Gaming as Management Training
Training Software Engineering Managers with Competitive Toxic Team Based Online Video Games
Note: The comments section on this post are as good if not better than the main post. So make sure to read them as well.
When people discover I used to be a chess player they get excited, they’ll say things like “makes sense, this is why you’re into math and programming now and yeah you probably don’t make dumb decisions since you can think ahead”.
I’ve spent more time during my teenage years studying chess than I did studying for school. I never quite went pro especially after I started undergrad engineering and found a girlfriend. But the above comments make me lol, chess is kind of like the SAT - I didn’t get better at tech leading teams at Microsoft and Graphcore by playing chess.
Chess is a popular metaphor in corporate America, something about planning ahead to outplay your opponents is an appealing narrative. You are the CEO, your team are your pieces, armed with the Gartner technology report, you are unstoppable.
For the most part while you’re playing chess you can only experience a finite set of emotions. You can be surprised by your opponents opening. For the midgame you’re either feeling slowly more confident or slowly fall into despair and for the endgame you’re hyper-focused trying to win or see where you can find a draw.
However, chess is not a team game and the downfall of most software projects is not competition but the team itself. Corporations are not in service of their customers, they are in service of their employees. Employees with chaotic wants, needs and constraints — most of which are totally opaque to management.
So if chess doesn’t capture the spirit of business, what does?
DOTA (Defense of the Ancients) is a 5v5 competitive team based multiplayer game captures some of the pitfalls teams fall into. You and your teammates need to each pick one of over a 100 unique heroes that synergize well together so you can destroy the enemy Ancient.
I’ve played a few thousand hours of DOTA and am happily in the top 85th percentile of players globally. It’s an amazing game with endless variety, real time tactical execution and long term strategic planning.
This is a decent 10 minute video summary of how DOTA works
You may opt for a quick assassination strategy where you pick off key enemy targets. Or maybe a flashy team fighting strategy where you stun the entire enemy team at once. You can choose to avoid fighting entirely by playing an economic game, collecting resources faster than your opponents or distract the enemy team while you destroy their buildings.
Regardless of what you pick it’s very difficult to execute on any strategy without some amount of coordination with your teammates.
Which brings us to the main point of this post
If you can convince 4 people you’ve never met to trust you for an hour, help each other overcome a difficult opponent, then you have what it takes to be a leader in the real world
Management seems to be one those skillsets without a clear study roadmap. Generally you’re a successful and consistent independent contributor, your scope increases a bunch and it’s critical for the business, you hire a bunch of people and you manage them.
But being a good manager is a tangential skillset to being a good engineer, sure you can read a bunch of management books but there’s no substitute to learning on the job.
But that means you are going to make tons of mistakes which will make impact peoples lives until you get better. You need lots of iterations and quick feedback loops which are hard to come by in the real world but easy in a game.
So in the same way the military does scenario driven simulations we can use
DOTA as the ultimate management training
DOTA is notoriously toxic, players get upset especially when they’re losing (my girlfriend can attest to this) but few things feel as good in life as executing on a DOTA fight perfectly. Doing so reliably requires exceptional leadership skills so there’s a lot we can learn.
DOTA Lessons for Managers
Shared resources lead directly to toxicity, in DOTA experience and gold are shared resources that usually need to be directed towards your carry (heroes that become better in longer games). But if your carry is bad then this is a frustrating experience, you’ll want to take your carry’s resources which will make them even angrier and the toxicity feedback loop continues.
I can’t remember hearing anyone say they like stack-ranks, ideally the scope and success of teams need to grow rapidly so the pie keeps increasing and fixed sum games can’t be played. Time, attention, salaries are fixed resources - projects are not.
It’s easy to be positive when you’re winning and it’s hard to be positive when you’re losing. Letting a negative mindset take over you while you’re losing will make you launch into all sorts of passive aggressive comments towards your teammates who will dislike and distrust you which creates a reflexive loop where you’re more and more likely to lose.
If you’re on a losing streak and easily irritable, you’re burnt out. Take a break, go for a walk outside and maybe play a different game for a bit.
Every team needs a chief meme officer they create a strong in-group identity by undyingly supporting their own team and playfully throwing shade at their opponents.
