Attrition based warfare with superior farm, superior carry, (hard skill function/mathematical superiority), and defensive play will usually beat a flexible, gank-oriented team.
I disagree with the above poster on the grounds that he has essentially posted a truism.
You are basically saying that a better team will beat a worse team. What it comes down to is how both teams shape the battle so that their strengths are brought to bear more than the other team's strengths.
An attritionist team like you describe has two very specific strengths, both of which have to survive in order to win the game. They have to defend and they have to outfarm in order to outcarry.
A rounded team has more strengths, though they are less pronounced. The first step is for the maneuvre team to identify the requirements for enemy strengths to function and break those down into the vulnerabilities for each one. From there you identify which ones your strengths are most effective at destroying.
Then you use your strengths to target the prerequisites for his strengths. For example he likely needs at least 4 heroes apart from the carry to defend. If you are able to isolate and gank 1 or 2 of them, you can probably then overmatch his defence. The game is then whether your team can isolate and gank in order to push, or whether he is able to avoid this, countergank etc etc etc.
As he needs to farm, you assess that one of his heroes will always have to be alone in order to do this effectively. One of the requirements for the farm to be effective is that he doesn't die while farming, causing him to lose gold and farm time in relation to enemy carries. The vulnerabilities of this are the wards he is using to see you coming, the reaction gap for him to start running before decisive engagement, escape abilities and the routes of withdrawal. You counter the wards using unexpected routes or counterwards, you counter the reaction gap with blink daggers, invisibility or very long range abilities, you counter the escape abilities and withdrawal routes by disabling him so he can't effectively use them. Meanwhile you exploit this lack of farm by pursuing your own design for battle with one of your semicarries, or pushing action when your core is done and his isn't.
At the end of the day if attrition strats based around stall and farm usually beat flexible strats it's because the players running the flexible strat aren't levering their strengths well enough, either through poor tactics, or poor play. Your argument isn't so much that attrition beats maneuvre, it's that good drafts beat bad drafts or that good players beat bad players. It has very little to do with maneuvre vs attrition or flexibility vs specialisation.
First of all very nice post, i've been playing HoN since beta beginning and Dota 5 years before that, and I never really thought about seperating my actions in this way. However I do believe that JeHaze has a point. A decent setup for HoN usually contains:
Depending on what psr bracket youre playing on. So every team will have more or less similar resources, but what you choose to do with them, is often battlechanging.
Pzkw you talked about manourvering warfare, and what Jehaze wrote, I see as exactly that. To have a roaming supporthero gank a lane at lv 3, even if just once, can break the lane completely. Furthermore the game is based of different phases which, if broken generally gives you a strong game advantage. And isn't this a fundamental OODA predicting people to choose a 2-1-2 strategy and sticking to a 8-10min laning phase in the beginning of the game. Prior to a choice of interaction with your adversaries you can also
choose to break them purely out of prediction. And im not thinking about taking them due to attrition or skill level, but simply predicting their subconcious choice of tactics and then using your ressources more effectively.
Last edited by GorZul; 12-02-2010 at 11:46 AM.
Understanding this concept and how to apply it or break it is probably one of the last hurdles a [mid] level team/player has to jump before becoming [high] skilled. Great observation and portrayal.Everyone has played a game where every single thing they tried failed. Ganks ran into counterganks with more heroes. Other ganks arrived to find no one there to kill anymore. Pushes got countered before they even got off the ground. Carries got ganked all the time. Every time you tried to do something the enemy had already countered it. The enemy was inside you OODA loop - his decision cycle was so far ahead of your's that he was effectively predicting your decisions before you were making them and shaping his courses of action to destroy your's before you had met the prerequisites for them to function. Your decisions based on Observations and Orientations that were out of touch with what the enemy was doing and were resulting in bad Decisions and Actions which didn't effectively cater to what he was doing now (as opposed to when you made the decision).
Ignore the user who posted the topic he basically, has to much time on his hands and wants to promote his bullshit. He has hardly any experince in Dota and HoN itself.
YOU GAIN BETTER AT DECISIONS THROUGH EXPERIENCE.
Can't get it better till you have experienced it, this is true for academics and sports.
Cut the obamanomic bullshit please:Nullstone:
if they fall for your simple 3rd world mind games you need to get out of the scrub games
I love the original post. Very informative and thought provoking. I appreciate the different element to looking at the tactical aspect of HoN.
