believe) they were being pushed hard with a great series of ganks against a team using lina, sand king, ES, sven, something, very aggressive heroes. In this game, spectre was not doing great with furion having a hell of a time trying to protect her in lane that suffered from ganks and double stun. In the early mid game she dispersed to separate laguna blades that killed sven letting her live, get a double and single kill, and have time to farm up. Sure she probably missed dodging relevent spells later; but this was all she needed to get into the game and finish radiance.

Now of course this is a ridiculous anecdote, but the point still stands that DOTA/Hon has only a limited number of relevant engages before one window or another closes. Finishing radiance against a team like that is the end of the window if they don't already have you really out farmed. This was a HUGE random effect that cannot be accounted for or played around. You must gank specter, you must kill her quickly, because the other team is never far from there carry especially with a global strat, and to do so sven has to get in her face and lina ult. Situations like this are what people see as an issue, where a game is hugely effected by true luck

kinda off point but i once saw a specter disperse techie's suicide for a quintupple kill. was just inhouse not pro, but they lost the entire game due to that one mishap

fun fact: the specter was red health and running

[GaIactic]

Multicast does not decide a game.

It COULD, if you knew every 4th cast was gonna be a guaranteed one.

^_-;

[Karumph]

I feel like what i'm about to say is very important so actually read this.

Luck should be there, like everyone else said if it isn't extreme, obvoiusly.

READ ALL:

After reading everyone elses arguments about it I think that no matter what it should stay in the game, but it should have

DEMINISHING RETURNS <---- :O

if you had something OP like 90% chance of dodge, it should go down a certain amount every time, so that you couldn't just get lucky and have 10 in a row.

Or for example if you only had 20% it could go down 5% every time, ensuring that you couldn't get more than 4 dodges in a row, even on luck. It should ONLY APPLY if you do dodge though.

WoW did a great job of implementing deminishing returns on stun locking mace-specced rogues. They would have 5% chance to stun from the talent, and then the other %'s added to it. For a while it was luck, and sometimes they would just get like 10 stuns in a row and u could die from it. As soon as they put in deminishing returns, stun lock could no longer happen. After 20 seconds, the deminishing returns would wear off and reset the counter at its max, but It would keep it from being too OP and at the same time allow their spec to still be useful.

[ubercrombie]

Epic math fail in this entire thread (and in other threads talking about crits, evasion, blacksmith multicast etc.).

Saying that because a critical strike has 20% chance to trigger DOES NOT, in and of itself, imply that it adds "luck" to the game. Most of these events (a crit triggering or an evasion or a bash) occur a thousand times every game, so things average out.

Think about it this way: Swiftblade with level 4 way of the sword has a 36% chance to do a 2.0 X crit on every attack. But in an hour long game, lets say he does 1000 attacks (conservative, assuming a decent player who doesn't spend half his time in the spawn queue). When you look at it in this scope, all the critical strike skill does is increase his AVERAGE damage per hit by (.36*2+.64) = 1.36. So what that crit skill really does is increase swiftblade's base damage by 36%.

The same analysis works for evasion (reduces the damage a hero takes) and for blacksmith's multicast (it increases the average damage of each fireblast). When you're talking about things that occur a hundred times a game, the average effect becomes more important than one instance of "luck."

[BLUEPOWERVAN]

It really doesn't average out at all, and in fact matters very much.

Say swiftblade is attacking a jereziah that is healing himself. The Jereziah has 3 full hits left to kill him. The swiftblade has time to perform 2 attacks. With "average" damage +36%, he has a 0% chance of killing jereziah pre-heal. With 36% chance of double damage, he kills Jereziah over 59% of the time.

If you think a statistical distribution is equivalent to its mean, you are #1) very pompous for considering yourself better at math than everyone posting before you, and #2) not very good at math.

The chance for bursts of high or low performance can have dramatic situational importance... situational considerations are what dota and hon are all about.

[ubercrombie]

The point is that those situations even out. When you're talking about a balance/gameplay type issue, you can't pick out one very very specific scenario and say "oh look this proves my point." This thread is talking about how much of a "luck" factor %chance abilities add to the game and my point (which you completely missed) is that you have to look at ALL the situations in aggregate. And btw I didn't come close to saying the distribution is equivalent to its mean. I chose a large trial number, 1000, specifically so that we would be guaranteed that the number of crits would approach the expected value. LLN ftw.

[Trasen]

Love you!

No, but removing the luck factor would make the game incredibly skill based and very much more fun.

Getting a critical hit in.. well.. boxing, is a situational luck of a fighter. When the opponent looses his grip and does something stupid/ unexpected, this is where a match can go wrong, for both fighter´s, if you are good enough or your opponent is bad enough you will get that extra damage you need to finish him of.

