Climbing The Ladder Part 1

Intro

PART 1: Impressions of the 3k MMR bracket
- 1. The Smurf
- 2. The Rules
- 3. Disclaimers
- 4. The Stats

General impressions of the 3k MMR bracket
- 5. Player have bad mechanics
- 6. Players are toxic
- 7. Support are significantly worse than cores

Models of learning and their implications in the 3k bracket
- 8. The Dreyfus model of skill acquisition
- 9. Reliances on routines/shifting focus to people

Implementing the Dreyfus model in climbing
-10. Self-Evaluation
-11. Evaluating others
-12. Winning the game before you

In this guide, I will discuss my impressions of the notorious 3K MMR bracket- a matchmaking bracket containing players of above-average skill that proves extremely challenging for many to break out of. Herein you will find a breakdown of the skillsets and trends in playstyle of players across a range of MMRs in the 3K bracket, and the differences that separate a 3K MMR player from a 4K MMR player. I will also outline effective playstyles in each sub-bracket within the 3K MMR range, which you can use to more effectively climb your MMR through to 4000 MMR and beyond.

OVERVIEW

The intent of this guide is to cast light on the shadow land of the 3K bracket. As I mentioned in my last guide “Mindset and You,” matchmaking in Dota 2 is a source of great frustration and angst for many players. I believe that many players (especially in the 3K bracket), although passionate about the game and thirsty to improve, lack the proper direction to do so, and are distracted by the toxic atmosphere around them. By breaking down the 3K bracket into manageable ranges, and explaining the trends and effective play styles in each range,the most important areas to focus on in order to improve become very clear.

This guide consists of four parts:

This first release will contain Part 1. We will release Part 2 and then Part 3-4 at a later date

In Part 1, I will give a general overview of my impressions of the players, play styles and skill level of the 3K MMR bracket. I will explain the relevance of these trends to various models of skill acquisition in order to both ease frustration concerning the players that are commonly encountered in the 3K bracket, and to direct 3K players to better self analyse their own position as a learner.

Part 2 contains a breakdown of the major sub-brackets within the 3K MMR range, and will explain (with reference to part 1) the different skill levels, play styles, and effective strategies to play with and against players in each sub-bracket. I will explain the common mistakes players in each sub-bracket make, how to exploit them, and how to avoid making such mistakes yourself.

In part 3, I will explain the importance of being an effective leader, and how to increase your impact in games by effectively leading your team. With reference to parts 1 and 2, I will explain how simple tactics and practice in being an effective leader can easily rectify many common problems in teamwork which arise in the 3K bracket.

In part 4, I will outline a three step process for analysing your games-both pre-game and postgame-in order to establish clear direction and goals when you play, and develop good learning habits for improving as a player.

After the release of “Mindset and You” I received a variety of questions from players on Reddit and Liquiddota pertaining to the problems they face in the 3K MMR bracket. Throughout this guide I have addressed these questions where relevant (questions are anonymous and paraphrased for convenience).

PART 1: Impressions of the 3K MMR bracket

THE SMURF

Zip Zap...

In order to write this guide I played on a smurf account for a period of approximately one month. I chose to play on a friend’s account rather than calibrating an account myself. The account began at 2850 MMR, allowing me to experience the entire 3K MMR range as I climbed.

Here is the Dotabuff for the account: Dotabuff

THE RULES

For the duration of the climb I chose to play only Storm Spirit. I chose to play only a single hero so as to maintain consistency in my impact on drafting by first picking Storm every game. I chose Storm Spirit in particular as he is one of my favourite heroes, is able to mid, snowball, and is effective at every stage of the game- qualities which I believe are extremely valuable at every level of matchmaking. Out of approximately 90 games on the SEA server I was able to pick Storm every game except for 2-3, in which case I picked TA- a similar hero with similar qualities.

DISCLAIMERS

It is important to address the fact that my experience playing in the 3K bracket is biased. However, I believe that my play style can be emulated by anyone and is just as relevant to an actual 3K player as it is to a smurf, although it make take time to learn, and your climb may be slower.

I also believe that my ability to observe and analyse the play styles of players in the 3K bracket is independant of my skill level. It is obvious that in many cases I was able to carry my team purely due to my own personal skill; however, I do not believe that impedes my ability to observe the trends and common mistakes of 3K players. In fact, many of my wins came as a result of exploiting those observations, a tactic that anyone can use.

Finally, even in games where I was unable to carry my team entirely on my own, my ability to be an effective leader was also able to win me many games. I believe that, with enough practice, anyone can learn to be an effective leader and implement this strategy at any MMR.

THE STATS

This image encompasses stats from 2850 MMR up until ~4300 MMR and so does not perfectly represent my experience in only the 3K bracket, but it gives a good general idea.


