Awkward is Right: Proof that Ana is a DPS, Not a Healer
Awkward/Artif3x coaching challenge results
TL;DR
After I challenged the advice in Awkward’s “Educational Ana Unranked to GM” video in a Reddit post, he offered to coach me and prove his point. I played through the entirety of Season 2 on Ana, using his style, and managed to climb from a low of Silver 3 to a high of Platinum 5.
I kept a detailed match log of the entire season, then wrote a data analysis program to determine what Ana’s most effective route to carrying matches was. The result was that Ana’s level of healing has almost no correlation to whether you win a match or not; instead, killing enemies and not dying gives you more control over whether the match is won or lost than any other measurement.
I will be sharing the charts that prove this below. In the future, I’ll also share the templates I used to track my match data, and even correct Blizzard’s flawed data that is shown in the “Statistics” screen.
My original post was wrong (partially)
This is the original Reddit post that I created: https://www.reddit.com/r/OverwatchUniversity/comments/z6ig1s/awkward_is_giving_90_of_ana_players_bad_advice/
Here are the assertions I made in that post, and my revised thoughts after spending time under Awkward’s coaching:
“Awkward is giving 90% of players bad advice.” - False
This is false, and is the easiest part for me to now refute. Awkward doesn’t give bad advice. He gives great advice. He has quite a few students that he’s helped improve, and he did the same for me.
Why did I originally say that? Because I took away from his video an incomplete and improperly prioritized list of changes to make to how I played. It’s very easy to watch that video and come away thinking only “Damage, damage, damage”. If you took the whole 11+ hour video, transcribed it, and did a word count for “damage”, I bet that would be 10x more than any other word he said.
If, after his coaching, I go back and rewatch it, I can easily pick out all the other foundational advice he’s giving that’s crucial to understand and internalize prior to focusing on shooting the enemy. All of it is far less emphasized and dramatic than his seemingly-insane DPS’ing, but it’s all there.
“Awkward’s play style [only] works when you’re Awkward, or close to it.” - False
This is also false and easily refuted. I’ve seen this not just in my own gameplay, but in many of his other students during live coaching sessions in Awkward’s “Rank Up Academy” that I spectated. The more of those sessions I participated in, the more common themes I saw in his advice, and my own ability to see mistakes increased.
You can see this below in my win rate and rank chart. The improvement is plain.
Above, matches 1-111 were Season 1, and the win rate there (~52%) was heavily inflated by my old play style, where I would:
Always swap to counter pick the other team. This meant that if the enemy team were dive-heavy, I’d swap to Moira (~57% win rate), or Zen (~62% win rate). That pushes my Ana win rate up quite a bit.
Heavily focus on healing my own team, including defensive grenades.
Keep my cooldowns to defend myself.
Stay in comms.
Nano or pocket those who requested it.
Compare that with my coached Season 2 gameplay:
Always on Ana; never swap, no matter what. The win rate you see there is unboosted by other heroes.
Heal my team when required, saving grenades for offensive use.
Use my cooldowns as soon as they were up and had a target.
Dropped out of comms and/or chat as soon as anyone got negative.
Nano frequently and reactively.
“[It will be a] Disaster” - True, but only at first
Looking at that same win rate chart above will show you the truth of this. My win rate got crushed. However, this is to be expected whenever you try to tear down a skill and rebuild it from scratch. Your ego is similarly going to get crushed, and you’re going to be depressed, or mad, or on tilt, or however you usually react to getting stomped in comp matches. That will happen a lot.
There were several sub-points that I made in that original post that I won’t re-list here, but most of those were also correct, but again, only at first. You’re going to be doing more difficult things trying to play in this style, but the more of it you do, the easier those difficult things become.
Doing damage more often is going to do all the things I said, like making you more vulnerable (you’re not hiding behind a wall all the time), and you’re going to get hunted mercilessly by annoyed DPS, or even enemy tanks, (tanks hate being slept), but you’ll get much better at duelling as a result.
Still, prepare to die a lot as you make the transition.
“You will be flamed in spectacular fashion” - True
Oh, so, so true. I struggle to describe the unholy shitstorm you will go through trying to play like this if you stay in comms. Your teammates will hate you more for not healing their damage than they do the enemy player that did the damage to them in the first place. All they have to gauge why the match is going wrong is the scoreboard, where your healing isn’t high enough for their tastes.
