Dota 2

Improvement plan and mentality

Content of the article: "Improvement plan and mentality"

Hello friends!

It’s a new year, and I imagine many people here have some new MMR goals! It’s a common question here on how to improve and gain MMR, so I wanted to share my own mentality and method that I used to get to Immortal (now roughly ~6.7k NA). It’s advice you’ve probably heard before, but I see the question all the time anyways so here it is again. For those wondering, I played only 5 to get from low Divine to Immortal, and now play a mix of 4 and 5. I believe this advice is relevant to all roles though.

I've made some videos on it if you want more details, but the main ideas are written below.

Grinding games is a slow method of improving

Thinking you need to grind games to slowly climb the ladder isn’t wrong, but it’s most likely going to be very slow. What you actually need is a high win rate. Sounds obvious, yet many people continue to simply grind games without trying to improve outside of ranked games.

We want to play fewer games at a high win rate rather than many games at a low win rate.

If you want to gain 210 MMR:

Win Rate Solo games to play Party games to play
50.1% 3500 5250
50.5% 700 1050
51% 350 525
52% 175 262.5
55% 70 105
60% 35 52.5

For comparison, gaining 1000 MMR takes roughly 333 games at a 55% win rate. An 80% win rate would get you from 0MMR to 6K in 250 games (aka boosters and smurfs).

If you want to play around with how long a certain MMR goal will take you at a given win rate, you can use this Google Sheet I’ve set up. Go to File>Make a Copy; Save it to your own drive and then you’ll be able to play with it yourself.

Even a small difference in win rate can save you hundreds of hours, which adds up if you’re trying to go up an entire medal.

Improvement, then MMR

We want to improve first, and then win MMR later. In my opinion, you need to do something to improve besides grinding ranked games. Practicing mechanics and watching educational content may seem like a boring time commitment, but if a 1 hour video helps you win a few extra games, it actually saved you hundreds of hours grinding MMR. Again, tackle this however you’d like.

Read more:  Mithia Hail, the Dusk Wastrel

A simple improvement plan

I’m going to briefly describe my own plan here (more details and explanations in this video). If you want to improve in a different way, that’s fine! I think it’s great to come up with your own plan. This is what worked for me (and I imagine many others).

Prepare a hero puddle

  1. Play one role
  2. Pick one main hero
  3. Pick 2 back up heroes

Prepare your hero

  1. Read the Dota 2 wiki for your hero
  2. Last hit in an empty lobby
  3. Last hit against insane bots
  4. Practice combos
  5. Theory craft what each spell can optimally be used for
  6. Get a feel for how much damage your spells (and attacks) do
  7. Look for guides on the hero
  8. Watch high-level replays (at least a couple before even playing)
  9. Plan a generic build and game plan

Play the game

  1. Pick your main hero
  2. Do your best

    • Prioritize the laning stage and early-game
  3. After every game:

    • Why did the game go the way it did
    • Something you learned
    • Something you need to improve
    • Something you did well (Way to go!)
    • Can't think of anything, watch the replay and find something
  4. Repeat as long as you feel good about playing

  5. If you don’t feel like you can play your best:

    • Take a break
    • Go practice
    • Watch educational content/replays

Mentality Tips

  1. You suck. Ditch the pride. Start this process with humility and it’ll go faster
  2. Do not focus on other players
  3. Only focus on what you can do to improve
  4. Assume any game that isn’t on your main role/hero is a loss
  5. Losing is part of the process
  6. If a game looks like a loss:
    • Just try to learn something
    • Experiment with something new
    • Have fun and de-tilt

General Improvement

  1. Read the Dota 2 Wiki Mechanics
  2. When you experiment:

    a. Actively pose a hypothesis

    b. Do it

    c. Actively explain the results

One role, one hero

Stop playing so many different things at once. When you keep doing different things every game, you are always focused on a new hero. You need to play as few heroes as possible (one) so that you become VERY comfortable with the hero. Then you can play DOTA instead of playing the hero. You don’t have to be a hero spammer forever but give it a shot. Getting really good at one hero will passively help you improve at other heroes as well making it easier to add heroes to your pool later.

Pick one main hero to play every game. The goal is to improve this hero and maintain a high win rate (55%+). The improvement process will blend into gaining MMR.

Read more:  Namu, The Wood Spirit Bounty: Team Spirit

I don't care what role or hero you pick. It's probably a little easier to pick a hero strong in the meta, or maybe one that is just generally good for how you play the game. You are going to be playing this hero a lot though, so I strongly recommend picking one you like. I believe any hero can work, it might just take a bit more work.

Do this for as long as you can make yourself spam heroes or until you are happy. I personally think you’ll need at least 30+ games to start to really be good at a hero for your given level. I chose Jakiro. I have almost 300 games of Jakiro. I’m pretty good with the hero, but I know there’s still a lot of stuff I can improve on – stuff I wouldn’t have even been aware of if I hadn’t spammed him in the first place.

Freedom and confidence with one hero

I found that having one tryhard hero that I was really good with helped relieve a lot of the stress from losing games and nerves of learning new heroes.

Let’s say you make your way to Ancient 1 by spamming Juggernaut. You decide to learn a new hero, have some fun picking whatever, etc. You drop down to Legend 3, but I don’t think it feels that bad anymore. You will always know that you can just go back to picking Juggernaut if you want to be Ancient again. The lessons you learned from spamming Juggernaut will speed up the improvement process for learning new heroes. Plus, if you queue into someone who flamed you last game, you’ve got a tryhard pick to kick his ass and there’s nothing more satisfying than that.

Playing to improve

You’ll need to play to improve as well, but you should mentally separate playing to improve vs playing to win. We do not care about losses as we improve. We are here to practice and try new things. Pick a few things (A FEW! Trying to do too much at once is just like trying to play 20 heroes at the same time) you are working on each game. Set goals like:

  • I will get every Rosh timer
  • I will secure every ranged creep
  • I will stack at least 5 times
  • I will try to be aggressive as possible this game
  • I will aggressively cut waves this game
  • I will be in every fight this game
Read more:  Early Game Supporting Tips [Mostly Pos 5]

Not all of these things are good to do every game. However, you can only learn to identify when it’s good or bad by trying to do them. YOU WILL LOSE GAMES DOING THIS and it’s absolutely fine. The improvement process begins with losing and that’s normal.


That’s it. The videos have more details if you’re interestednudgenudge . I’m probably more helpful to support players, but I’m happy to try to answer any questions I can.


Similar Guides

Top 7 NEW Games of January 2021

New year - new month - new games. Take a look at the first 2021 games you’ll be playing on PC, PS5, PS4, Xbox Series X, Xbox One, Switch, and more.

More about Dota 2

Post: "Improvement plan and mentality" specifically for the game Dota 2. Other useful information about this game:

Top 10 Best Video Games of 2020 (So Far)

In times of uncertainty, video games allow us to escape from the stress of the real world. For this list, we’ll be looking at some of the best games released in the first half of 2020.

You Might Also Like

Leave a Reply

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