Of all the sports underdog stories ever written, nothing comes close to the Leicester City win of the English Premier League in 2015-2016. Ignored and underestimated by their peers, financially limited, and on the verge of relegation, Leicester defied 5000-1 odds to clinch the most astonishing title win in sports history.
To put this in perspective, the odds of victory for the 1980 U.S. Olympic “Miracle on Ice” hockey team were 1000-1. Like the U.S. hockey team, Leicester did not have the most talented or experienced players. They had the third-lowest average possession in the league, the second-lowest average pass success rate, the third-lowest number of short passes and the highest number of penalties received.
Unlike the other top teams, Leicester wasn’t bothered by this. Why? Because they took a different strategy, one that looked beyond the usual shots on target and high-tempo attacking tactics. One that provided them with a winning differentiator: an innovative sports science and medical team, carefully integrated into the decision-making process.
Similar to winning sports teams, high-performing technology organizations are changing how they measure performance. This article compares two key areas that winning sports teams and high-performing technology organizations measure to understand what it takes to level up performance: workload and fitness/safety.
Instead of relying on generic data, such as shots on target or possession statistics, top sports teams create their own metrics and build algorithms to fit the club’s philosophy and tactics.
Leicester plays a defense game, using tactics that allow them to play with fewer players, a benefit for a financially strapped team. Tactics include:
- Kicking the ball farther up the field than most teams
- Making passes predictable
- Developing effective partnerships on the field (e.g., relationship between the right-back and right-midfielder)
- Constant evaluation
- Effective relationships of players, medical team, managers, coaching staff, analysts
So, what do they measure? Paul Balsom, Leicester’s head of sports science and performance analysis, focuses on two things: injury reduction and performance improvement.
As such, they pay particular attention to optimal load, which includes all the games, all the trainings, all the gym time, conditioning and medical time. If the load is too high (or too low), they won’t have optimal performance. Too much load on players causes injuries.
Players get 48 hours off after each game to recover. Unlike other clubs, they also get an additional day off mid-week. Leicester understands the benefit obtained when players are not fully loaded. This idea should sound familiar to tech organizations that benefit from visualizing work-in-progress (WIP) and focus on flow time versus resource utilization.
Leicester employs a medical staff plus a 10-person team of analysts that monitor everything. While the club spends a fraction of what the larger clubs do, they recognize that in order to win, the right tools and support staff are paramount.
Traditionally, technology companies have measured resource utilization with the idea that developers and engineers must be kept fully utilized to receive maximum return on investment.
But operating a software development process near full utilization actually increases delays due to dependencies, conflicting priorities and unplanned work.
Additionally, high utilization contributes to stress and fatigue (from long hours hunched over a keyboard), to a load that is unsustainable — no matter how much people love their job. Eventually, overloaded workers exhibit burnout symptoms: cynicism and detachment, emotional exhaustion and feelings of ineffectiveness and lack of accomplishment — factors that undermine high performance.
Instead of resource utilization, one area that high-performing organizations are looking at is flow, the continuous smooth and fast delivery of business value. From the initial business request all the way across the value stream to production, flow is how business value is delivered quickly.
The single biggest deterrent to flow is too much WIP. Why? Because WIP and flow time have a relationship. High WIP means that some items sit waiting in queues longer. There is science behind this called queueing theory, which is a field of applied statistics that studies waiting lines.
Queueing theory allows us to quantify relationships between wait times and capacity utilization even when arrivals and service times are highly variable. It’s like what happens when we’re in the middle of a deployment and a switch goes bad, taking out 1,000 servers. Or when intruders hack their way onto your now unsecure database servers.
Queuing theory is the math behind why 100 percent capacity utilization doesn’t work, especially in a domain with low predictability, such as software development and delivery. Statistically, once we get past 80 percent utilization, things slow down and queues build up. Freeways utilized at 100 percent utilization come to a grinding halt.