I was curious as to your opinion on the reflection of empirical data post-game. I don't mean the simple stats that HoN spits out after the game. I mean the theory crafting, mathematical models, and real numbers generated from a game. Things like, time stamps of completed items, damage output in certain ranges because of certain items, effective mobility (leaps/blinks/movement speed), the effectiveness of kills vs farm, etc.
Brilliantly written, a great read, and generally explained to me what my mind was instinctively going through during gaming.
I propose that this thread be stickied.
Can you include HOW to get better at the OODA process? I'm doing it slowly by test and fail strategy (trying to act quicker based on instincts instead of thinking things out) and it's kind of tiring.
Thank you for taking the time to write up on this. Every player becomes aware of the threat that a confident roaming hero presents and every player knows the feeling in a team fight, when your team's lack of proper planning or insecurity with their planning leads to fights with low commitment. I would say in good games these situations ONLY arise because the other team is "thinking ahead" of you, something that is well explained with the OODA cycle.
In lower rated games and non-competitive play, the team's grasp of mechanics and picking a decent setup seems more important to me. It is true that once you have properly learned the game mechanics, there is only a small margin of mastery...but until you get there, learning timing, effects, cooldowns, animations etc. etc. is more important.
Fight or flight. Regarding decision, it is extremely true. People who are indecisive and are turtles regarding reaction time are usually thoroughly flamed by teammates. Timing your fights and knowing when to back off is also crucial to ganking/teamfights.
You also have to be heartless in this game sometimes. I've noticed being a selfish player can be beneficial. You shouldn't try to save those who aren't worth saving. Like when theyre sure to be dead or it willl result in you dying.
This was such an interesting post that decided to add my two cents. I would like to add to this post and talk more about our subconscious decisions and how we can be better at deciding.
I would argue that most ingame decisions are done subconsciously. The game is so fast paced, that you rarely have time to make fully rational decisions. You might have time to rationalise some part of the decision, but rarely the whole decision.
For example, using Bombardier ulti on people that are in fog of war. You can quickly decide that you can kill someone who enters fog with your ulti. But there is absolutely no way to rationally calculate where the hero will be. What you do is you estimate the position. This estimation is based on your accumulated game experience and is made by your subconcsiousness.
People make decisions using heuristics. (http://en.wikipedia.org/wiki/Heuristic) Imagine the calculations needed for solving the problem in the example. All the possible paths the hero can take, with stopping without stopping, understanding the intent of the other player, there are multiple variables that need to be considered.
The heuristics build into our brain allow us to make fast decisions about extremely complex problems. The pitfall in this is that we easily do these decisions with faulty logic.
To get better at making these snap decisions, players have to train their subconcsiousness. This is done through repeating the scenarios multiple times and analysing the game play in each scenarion. None of these strategies to get better work very well on their own. If you just play the game, and thus repeat the scenarions, you easily repeat your mistakes. If you dont reflect on your decisions, you are very unlikely to know when you actually played well and when you made poor decisions. If you over analyse your game, you are bound to rationalise poor decisions.
Thus to get better you analyse what was done and try different approaches to the problem.
Just a small personal remark in the end. I feel that the major reason why many players do not learn from the game enough is that they blame other people too readily. If every poor decision was made by other players, how can you learn from the mistakes made?
I hope this post was understandable. Wrote it after waking up and the idea needs more conscious work. I hope you get something out of it.
Firstly, never stop thinking. Your instincts should be able to provide you with reactions to situations faster than you can consciously think them up, but your instincts generally can't compete with rational analysis of larger scale issues like how your team should handle the gank phase of a game, where ganks need to be targetted, where a team should push and when...
You should never be consciously attempting rational analysis in any kind of stressful situation where decisions need to be made and actioned within a few seconds since there's no time for reasonable analysis. You will make a lot of mistakes either way.
The trick is to learn from them by watching your own replays or getting a mentor to watch your replays. Alternatively, which is very difficult to do in HoN, find a few much better players to play with. They won't hesitate to tell you how you should have handled a specific situation. A game like HoN is formed from thousands of combinations situations happening concurrently in different places over the course of the game, those situations once you've played a lot of games start to reveal stark similarities even under very careful analysis. There is only one optimum way to deal with two advancing stunners with higher MS, no other missing heroes, one on low health as a lowish level DSham with a TDL on high HP, mid mana and a creep wave arriving in about 10 seconds and nearest allied tower 1100 units away, but you won't identify that and get it right every time until you have it confirmed for you as completely right at least once.