Win rate of the account during the smurf period


Progression of the overall win rate of the account during the smurf period

The trends in the images above were as I expected them to be, with a winrate of 100% when the account was at a low 2850 MMR, followed by a slow and steady decrease in winrate as the account started to climb. An overall win rate of ~90% is about what I expect at a minimum for a smurf account 1000-1500 MMR lower than the player, a trend that is common at least anecdotally from players I know who frequent smurf accounts. It is my belief that any player significantly better than the MMR they are playing at will inevitably climb, and smurf accounts present good evidence to support this idea. So if you are stuck at 3K MMR and you think you should be upwards of 4K MMR, you should reconsider your position, and look for areas of improvement which you have failed to notice.

GENERAL IMPRESSIONS OF THE 3K MMR BRACKET

The general trends of 3K players I observed whilst playing on SEA server fall into two overarching categories. In the first category are trends common in higher brackets; aspects that players struggle with just as much as higher level players relative to their difference in MMR:

1. Players have bad mechanics

As I will explain further in Part 1, players develop mechanics in a mostly linear fashion as they develop their knowledge and experience in the game. 3K players are in a stage of development where they have learned the majority of the content of the game, and are learning routines, generalised rules, and developing fluency in their mechanical execution in the game. You can take advantage of this since s you can afford to make riskier plays, as other players lack the polished mechanics to punish mistakes.

2. Players are toxic

Same toxicity, different bracket. SEA is considered the most toxic out of all the servers as a huge variety of cultures overlap and face challenges, both through language barriers and play style clashes. The only real difference in the 3K bracket is the flaming tends to be “more general” as players obviously have less sound reasoning for why a game has been lost, which gives even more credence to simply ignoring criticism.




3. Supports are significantly worse than cores

Support is by far the most difficult role to play in public matchmaking. There are many reasons, but much of it has to do with supports being highly reliant on teamwork to be effective. 3K players lack the consistency and extent of overlapping knowledge to be able to coordinate effectively. Support players need to be able to make assumptions about how the other players on their team will play, which is difficult at lower levels where players have less predictable habits. Also, the breadth of potential knowledge of the support role compared to other roles is so big that support duos may not have sufficient overlapping knowledge to be able to coordinate effectively. Even professional teams can take a long time to develop synergy in their supports, and truly effective supports (think TI4 fy and fenrir) are in fact quite rare. As such, the discrepancy between the ease of playing support compared to other roles is just as relevant at 4K MMR as it is at 3K. Finally, players tend to play supports less often compared to cores, for a variety of reasons, and as such, most players are much less effective at playing supports rather than playing cores.



In the second category are things which players do significantly worse than higher level players.
There are a range of different types of players in the 3K bracket, which I will discuss in detail in part 2. However, the general trend of 3K players is that they are less “tryhard” than 4K players. This results from a combination of-

• a) Players not knowing how to create and utilise advantages
  
• b) Players being generally more interested in ‘style points’ than consistently winning and improving
  
• c) Players having a dissonance between their actual goals as a player and the style with which they choose to play, which I will explain in more detail in part 2.

The result of this is that 3K players’ ability to create and utilise advantages over the length of a game is considerably worse than higher level players. This in turn leads to teams playing as though they have much more of an advantage then they really do, which causes over extension and throwing. For similar reasons, players who are at a disadvantage tend to overestimate how far behind they are, play far too reserved, and fall into the standard toxicity that comes with giving up much too soon.

MODELS OF LEARNING AND THEIR IMPLICATIONS IN THE 3K BRACKET

The Dreyfus Model of Skill Acquisition

According to the Dreyfus Model of skill acquisition, learners progress through five general stages of learning:

• Novice: Players need exact rules and guidelines to direct their play. At this stage of learning, ‘Training and Education’ have the largest impact on learning, as players are still learning mechanics and the content knowledge of the game.
  
  
• Advanced Beginner: The majority of the 3K bracket falls within this range. At this stage of learning players move beyond concrete rules, and begin to learn to adapt to situations. At this stage the bulk of content knowledge has been learned, and the largest impact on learning comes from acquiring situational experience, such that players build the ability to create exceptions to rules for common situations.
  
  
• Competent: This stage of learning is what separates the 4K bracket from the large majority of the 3K bracket. At this stage of learning, players learn to troubleshoot. Although at this stage the largest impact on learning still comes from experience, experimentation comes into play; players begin to look through the crack of the door at the true potential of the game. Players display deliberate planning, and dissolve their focus on rules in favour of goal oriented, adaptable routines.
  
  
• Proficient: At this stage of learning, experimentation becomes much more key, and players are proficient at self-correcting and problem solving in complex situations. Proficiency is most commonly displayed at MMRs over 5000.
  
  
• Expert: The highest level of learning, where players dissolve rules in favour of relying on intuition and instinct. In the eyes of a lower level player, the ability of players at this level to achieve amazing results is akin to magic.

Note that there are large overlaps in the stages of learning in the Dreyfus model, perhaps more so than is displayed in the image above. However it is the impact of each learning area which is the most important thing to consider. For example, at the expert level, the impact of learning new mechanics and content knowledge is practically irrelevant compared to the impact of experience and experimentation.

