The 10x engineer and the 10.0 WAR baseball player

tldr; The best engineers bring more to the table than just coding. Like the 10.0 WAR baseball player, 10x engineers are few and far in between. But they definitely exist. This post shares how baseball approaches identifying its best players and how we can attempt to do the same in software engineering to identify 10x engineers.

Disclaimer #1: Of course the 10x engineer is a big generalisation! But let’s suspend our doubts and entertain the idea for a moment. Hopefully there’ll be one or two things we can take away along the way.

Disclaimer #2: A lot of this post is subjective. When I talk of a “great engineer” it will be about a mould of engineer that I personally consider extraordinary.

A definition

We’ve all heard about the mythical 10x engineer. The engineer who can produce at a rate that far exceeds his/her peers. But what exactly does it mean to be a 10x engineer? A Google search yields the a variety of definitions:

  • …an individual who is thought to be as productive as 10 others in his or her field [technopedia]
  • …a highly productive individual who produces about 10 times the work as your average developer [GitClear]
  • Jeff Dean and Sanjay Ghemawat (early Google engineers) [The New Yorker]
  • …their net effect on an organization is at least 10 times greater than the median programmer [quora]

Definitions seem to mostly centre around productivity and impact. For the purposes of this post, let’s use one of them:

A 10x engineer’s net effect on an organization is at least 10 times greater than the median programmer

The median engineer is slightly harder to find definitions for. Short of a study (if you know of one, please let me know!), my gut feeling — based on organisations I’ve worked for in the past — tells me that the mid-level engineer is representative of the median skillset. Suppose you have a (fictional) organisation that has one CTO, three team leads who each have three engineers in their charge. The median engineer in that organisation would be one of the non-leads who will likely be a mid-level engineer (assuming team leads are senior!). Hopefully, qualities of the median engineer can be inferred from the discussion below.

With these definitions, let’s move onto the 10.0 WAR baseball player.

The 10.0 WAR baseball player

Mike Trout, widely regarded as baseball’s best player [Image source]

Baseball is one of those sports where you can get statistics on just about anything: pitch speed, time to run from home plate to first base, speed at which the baseball leaves the bat… the list goes on! It wasn’t always this way. But with technological advances and the rising popularity of sabermetric baseball statistics (think Moneyball), baseball now has some of the most advanced metrics in sports. One of them is the revered Wins Above Replacement (WAR):

A player’s WAR value is claimed to be the number of additional wins his team has achieved above the number of expected team wins if that player were substituted with a replacement-level player: a player who may be added to the team for minimal cost and effort. — Wikipedia

There are several ways to calculate WAR (FanGraphs, Baseball Reference) but the underlying principle in all variations is the same: WAR is a single value that’s calculated from the player’s success on the field and considers metrics around batting, baserunning, fielding, and pitching statistics. It’s also adjusted for their position: comparing a pitcher against a first baseman wouldn’t be a fair comparison. WAR is often measured over the course of a season, so players who play well consistently and play often earn higher WAR. Due to its additive nature, it can be used to compare careers too. Here are some examples of WAR:

  • In Clayton Kershaw’s 2014 season, he had a 7.2 WAR, suggesting he won 7.2 more games than would be expected if a replacement-level player had played instead. He achieved this by pitching many innings and striking out many batters while keeping opponents’ walk and homerun totals low.
  • In Mike Trout’s 2016 season, he had a 10.5 WAR — one of the highest in recent times. He did this by getting on base almost half the time (incredible!), scoring lots of runs and playing solid defence.

In a baseball season of 162 games, a team consisting of only replacement-level players is theoretically expected to win just 52 games. So, adding Kershaw and Trout to a replacement-level team could increase the win total to almost 70! You might rightfully have some doubts about the effectiveness of such an all-encompassing metric but WAR does seem to have a strong correlation with actual wins.

So you might be thinking, “just get the highest WAR players” (indeed general managers think this way too), but players who can generate a high WAR are rare. In the 2010’s, there were only five instances of a player producing 10.0+ WAR. Three of them were Mike Trout (maybe that’s why he owns the largest contract in all of sports). If there were a unicorn in baseball, it would be Mike Trout. To give you an idea of how special a high WAR is — let alone a 10.0+ WAR — here’s a graphic from FanGraphs:

WAR distribution in 2010 season visualised [Image source]

In the 2010 season, only 0.5% of all major league players had a WAR above 7.0. There were no 10.0+ WAR players that year.