It´s a very very veeeery valid point, and should be overlooked by the developers!

[uranows]

In WC3, Blizzard reduces randomness for probabilistic abilities (crits/bash/evasion) by artificially localizing the distribution graph based on the percentage.

Could you give us a proof of that?
Don't get me wrong, after many years playing WC3 (and reading the Dota forum) I probably agree with it.
It's just because I couldn't find anything on google to check it.

S2 members should reply this thread telling if something like that were implemented before any crazy and lifeless HoN player like me start to test the randomness of lucky events to see if it tends to average.

And yes, I agree that despite the Ogre multicast, others lucky-based events don't decide an entire match.

[Guile]

being "lucky" is just knowing how to take advantage of opportunitys

[purehybrid]

People seem to misunderstand the effect 'chance' has on strategies.

When there are 0 luck/chance based effects in the game, a single encounter can start to feel very scripted. If you know (from experience) exactly how much damage, how much stun, how much mitigation etc is going to take place from each character in a situation, then you will know your strategy before you start.

Add in a couple of random elements to that fight, ie critical-strike... and people are then required to not only have an initial strategy, but think on the fly, and carry through contingencies as well.

Chance (when kept to a reasonable level) does not necessarily impact the skill requirement itself, but rather, where that requirement is placed. In this case, it shifts the focus more toward reaction time and adaptation, whereas with no random elements, the skill requirement would be focussed much more heavily on pre-fight strategy.

[ihatewalls]

I have to disagree that the "luck" factor is bad.

I don't know if anyone posted this already, but consider the example of aiming in FPS games. Many highly competitive games (counter-strike) already have the luck factor as a integral part of the game. That is, where you aim is not necessarily where you will hit, especially if you are moving.

I feel like crit strike chance is more or less the same idea. Although it's not controllable, weighing chances/risks is simply another part of the game.

[campster123]

I think luck adds a little flavor to competitive and non-competitive games (critical hits, etc.). But, luck can´t act like a decision point in the game. Players must have a chance to minimize the effect of luck. For example; Poker, Magic: The Gathering. These two games have luck involved, but the same players are world champions every year. Why?
Its simple, with strategy they can minimize the luck factor of the game.
Resuming: Luck is important for fun, but strategy and training must minimize the luck factor.


This (although argued terribly but it has a lot to do with it)

DOTA would be a horribly boring game with "Luck" eliminated. It is also this luck that early can entice weaker players to keep playing. This sort of effect is more pronounced in poker. Any player can win at anytime in poker - this encourages new players to keep playing or to try out the game - and yet there is enough skill factor to create a balanced competitive environment.

DOTA/HoN have what I feel is a nice balance between luck and skill. The game (and most games) would just would not be as enticing without it.

[Karmashock]

I'd just like to come out in support of reducing luck in the game. I think all or most of the luck like abilities/items should be changed to something like the puppet master's third skill that does area damage every 5th attack. Just have them trigger automatically at a rate that is similar to their probability level. Lego's ability would trigger on about every 6th hit, wingbow would evade something like every 3rd attack, etc.

no need to change blacksmith though. He's kind of the exception. IF you don't want the luck, then don't pick him.


Elsewhere luck is just too pervasive.

[Hugh_Jorgen]

I think the OP has a good point. *


It's all down to degree though. *Having a range of 45-57 damage is fine, since it's a 20% range and you'll be hitting a LOT over the course of a game, so it'll average out. *The people saying that taking all the luck out of the game would make it boring are missing the point. *Showing that the extreme end of the argument would be bad doesn't help. *I could easily argue that on the other extreme end, having a character with a 1% chance per game of winning the match automatically, would suck. *So would a character with a 1% chance of hitting for 20x damage.


When Blacksmith gets lucky and gets multiple quad casts, you can't plan for it, no matter how good you are, or how bad the other player is. *You can't go into a fight thinking, if he gets 3 quad casts in a row I'll do X. *When that happens the decisions you've made don't matter any more. It's not fun, you can't learn from the experience, it just sucks. *What makes a game fun IMO is making decisions and feeling that those decisions affected the outcome of the game, for better or worse. *Excessive randomness reduces the impact of your good, and bad, decisions. *You might as well play snakes and ladders.


I would argue that a lot of randomness discourages players from thinking at all, since it won't help, and that's no fun at all. *I'm all for limiting the randomness in HoN. *Not completely eliminating, but cutting back the worst of it.