Reliance on Routines/Shifting Focus to People

As you can see from the images above, the threshold between the Advanced Beginner and Competent stages of learning are gated by 2 major maxims which are the source of a large part of the frustration associated with the 3K bracket. Players who lack appreciation of the impact of these maxims in separating them from higher level players become bogged down in frustration, oftentimes becoming victim to confirmation bias or the Dunning-Kruger effect.

• Switching to a people focus: For players to advance past a focus on content knowledge and rules, they need to switch their focus to reading the other players in the game. Being able to read other players both on your own team and on the opposing team is a huge component which separates 3K players from 4K players. As I mentioned earlier, the ability to read the players on your team is important for support play; similarly, it is important in a general sense for all positions.. Players in the 3K bracket are beginning to learn how to read the other players in the game, predict their actions, and react accordingly (this applies to teammates and enemies alike). However, this can also be a source of frustration, as the 3K bracket is a tipping point for this skill, and players who have begun to appreciate and build upon this skill become frustrated at players who have not and therefore make plays which seem stupid to players who have more awareness.

A good example from Roro of how advanced players utilise their ability to read and predict other players.

• Reading context: One of the largest differences between effective players in the 3K bracket who climb and ineffective players who yo-yo in MMR and remain stagnant is a lack of ability to read context. Players who are stuck in the advanced beginner stage of learning are too reluctant to attempt to read the game state and troubleshoot, in favour of relying on the safety net of rules and routines. Let’s illustrate this with an example:

IMPLEMENTING THE DREYFUS MODEL IN CLIMBING

The Dreyfus Model is best used as a general yardstick to gauge both your own progress and evaluate the play of others. Many of the major problems people encounter in matchmaking at any level come down to the mindset with which they approach the game, as discussed in my last guide. With a good mindset, and the Dreyfus Model in mind, you can optimise your strategic decisions for both yourself and for synergizing with the given group of players you are playing with in any game.

1. Self-Evaluation

The 3K bracket can be very confusing for learners. In lower skill levels,players are primarily learning the general content of the game, and the learning curve is mostly linear;however, this is not the case in the 3K bracket and above. Players around 3K MMR hit various walls as they work to understand and evaluate complex situations and begin to think critically of the game. As a result, learning comes in waves, with hills and valleys where players struggle with a concept, then come to find the game much easier until they reach the next hump.

As the concepts you are learning become more complex, the Dreyfus Model will serve as a good guide to direct your attention.

• Give conscious and objective thought to your strengths and weaknesses as a player; post game analysis is very important for this. Watch your replays and think about your recent games to identify general problems you encounter or general situations you find yourself in which you don’t understand.
  
  
• Then compare these problems to the Dreyfus Model to get an idea of which stage of learning you are stuck at. This will give you a more clear idea of what the likely source of your problem is.
  
  
• Finally, compare the way you play to high level players through watching live games, replays, streams, tournaments etc. or by making friends who are better than you and asking for them to critique your play. Even enlisting the help of a player at the same level as you can be helpful; as I discussed earlier, players in the 3K brackets are at various stages of learning in different areas, so your peers are likely more knowledgeable in some areas than you are.

2. Evaluating others

Know thy enemy, and thy teammate. As I mentioned earlier, the 3K bracket is the tipping point where players begin to shift to a people focus. Being able to judge and adapt to the players in your game is an involved process with many factors to consider, and takes time to learn;o, don’t waste any opportunity to observe players and learn about how people approach the game. I have already discussed the value of learning how to evaluate and predict your enemies, but applying the same concepts to your teammates is just as important and often overlooked. EVERY game can be learned from, and one of the biggest mistakes (which even I am guilty of) is wasting lessons I could have learned from a game because I considered the game to be ‘silly’ or not worth analysing.

A common situation all players encounter is having a teammate who feeds exceedingly. Again, your instinct might be to generalise it as bad luck and not think about the game further; but, if you take the time to look more deeply at games like this, you might find trends in how these players act in the game. Do they seem stressed? Are they making excuses? Are they picking fights? If you understand the nature of the beast, you can take action to lessen the impact of players like this. In parts 2 and 3 I will discuss various strategies to accomplish this objective.

3. Winning the game before you

This isn’t The International, and you are not a professional player. At the end of the day, you are bad, you make mistakes, and you make them often. There are mistakes you make that you don’t even know about yet. Your teammates are the same.

Unfortunately, most players seem to live in a fantasy world where they expect near perfect play from their teammates, and react viciously when they don't. Your teammates will make mistakes and you need to factor this into your expectations. Just as you make many mistakes that you aren’t aware of,, when you notice your teammates making mistakes they may also be unaware; therefore,you should exercise leniency in your criticism. This is the main impact of learning how to evaluate others.

Win the game before you, not the game you think it SHOULD be. This takes time to learn, and requires you to re-evaluate the game at various points, and refocus your goals and assumptions.

Next Time...
That’s all for Part 1! Stay tuned for a breakdown of the 3K bracket in Part 2.

My Twitter
My Stream

---