The 10x engineer

Baseball’s WAR measures the player’s contribution to the team’s win total over the course of a season as well as their career. It considers many dimensions (e.g. hitting, fielding stats) and is adjusted for the player’s position. What if we could apply a similar approach to attempt to identify a 10x engineer? Can we break down software engineering into dimensions that we can examine to assess the engineer’s effectiveness? Let’s take a look at a few that can potentially increase the net effect of an engineer on their organisation. We’ll also — to make the discussion more fun! — come up with our own simple model (the “Net Effect Above Median” metric) to measure the engineer’s multiplier effect.

1. Is extremely productive

Output is by far the most common dimension by which engineers are measured in any discussion around 10x engineers. A 10x engineer is often said to be able to do a task 10x faster than the average engineer. Indeed, there are engineers who can churn out a feature at a pace no other engineer can. Experience and work ethic matters.

Where this might fall short however is when you compare a senior engineer and a junior engineer’s output for a very basic task (like updating copy in the front-end): there won’t be any 10x effect, no matter how much experience or effort you put into it.

But let’s consider a more open-ended task, such as designing a system or product from scratch. What pragmatic shortcuts does she suggest to deliver value to the business faster? If the company succeeds, the shortcuts will be directly attributable to the engineer’s pragmatic decision making. But taking too many shortcuts may lead to a maintenance nightmare or tech debt, so good judgement is required to sustain a good balance. The ability to ship value quickly but sustainably matters.

Our Net Effect Above Median (NEAM) metric:

  • Median (1.0x): Can take on tasks that are open-ended but only up to a certain level of complexity. Ships value consistently but may not go out of their way to fix issues that can lead to tech debt. May be very good at one type of task (writing queries) but not at others (data modelling).
  • 1.33x: Can take on all types of tasks in their domain of work. Ships value quickly but may be doing so without considering sustainability of code. May do things only to a point where the output works.
  • > 1.5x: Ships value consistently and quickly but considers how their work will impact others. Reduces the effect of shortcuts taken by making sure to address them at a sensible time. Their work can be re-used by team mates and increase the team’s output time and again. May research and introduce new tools that make the team even more productive.

2. Maintains high quality work

For open-ended tasks again, what design choices does the engineer make to promote contribution and productivity? Here, the better engineers would weigh the short term costs of shipping early against the long term effects of choosing the right method, framework and models. Where the average engineer might copy-paste code just to make things work (we’ve all been there!), a strong engineer might refactor code to make copy-pasting no longer necessary (saving others time). If the company scales, the early design decisions might make future hires more productive. But if too much time is spent future-proofing code, it will slow down shipping code. The ability to make sound design choices while delivering value matters.

Our NEAM metric:

  • Median (1.0x): Will follow examples of similar work. If similar work doesn’t exist, will still eventually deliver value but, left on their own, may introduce bugs or potential maintenance issues for team mates.
  • 1.33x: Is aware of how their work might affect others’ work. Has the capacity to produce good quality work on their own but may often need more senior colleagues to implement best practices. In their delivery of value, may produce good code but not have considered the user’s experience of the product.
  • > 1.5x: Their work is often the organisation’s first example of the type of work they do. Increases the team’s output time and again with their work. Tech debt is addressed at a sensible time but their sound design decisions also reduce the potential for debt. Work is of a high quality for both team mates and users of the final product.

3. Communicates well

Whether it’s documentation, communicating progress to stakeholders or public speaking, communicating your work is an important skill for anyone who might be considered 10x at their job.

If an engineer doesn’t communicate well, there’ll be less awareness of their work, a higher risk of misalignment, risk of implementing outdated specifications and less help coming their way. The final point is important: even the best engineers need help from others. For instance, business folks often have the best ideas to deliver value faster, as they might come up with creative (frequently non-technical) ways to solve a problem. Not keeping them informed is a missed opportunity. Communicating well can also attract organisational support, which can help accelerate the delivery of business value. The best engineers communicate regularly and collaborate with others to ensure value is delivered.