As you start to get more experienced and exposed you learn more and more of these situations optimally and that makes your OODA loop much quicker and more accurate since you're just templating everything into optimum templates. The problem comes with you start playing with players who you are relying on to fit into your plans and understand the same things as you. That's the most frustrating thing for better players about playing with worse players - you die to their lack of simple understanding so much (If you just stunned after I went in, we could have killed them both - common enough situation).
Eventually you start to distill all of these individual situations into generalisations. For example, you always pace wide as QoP to avoid the initating stun. The reasons for are wrapped up in layers of metagaming, dozens of individual situations and assumptions, the simple reason (that you have low base HP) is only one small part of the overall reason. Another is that you always leap immediately after an Elune's Arrow if there's even a decent chance it will hit (if in an offensive situation). Again, the simple reason belies dozens of generalisations and considerations that underpin this rule of thumb. Justin (I think) used to have a little tip of the day masterclass segment on dota-allstars.com where he would share his generalisations as a high level player, but really you should form your own so you actually have your own style of play and not just the level of play you want. And again, you do that with focussed and guided reflection as well as fair and rational analysis.
Also, always. ALWAYS take complete responsibility (internally) for every single thing that ever goes wrong even if it clearly wasn't your fault. That way you start to think of strategies to deal with that and similar situations again because you will always have teammates that don't do quite how well or what you wanted them to, that's a fact of the game even in organised play. You as a player need to have strategies to deal with situations that involve teammates being part of the problem as well (farming harder usually isn't one of the good potential candidates, though it's an easy plan to form)...
Best of luck. Sorry we disagree about Vindicator's balance.
A very important point that isn't learned until a lot of experience is forcing people to go in a particular way. If you stun with an AE stun, you stun where someone is going to be when it lands. At some point, good people know this, so you have to stun where they are; since they will fake movement in one direction, and either stop to engage you, or go another direction. If you find these people, you need to remember that you have more time to kill them, and start with an auto attack/2 before stunning.
Same goes for Devourer. If an enemy enters fog of war, they normally have at least two choices. Stop, Left, Right. If right is back home, I will hook right, and miss if they stop or left. If they stop or left, they die to rot. Whereas, if someone has a portal key and is good at using it, I might hook stop/left assuming that they will juke stop/left and portal key away. This is what makes games against people so much more enjoyable than games against AI.
Additionally, PzKw makes a lot of good statements in his previous post.
i think my brain exploded, anyway ill start reading now
In this world gone mad, we won't spank the EM players, the EM players will spank us.
you are right: strategy has to encompass not only the surface facts and observations, but also the deficiencies of your team. for example, if you're initiator and realize that your team isn't joining the fight appropriately, leaving you to die a horrible permastun death, you should perhaps look internally first and question whether
a) your timing is off
b) your not communicating your strategy effectively
c) you're being greedy
and the worst realization...
d) that you're team is simply not capable of working together.
if you end up at D too often, then there's always
I always did like to call Dota a "decision engine" game. nice post.
Going aggressive while your mid miss (but isnt near you) or any other circumstances can give you some extra kills/LHs/free hits in harass battle/whatever.
You need to know the ennemy to predict his actions.
A bad player may just engage you if you fake a gank situation, while a good one may retreat to his tower.
Very interesting, I went and did some more reading on it.
My only criticism would be your "draft".
A faster OODA loop would mean that you should have no draft in mind.
In organized play, your draft should be based on what you know about the other team, therefore leading you to have a general plan, however, what you pick would be improvised on the day.
In a game where you are 5 friends against 5 randoms, you have already, at an organisational level, a far higher OODA. You could likely pick any type of strat and have a high level of success.
However, in an environment where you are meeting "equals" sticking to a strat is not advisable under OODA. You then become predicatable and liable to being unsuccessful. You need to stay unpredictable in the picking phase.
The practicality of this theory to how to be a better player, is likely difficult. It is easy to say you need to get inside your opponents OODA, however, how you do this is really a culmination of natural talent for learning, visio-spatial awareness/co-ordination and how much you really want to learn.
Making people rage for over 400 games
PzKw.... will you marry me?
It has been a long time since I have seen such a well prepared post, which actually had some of the best context of all time as far as I am concerned.
Props to you, many props to you
It is critical to remember that there aren't two sets of decision making cycles at work, there are ten in a 5v5.
One of my very favorite things to shout as one of my teammates dies a screen-length away is, "Everyone, Leave him to Die. HE CHOSE HIS FATE. Let us honor his memory by living well."