[Hugh_Jorgen]

Oh, and about poker. *How many people would play poker if money wasn't involved? It's not that it's such an amazing game. *It's gambling that's less random than dice. *That's why people play it. It's the comparative lack of randomness that makes it popular.

[HonStinks]

At the very least the evasion on Chronos and Blacksmiths ult should be remade into something that is less extreme, there's so many times when I've evaded a ult from Pyromancer that would've otherwise killed me and few times it lead to us being able to keep up the attack and destroy raxes.

Blacksmiths ult is pretty dumb although it's less ridiculous compared to the dota counterpart because the stun time doesn't extend with proc, unlike in dota. Just change the buff/stun parts of the ult to something like +0.5 sec stun per level and +4 seconds duration on the buff per level. The AoE slow/burn already improves nicely.

[CruS]

I have some opinions in this matter..

1. Luck is needed since it will induce reactive gameplay.
You have to play differently and you can't calculate everything in a game when there are some random factors to it.

Example: If you know you will crit every 5th hit with your scout backstab. You wonder off and hit a creep 4 times, then vanish and move straight for the target hero. Knowing you will crit him for 50% hp and you will hit your ulti for 50% hp you can just fire them off instantly after eachother without any thought process at all.

2. Luck is removing imbalance
Games without any luck factors needs to be 100% perfectly balanced which will never happen with DotA because there are too many factors involved.
Remove the luck and try to balance the game


3. Luck evens out in the end.
As mentioned, poker is a game of luck. But play enough poker on the same stakes / limits and the luck % of game quickly moves towards 0.
In some sports with luck involved, such as pool (where the first strike can decide the whole game) you simply play many games to even out the luck.
In serious competative gaming, to even out the so needed luck you simple make the game BO3, BO5, BO7..

my 2copper

[Karmashock]

I have some opinions in this matter..

1. Luck is needed since it will induce reactive gameplay.
You have to play differently and you can't calculate everything in a game when there are some random factors to it.

Example: If you know you will crit every 5th hit with your scout backstab. You wonder off and hit a creep 4 times, then vanish and move straight for the target hero. Knowing you will crit him for 50% hp and you will hit your ulti for 50% hp you can just fire them off instantly after eachother without any thought process at all.

2. Luck is removing imbalance
Games without any luck factors needs to be 100% perfectly balanced which will never happen with DotA because there are too many factors involved.
Remove the luck and try to balance the game


3. Luck evens out in the end.
As mentioned, poker is a game of luck. But play enough poker on the same stakes / limits and the luck % of game quickly moves towards 0.
In some sports with luck involved, such as pool (where the first strike can decide the whole game) you simply play many games to even out the luck.
In serious competative gaming, to even out the so needed luck you simple make the game BO3, BO5, BO7..

my 2copper
while you make some good points there should be a way to have percentage based activations that are more reliable. I've been lego, been attacked pretty savagely, and not had the blade activate even once. I was just unlucky here but the system needs to be made more predictable. Maybe keep some level of randomness in it but it should have base levels that it forces activation if it hasn't happened after so many hits or cut offs if it's happened repeatedly.


For example, if something is supposed to happen 1/6 times then after it's happened 10 times in a row maybe stop it for a second. Further, if it hasn't happened after 12 times then activate regardless of the roll.

[thugg_life]

I really don't get the poker analogy. The sort of luck that the OP is talking about, the Chronos/Blacksmith/Legionnaire type abilities, would be more like seeing a full house in your hand in poker, but having one card have a fixed chance of becoming another card when you go to lay it down.

The fact is, you never know when you have a 'full house' with these heroes or not. In poker, once you have drawn the cards you know what they are and they will never change. Luck stops influencing your decision making at that point. In HoN, all of your interactions will be influenced by this type of random luck.

[DeimosTA]

percentage % do exist in basket ball.

the percentage % for a player to shoot a 3 point ball is never 100%

[Karmashock]

percentage % do exist in basket ball.

the percentage % for a player to shoot a 3 point ball is never 100%
Sure, but it isn't a random roll either. No one is rolling dice. If he misses the first shot he tends to pay more attention for the second shot. You're not going to have a good player miss 10 shots in a row. In HoN these are truly random percentage based rolls. Where it is possible to NEVER make a shot EVER. Now, that's unlikely but it is possible. That is not possible for a good player in basketball or any other sport for that matter.


There should be minimum and maximum activations on the luck system so that it does not for whatever reason deviate radically from it's desired activation percentage.


Perhaps have a scaling system that increases the likelihood of an activation every time it doesn't happen and deceases the likelihood every time it doesn't happen. Set it up so on average it happens at the base percentage but this way system activates at a more predictable rate.