Do I suck? No! It’s my team
The worst and most toxic players always think they’re the best player. Even if that were true (usually it is not), blaming your team is an excuse that prevents you from stepping up. Would a better player salvage the situation? If every team you’ve ever been in on is bad, then you’re weakest link.
Macro level communication as in “here is the plan” is very helpful, micromanaging is irritating. Almost universally if you micromanage someone they will tell you to “respectfully fuck off” especially if you’re not paying them.
In the heat of the moment it’s very difficult but much more useful to give a plan. Saying “Stop dying guys” or “We are not selling enough” is not actionable. This kind of feedback often showcases a lack of imagination on the side of the leader. Try instead “We need magic immunity vs their shaker” or “Why are we losing, what are we missing?”
Everyone has a plan until they get punched in the mouth - Mike Tyson
Complex plans are brittle, in real time if there’s a hitch, someone mistimes their spell usage and the whole plan falls apart. Simple plans are modular and reliable and allow for emergent complexity via composition. Teammates can fit their plans and strengths more easily into simple plans.
10xers exist, if a player is 1v5’ing the enemy team, listen to them and see how you can help them. Accept that they are better and you have more to learn from them than they do from you.
Your teammates will listen to you if they perceive you as the best player or the best shot-caller on the team
And the corollary of that rule, is crybabies get muted.
Negative feedback needs to be framed to the whole team and positive to the individual and the team
Trusting your teammates even if it results in a bad play in the short term is still the best long term plan
After a negative encounter, laughing it off while analyzing team level mistakes is invaluable, picking on a specific teammates mistakes no matter how egregious is not a good idea. If your entire ads system goes down, telling the sleep deprived engineer why did you take so long to figure out the bug is not useful, asking them what they would have done differently is.
Positive feedback needs to be specific: “Holy shit man, that Echo Slam was insaaaane”
Some people can’t be saved, a DOTA meme is Tiny airlines where tilted (cartoonishly angry) players will pick a hero called Tiny and toss you over and over again to your enemies to your death for 40 minutes straight. Accept that sabotaging teammates will happen sometimes and be thankful that it only lasts 40 minutes.
Nerves not skills is the true bottleneck. Nothing exemplifies this more than techies, a hero that can instantly kill you with bombs from anywhere on the map and routinely makes games drag out to 90 min+. Techies players are trying to deplete a different resource from the usual DOTA one, your sanity.
If someone is apologizing a lot for misplays, they are often having a hard time in real life. Be extra supportive to these players and they will be in a much better mood if they don’t feel judged by their team and hopefully play better.
People are different, some will ask for constant feedback some won’t
It’s often not your enemies or competition that are outplaying you but your teammates not trusting each other. Self sacrifice is the most iconic form of building trust, no-one forgets it, even after thousands of years.
If you’re a gamer and manager I’d love it if you could reach out and let me know if this article resonated.
If you enjoyed this you’ll probably enjoy
My talk on Youtube on How Games Teach where I go over tons of examples of games that do the best job I’ve seen of any medium at teaching you logistics, factory management, diplomacy and much much more
My beginner’s guide to Dungeons and Dragons will show you my DnD is the ultimate game to coach you on how to become a better storyteller
Why games not algorithms will help us build intelligent systems - see my guide on creating your own Reinforcement Learning environments with Unity ML agents
Games are the most efficient teaching medium we have today.
Cats learn how to hunt by playing. We’re the same, we like the sun, we like hugs but we differ in how we over-intellectualize what we consider to be valuable.
As for me, after 15 years I’m taking a break from competitive strategy games for a while. Hopefully I’ll find something I’ll be as excited about and if I do I’ll be sure to write about it.
Thank you Daniel Bourke for reading and sharing heartwarming stories about what games taught him throughout his life., these kinds of stories are why I started writing in the first place. Vera for a careful analysis of toxicity and leadership in DOTA vs the real world with lots of specific examples. Diamond Bishop for long detailed feedback reflecting on his own management and gaming experience. CharonAbol for picking the catchy title I ended up using which was a much needed improvement over “Toxicity Exposure”.