Our NEAM metric:

  • Median (1.0x): Communicates their work in daily stand ups. When they have issues, they’ll know to ask. Sometimes might get stuck and not communicate it well.
  • 1.33x: Communicates their work well. Issues are flagged in a timely fashion and when there’s an issue, they’ll pull in people from their team or outside their team to resolve the issue. When requirements for their work aren’t clear, they’ll attempt to clarify them with the right stakeholder. If juniors in their charge have issues, they’ll discover the issues early through regular communication.
  • > 1.5x: Keeps team mates but also external stakeholders in the loop. Documents and explains things in a way that new joiners or juniors can learn quickly. Others don’t need to review their work long: they want the audience of their work to not only see but also understand it (think sparse vs. detailed code reviews). Sets best practices and communicates their justifications well. If there’s a call for action, they’ll state their recommendations and recommend the best course of action. As a senior engineer, they’ll make sure to regularly check in with people in their charge (1-on-1’s) but also communicate initiatives and achievements with the wider team in all-hands.

4. Has cross-functional impact

Most of us like to stay in our own domain of expertise and rarely venture into a different area of business. That’s fair but sometimes an engineer’s small contribution to sales, support or marketing can have a several-fold impact for the business. Non-engineering functions might have small inefficiencies that accumulate over time. For example, support may be getting tickets for the same technical issue again and again that needs to be resolved in a very manual fashion. The engineer’s ability to listen out for such problems can potentially yield technical solutions that make other parts of the business perform better. At some companies, this is a big enough opportunity that they hire marketing developers or support tools engineers. Of course, not every business workflow needs optimisation. And not every optimisation requires a sophisticated solution. But a 10x engineer might pick up and find sensible solutions to a few interesting opportunities to improve business output outside their usual work. A great engineer might look out for opportunities to deliver value outside of engineering.

Our NEAM metric:

  • Median (1.0x): Rarely considers how their tasks will impact functions outside of engineering.
  • 1.33x: Considers how their tasks might impact other teams. Keeps support in the loop when there’s a risky release. Keeps product (and, indirectly, sales) in the loop about new features being released.
  • > 1.5x: Considers the impact and risks of their work but also others’ work. Doesn’t only keep other teams informed but also may go out of their way to make other teams more effective with tech. May introduce processes (e.g. key stakeholder meetings or newsletters) so that other teams are kept in the loop about the latest in product development. Is the go-to person in engineering to solve any type of problem.

5. Manages relationships well

Even if an engineer isn’t in a managerial role, there will be relationships to manage. This might come in the form of mentoring juniors or scoping work with product management. The very best engineers might also manage relationships with other functions: release engineering, QA, support, sales and even customers. Where this really shines is when the management of relationships makes others 10x better. Whether it’s helping a junior accelerate their career or helping sales manage customer requests, a great engineer will be the go-to person to provide advice. It can boost morale to have someone who understands others’ objectives and provide pragmatic suggestions to create more win-wins.

Our NEAM metric:

  • Median (1.0x): Has good relationships with the people who work closest to them. Doesn’t engage too much with people outside of their immediate sphere. Has the capacity to mentor but doesn’t go out of their way to help.
  • 1.33x: Has good relationships with people who directly work with their team. Engages in work with the product team. Regularly gives up some of their time to make sure their mentee is looked after.
  • > 1.5x: Has good relationships with many stakeholders. Shows care when communicating with people in their charge, customers and external partners. Can win (customer) deals. Is the go-to person for discussing opportunities and solving problems.

6. Can boost hiring

If you’ve attempted to hire good engineers, you’ll know it’s no easy task. They’re few and far in between, and often you’ll search many months to make your hire. Especially if you’re not at a big Tech Co (Facebook, Google, etc.), you might also experience issues getting enough engineers through your hiring pipeline. But sometimes, an engineer you hire can bring more engineers to the business through their personal referral (at a company I used to work at, one person was said to have referred 30+ colleagues successfully). We should of course make sure to have the right processes to vet those referrals but, if they are successfully hired, the initial engineer may deserve our thanks!

Another way in which an engineer can influence hiring is their ability to screen and interview. An engineer with a good understanding of the market will know where to look and what to look out for when considering engineering candidates. Good interviewing will reduce the likelihood of bad hires who can derail projects and lower team morale.

