OOPS DOTA / My stories

3 kills. 33 assists. What was I doing in this game?

The kill column is short. The list of things to do is not. This is the match behind my first OOPS DOTA episode — with the good plays and the decisions I would change.

MY GAME / CM01

Watch. Then break it down.

Episode page ↗

You can watch the full match right here. Below are three short moments from the same game.

Press play to load the YouTube player. YouTube then processes connection and playback data under its own privacy policy.

Episode, chapters and items →Open on YouTube ↗

Three moments from this match.

Browse the short stories →

Another short look at our Crystal Maiden: 3 kills, 33 assists and MVP.

Episode + breakdown · 10:10 ↗

The YouTube player loads only after a click. Audio tracks and subtitles are in its settings.

OOPS DOTA Shorts

I found the lane's limits through my HP bar

While my carry farmed, I tried to push their heroes away from the creeps. Cast, back off, repeat. The trouble starts when I forget the “back off” part. I left those moments in the edit: a strong late fight should not rewrite an awkward lane into a perfect performance.

Vision in my matchEXPLORE THE EXAMPLE ↗
Vision in my match

Look at the visible area and dark parts of the map: no opponent in the frame does not mean nobody is nearby.

OOPS DOTA / Ruslan Ukrainskyi

54 HP before the teleport

Jump to 03:38 for the escape. The confirmed 54 HP value is from just before the teleport, not a claim that it was the lowest health of the whole sequence. My job there becomes very simple: use what I can, find an exit and stay alive long enough to leave. Not every highlight needs another kill.

An entrance is not a shopping list

At 04:40 I still have no Blink. I have to walk into position. That makes the approach part of the play rather than a gap to skip. A little later, I let others commit before joining. The question I want to review is what I knew before stepping forward — not how obvious the danger looks after I die.

The team goes in. I bring the winter.

The fight at 07:01 is the moment from the opening. The team commits first; I find a chance to add control and area damage, and the follow-up arrives. That is the play I wanted. I did not produce every kill on the screen. Making room for someone else to finish is a perfectly good reason to play support.

Freezing Field / CMEXPLORE THE EXAMPLE ↗
Freezing Field / CM

How Freezing Field works: the area around CM and its explosions. This is Valve's demonstration; watch my entry into the fight at 07:01 in the episode.

Valve · official demonstration

After the fight, there is still a game to finish

The Tormentor at 06:30 is a team effort. Then we push the lanes, pressure the base and finish the Ancient. I end with 3 kills, 9 deaths, 33 assists, 31 301 hero damage and MVP. I'm happy with the result. I still want to review the risky entries: one win doesn't make every decision right.

One thing to try in your next replay

Pick one fight and pause before your entrance. Who can catch you? What has your team already committed? Can you contribute without being the first target? Then watch the outcome. Keep one change to test next game. This match is an example to think with, not a promise that copying my items will win yours.

Sources and context

3 KILLS… AND MVP?! This Crystal Maiden Changes EVERYTHING | Dota 2 ↗

I start with my own games. I link sources, give the team credit and keep mistakes in the story. One win is an example, not proof of a universal build.

Less fog. More answers.