The Book

Good Strategy/Bad Strategy: The Difference and Why it Matters is a book by Richard Rumelt. It is very very good. In it, Rumelt explains his "kernel of good strategy", which is the three core items a good strategy must contain:

• Diagnosis
• Guiding policy
• Coherent action

I couldn't recommend the book enough, it's a really enlightening look at what good strategy is and, possibly more importantly, what bad strategy is. There are plenty of summaries online if you want to get the general gist; Anna Shipman has a really succint one here, and Camille Fournier has a really good article that uses elements of the kernel here.

Whenever I'm trying to make sense of a new concept, like I was reading Good Strategy/Bad Strategy, I'll try to translate the ideas into some mental model I understand better. I think if I feel more comfortable with the surrounding context of a new idea, I'm able to process the idea itself more easily. For me, that's almost always football.

So. Welcome to the kernel of good strategy, if Richard Rumelt was mad into football.

Setting the Scene

There are infinite ways to win a football match, so players need a coherent strategy to limit the universe of options down to a set of actions they can implement. The diagnosis informs the players of the problem they are trying to solve. For a team that is defensively very strong, but draw too many games 0-0, an initial diagnosis might be that they don’t score enough goals. A diagnosis should provide clarity, which this initial diagnosis does not. Part of the job of crafting strategy, then, is to identify a diagnosis that provides a clear problem to solve.

Diagnosis

If we look closer at this defensively strong team, we can identify that their defenders are strong, experienced, and efficient at preventing the opposition from getting clean chances to score, but when they actually have the ball, they are weak at passing out of defence. This additional insight allows us to identify that one of the reasons the team struggles to score goals is because their defenders are not skilled enough in possession to start attacks once they win the ball back from the other team. With this diagnosis, we can start to consider a guiding policy (what Anna Shipman calls a “vision”) to overcome the diagnosed problem. This guiding policy does not tell us exactly how we’ll solve the problem, but informs us of the overall approach and should give a guide against which to consider whether the concrete actions decided upon actually contribute to solution the team is trying to implement.

Guiding Policy

As in most situations, the team could employ a number of different guiding policies to solve their diagnosed problem of not being able to start attacks once their defence wins the ball. Some potential approaches could be:

• Get better at passing the ball out from defence
• Win the ball back further up the pitch so it's the more attacking players starting the attacks
• Stop trying to pass the ball out of defence to start attacks, and instead get the ball up the pitch as quickly as possible

There are pros and cons of each of these policies.

Getting better at passing the ball out of defence would require either spending time improving their existing players so that they can pass as well as required, or bringing in new players that can perform this role. Improving the existing players could pay dividends in the long term as they would become more versatile members of the team (and probably more valuable assets), but the time and focus required to train them would take time away from working on the existing skills and tactics that are successful in making the team difficult to beat. Bringing in new players with the desired skillset might be quicker, but could be expensive and take time for the new players to gel with the rest of the team.

Winning the ball further up the pitch would be useful as it would mean starting attacks from closer to the opposition’s goal, and typically means the opposition has less time to re-organise after losing the ball. However, examining the existing defenders unveils that while they are strong and experienced, they lack pace, so trying to win the ball further up the pitch would leave them very vulnerable against teams with fast attackers.

Finally, abandoning the attempts to pass the ball out of defence completely could have downsides of upsetting the fans (as it is seen as an unattractive style of play) and the midfielders (as instead of trying to receive the ball from the defenders, the defenders might now be skipping them completely by passing directly to the attackers). It would, however, allow the defenders to keep concentrating on the thing they do well, and, after all, if they do their job right then the worst outcome is still a draw rather than a loss.

A good guiding policy acknowledges that picking one approach means making an active decision not to follow alternatives. It should provide enough supporting reasoning so that team members who favoured another approach can see why it was picked, and hopefully convince them that focusing on this policy will still achieve the goal of solving the shared problem identified in the diagnosis. It is decided that the third option will be selected. While everyone agrees that the team need to score more goals, the defensive strength of the team is too important to risk on options one and two. This may upset the fans and some of the players, but the expectation is that improving the actual results will improve morale enough that the fans will accept a less attractive style of play and the players will accept less interesting tactics, if it means they’re involved in an overall more successful team. With a guiding policy decided, the team must decide on a set of coherent actions that implement the policy.

Coherent Action

Potential actions such as purchasing more skilful defenders or pressing higher up the pitch are now discarded as they don’t contribute to the guiding policy. Instead the team decides on the following actions:

• Defend from deeper
• Allow the opposition the keep the ball for longer
• Replace one of the attack-minded midfielders with a more defensive player
• Promote a very speedy under 21s player to the senior squad

By defending closer to their own goal and allowing the opposition to keep the ball for longer before trying to win it back, the team is increasing their risk of the opposition creating chances to score. The calculated gamble the team is taking with these actions is that by allowing their existing defence to keep focusing on what they’re good at, rather than replacing them or taking focus away to train them to be better passers of the ball, they will still be good enough to stop most opposition attacks.

To help mitigate this risk, an additional defensive player is brought into the midfield at the expense of a more skilful attacking midfielder. Again, the team is happy to make this change as the guiding policy they have bought into says that they will be trying to skip the midfield, rather than pass to it, when they win the ball back. With these changes, the team is risking allowing the opposition more time to come forward in the assumption that this will leave more space in which to counter-attack when they win the ball. The new counter-attacking tactic implements the guiding policy as it requires moving the ball as quickly as possible from defence to attack.

To increase the effectiveness of this tactic, the final action involves promoting a new player to the team. This player was not previously deemed good enough to join the team yet, but their attributes now suit the guiding policy. While they are not as skilled a passer or goalscorer as other members of the team, they are extremely fast. The team now anticipates that the new tactics mean this player won’t need to be as skilled a passer, as they will be expected to run onto long passes out of defence over the top of the opposition. This should mean they will have easier chances to score one-on-one against the goalkeeper, rather than needing to find space to score against a crowded defence, so they do not need to be as good a finisher.

So

This (utterly contrived, of course) example hopefully shows the value of this structure for strategy. In identifying a specific and actionable diagnosis, using a guiding policy to get buy in on disregarding potential approaches in favour of a more promising one, and providing coherent actions to everyone implementing the strategy, they were able to address their problem in a clear and structured way. This strategy allowed them to:

• Build on the strength of their existing defence, by deciding against policies that might have weakened it
• Focus their whole team on a single tactic, by doubling-down on skillsets around the counter-attacking idea
• Impose disproportionate costs on their opponents, by making them consider sacrificing their better players to deal with this teams fast attacker

Last Thing

I'm not sure how much material I'll end up getting out of this, but I'm going to try to keep this going along the general theme of expressing ideas I come across in work through the thing I think about most often, football. I'm working with the title Creativity & Discipline, a nod to John Green describing football (soccer to him, of course) as the perfect balance of creativity and discipline (I can't find that quote, if someone can find the video it came up in please send me a link!), something I've always also thought rings through in software engineering.