Finally, a great interviewer may also make the interviewee want to work at the company. Great engineers attract great engineers. A good interviewer will often show professionalism —always prepared and communicative of the interview process — and leave a good impression to potential hires. An interviewee will often place emphasis on their impressions of the people they interviewed with when deciding on job offers. By having an engineer who can communicate well, excites the interviewee and acts in the interviewee’s best interests, the business might hire a few additional engineers of a high calibre. In basketball this happens all the time: LeBron James famously left the Cleveland Cavaliers to join the Miami Heat and brought together The Big Three that won multiple championships. Hiring a great engineer might lead to more great engineers joining your business.

Our NEAM metric:

  • Median (1.0x): Doesn’t participate in many interviews so doesn’t have much of an impact on hiring.
  • 1.33x: Participates in interviews, assesses technical ability well and makes a very good character judgement to reduce the likelihood of bad hires.
  • > 1.5x: Doesn’t only interview well but also can communicate the company’s vision and how the candidate will play a role in it. Leaves a good impression on the candidate and attracts great engineers to work with them.

10.0 NEAM = 10x engineer?

Our discussion made an attempt at measuring the engineer’s net effect on the organisation with the (made-up) Net Effect Above Median (NEAM) metric. It’s by no means a tested measure and was only created to add some fun to this post. But, for the sake of discussion, how might you identify a 10.0 NEAM engineer? If an engineer scores highly (>1.5x) in all six dimensions described above and assuming the dimensions are not additive but multiplicative (which should be a fair assumption), the engineer will score above 10.0 NEAM (1.5⁶ =11.4).

To summarise, a 10.0 NEAM engineer might exhibit the following qualities:

  • Is extremely productive: Ships value consistently and quickly; considers impact of work on others; addresses tech debt; their work can be re-used by team mates; introduces new tools to make the team more productive.
  • Maintains high quality work: Their work is often the organisation’s first example of the type of work they do; their sound design decisions reduce tech debt; work is also high quality for users of the final product.
  • Communicates well: Keeps team mates but also external stakeholders involved; documents and spreads knowledge; their work is easily understandable; sets best practices but also justifies them well; recommends the best course of action; regularly checks in with colleagues but also communicates with the wider team in all-hands.
  • Has cross-functional impact: Considers impact and risks of their work but also others’ work; makes other teams more effective with tech; introduces processes so others are kept in the loop; go-to person to solve any type of problem.
  • Manages relationships well: Has good relationships with many stakeholders; shows care in communication with colleagues, customers and external partners; wins (customer) deals; go-to person for discussing opportunities.
  • Can boost hiring: Interviews well; can communicate company’s vision; leaves good impression on candidate; attracts great engineers.

Does this sound like a 10x engineer to you?

We can all agree that 10x engineers — like the 10.0 WAR player — are difficult to find. They’re hard to come by because it’s not easy to score highly in all of the dimensions that define your discipline. But they definitely exist. Hopefully this post was able to demonstrate a (fun!) way of identifying the 10x engineer.

Notes

A note on brilliant jerks

You might ask, what about the engineers who are technically brilliant but don’t engage with anyone and dismiss every opportunity to work with others? At best, you might have someone who will produce an unbelievable amount of work but not engage much with the team. At worst, you might have someone who rubs the entire team in the wrong way: brilliant jerks. The above discussion is about the net effect of an engineer on the organisation, so individuals who don’t work well with others won’t score very highly.

A note on sub-median

What if an engineer doesn’t possess the median skillset (yet)? In the above discussion we’ve given median engineers a NEAM of 1.0x. Following this logic, engineers who lack the median skillet will have a NEAM of under 1.0x.

But a NEAM of under 1.0 might not necessarily be a bad thing. My interpretation is that a NEAM of under 1.0x simply implies that the engineer still has room to grow (this may be especially true of juniors). They may produce less than the median engineer today but may grow to be an above average or even 10x engineer over time. But let’s not forget: NEAM is something I’ve made up for this post and may not fully capture what it means to be a 10x engineer!

That’s as far as the discussion goes! Thanks for reading.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Mario Hayashi

Mario Hayashi

Product engineer, No-Coder, contractor, tech leadership at startups, indie maker.