Too much WIP across all the teams working in the value stream is responsible for late arrivals and deliveries, often due to conflicting priorities, particularly between teams with high dependencies. High dependencies equal high wait times because people aren’t available when you need them. Imagine a goalie not being available when you need him.
In a league where $330 million is paid out annually to injured players, Leicester takes injuries seriously. They measure the optimal load for every player using GPS systems and heart-rate monitors. Additionally, players complete questionnaires daily to identify how they are feeling in general.
If a trend in quadricep soreness emerges, coaches can adjust training sessions to reduce potential injuries. Sports analytics show that 40 percent of sports injuries are avoidable. Leicester has the lowest number of injured players in the league, which gives them an average player availability of 96 percent, the best in the league.
Soccer is all about energy management. Games are 90 minutes long, and few people can sprint that whole time. Leicester fitness levels make them hard to beat. Their capability to play at a high tempo throughout the whole game puts them in position to score towards the end of the match. You cannot wear them down or just wait for them to tire.
It’s worth noting that players advocate for the type of training they need; they are okay saying, “My quadriceps are sore.” The equivalent of that in tech is someone saying, “I’m tired from working late (or over the weekend).”
One benefit of looking at the sports domain is that it makes it easier to spot essentials that can be applied to other domains, such as safety in tech.
How does safety play out in the technology domain? By making it okay for workers to be honest about their workload, which helps with the capture of clean data allowing decisions based on valid data.
While at a conference, I overheard an attendee at a vendor booth ask if there was a way to exclude weekends from lead time reports. I politely asked, “Why do you want to exclude weekends?” He said, “My team doesn’t work weekends. I don’t want that time counted against us.” I asked, “How is it that lead time metrics count against you?” He said, “It’s in my goals – it impacts my performance review.” As Eli Goldratt said, “Tell me how you’re going to measure me and I’ll tell you how I’ll behave.” If people feel safe making their work visible, there is less incentive to game the metrics.
How are high-performing tech organizations gauging fitness and safety issues? Predictive analytics help. Similar to sports, self-reporting through survey data has its advantages. In their DevOps Metrics article, Mik Kersten and Nicole Forsgren state, “Research shows that good organizational cultures drive software delivery and organizational performance, and job satisfaction drives revenues. Monitoring these proactively (through survey data) and not just reactively (through turnover metrics in HR databases) should be a priority for all technical managers and executives.”
Understand the systems
A view of the whole system looks at everything influencing performance and increases the understanding of cause and effect. Injured soccer players can’t outrun stronger opponents.
Like football, tech organizations are complex sociotechnical systems with competing functions and relationships between individual components. There are multiple interacting human and non-human components operating within a dynamic and constantly changing environment. Traditional performance measures typically fail to consider this complexity, and instead often focus on components in isolation (such as passing in football or uptime in tech).
It does Leicester no good for the right-back to kick the ball to his most talented right-midfielder if the latter isn’t available. Tech teams who optimize locally and impact other teams negatively do not improve performance at the organizational level. Flow time is only as fast as the slowest moving part. Therefore, we must consider all the factors within the system that influence organizational fitness (workload or WIP, communication, policies, flow, dependencies, feedback and safety).
Winning sports teams and high-performing tech organizations invest in tools and analytics to visualize and better understand performance influencers and detractors from a systems perspective. As a result, they are in a better position to blow the competition away.
The game is changing
Data analytics is having a huge impact on how sports teams perform, as Leicester has proven through its cross-departmental relationships, a willingness to embrace new styles of working, and constant evaluation and feedback process.
Lean product management practices and measures help tech teams ship features that customers want more frequently. This faster delivery time enables a faster feedback loop with customers. The result? The entire organization benefits, as measured by profitability, productivity and market share.
While Leicester was unable to retain the crown the following season, the club has cemented its position in one of the most lucrative sports league in the world. And thanks to its progressive approach to sports performance analytics, it will likely remain competitive with its most powerful competitors. Leicester shook things up and innovated with a whole system approach. All tech organizations should take note.