“Ana is a HEALER idiot!” and “Ana stop shooting the enemies!” are on the friendlier side of what I’ve had directed at me. I’ve had people ask me to commit suicide in real life, or had them say that my mother should have aborted me in the womb. I admit I can chuckle about these now, but that kind of spleen-venting is really distracting during a match. I found that one of the factors in my lost matches is the high tendency of my own team to tilt based on what I’m doing. I’d estimate that if things got tilty, we’d lose the match around 70% of the time. In the worst cases, I had entire teams band together to report me for game sabotage for having more damage than healing.
I regularly ended up with more elims or damage than the DPS on my team, which was likely due to them expecting to have a pet Ana pocketing them the entire match, and wandering out into the open to be focus fired by the enemy team. Often my team would split and wander off to be killed while I kept the other fraction of the team up. If you’re annoyed by cries for “heals!”, then you need to brace yourself to play like this.
There were some cases where the criticism had some basis. Trying to find the new balance between damage and healing sometimes ended up with teammates right next to me dying because I had tunnel vision (literally with Ana’s scope) on an enemy I was trying to secure a kill on. Maintaining that dual awareness between my own and the enemy team was probably the most difficult part of this to learn for me.
The Proof
One person taking some coaching and improving isn’t a large enough sample size to prove Awkward’s point about how Ana should be played. What I’m showing next here is the closest thing to proof that I think anyone can provide that Ana is not best used as a healer if you want to win matches.
The data I’m showing below covers almost 700 competitive matches I played since taking Awkward’s offer. This is well beyond what’s required for a complete and definitive data set.
Ana is a healer - False
Ana is not a healer; she is a Support DPS.
That’s a bold statement. Allow me to back that up.
Here is a table that shows all my Losses, Wins and Draws, along with the amount of healing that I did during that match. The “Win average” is exactly what it sounds like—the average of all match healing, and similarly the “Loss average” are the lost matches. “Correlation” here means that drawing a line between the “Loss” and “Win” dots should be a steep angle.
As you can see here, the amount of healing Ana does means almost nothing when you want to win the match.
And, despite the “Damage, damage, damage” mantra of Awkward’s, the results are even more meaningless for damage itself:
Again, the imaginary line between the dots is almost flat.
So wait, what? How can damage nor healing affect win outcome?
So what does affect the win/rate loss? What is highly correlative? What do I do to win matches as Ana? Well, here you go, in beautiful data form, is your answer:
Answer: kill enemies and don’t die
The chart above shows the average Elimination/Death ratio over a period of 10 minutes. Just look at the difference between the win average and the loss average here. I never once had a E:D/10 ratio over 6.5 and lost a match. Not once in almost 700 competitive matches.
Need more? Here’s 400 competitive matches of data for you:
The above is the same E:D/10, but mapped alongside my Win/Loss trend, which is simply the average of my Wins (+1) and my Losses (-1) across the last 10 matches. A top score of 1.0 would mean I’m on a win streak of 10 matches or more, and a low of -1.0 would mean I’m on a loss streak of 10 or more. A score of 0.0 means I’m exactly 50% win rate—5 wins and 5 losses in the last 10 games.
Just look at how closely these two lines follow each other. Compare that now with the same chart, but with healing replacing the E:D/10:
You can see in some cases, when healing went up, I started losing, not winning, or vice-versa, when my healing went down, I started winning. There’s almost no link between the two; sometimes it follows, other times the line goes the opposite direction.
Closing Thoughts
First, I want to thank Awkward for his offer and the time he spent talking with and coaching me. It was educational, fun and gave me a peek into how a number one ranked support thinks.
If I get one thing out of all this, I want it to be shifting the community perception of the support role away from healbotting. It’s damaging to the game in multiple ways. DPSs and tanks are getting used to being pocketed and are denying themselves the opportunity to learn to play well without them. But worse, the idea that supports have no use outside healing is causing support players to get raged at and has turned voice and chat comms into a toxic waste dump.
What’s next
I’m planning to try and apply what I’ve learned to other heroes. I’m going to swap to DPS for a bit, and go back to some of the other supports to try them out now that my style has shifted significantly.
I’ll be keeping similar data on all those matches as well, and will post more of what I find about what’s most meaningful to each of the heroes for turning a match.
I’ll also look at some way to release the tooling I’ve written to track all this and produce charts from it, while avoiding Blizzard’s error-riddled data. I’ll write up why Blizzard’s data is broken along with data as proof soon.
Questions?
You can reach out to me on Reddit, or presumably here through Substack. This is a brand new substack, so I’m still getting this figured out.