Agile Periodization Manifesto [Part 2]

In the previous installment I explained the concept of Agile periodization as opposed to “traditional” one (i.e waterfall) and covered one of the two main cycles – Training block.

In this one I will cover the second major cycle called “Sprint”, it’s sub-components and two major principles:

• Embedded testing
• Minimal viable product

Sprint

“Micro dictates Macro”

Sprints are shorter cycles and usually last 1-4 weeks. The term sprints come from SCRUM methodology.

It is important to acknowledge that ONLY ONE sprint is planned in advance (in detail). This is in sharp contrast to traditional planning: in Agile Periodization planning is ongoing through sprint iterations, while in the “traditional” approach planning is just one stage that happens once in the beginning.

Using sprints we iterate through PDCA cycle and adjust (we adjust the apriori standardized work) based on observations (new evidence and information) through embedded testing, demonstrable MVP, better understanding of the context and problems at hand and engaging with the players themselves and massively using their feedback in improving the processes.

Each sprint has couple of elements:

1. Sprint planning – well, deciding on the duration, objectives, standardized work defined by constraint of the training block and so forth
2. Realization and monitoring – using embedded testing to adjust processes instead of outcomes (see Management by Means vs. Management by Results). This also involves daily standup meetings – very short meetings to realign the team to the defined objectives
3. Sprint review – Reviewing results/outcomes of each sprint
4. Sprint retrospective – Reviewing sprint processes and trying to improve them (kaizen)

There are two very important concepts in sprints – embedded testing and MVP.

Embedded testing

Embedded testing involves utilizing ongoing testing and monitoring of the athletes. This is in sharp contrast to the traditional approach where testing is done in batches and infrequently. In Agile Periodization, we seek to make tests embedded in the training process, do them as frequently as possible without negatively affecting the training program and process.

With the traditional approach, testing is done 1-2 times in pre-season and 1-2 times in-season. The problem with this approach is that we are not able to use the information gained by the test to affect our planning and modifications – it is already too late when we do the tests.

It is also important to make tests more prescriptive instead of sport-specific. The tests should give us actionable data and help us with identifying bottlenecks (limiting factors) and hence help with the planning. Performing sport-specific tests is also usable (especially when we track them for RTP protocols and we seek to hit certain thresholds), but since their scores depend on multiple factors, we cannot be sure what needs to be enhanced for the performance to improve. More general tests can give us this information. Take for example Yoyo test – changes can happen due to changes in CV (critical velocity or aerobic capacity), W’ (anaerobic reserve), intra-rep pause recovery, COD efficiency, mental toughness and so forth.

As coaches, we need to decide how deep down the rabbit hole we need to go with testing and decide what is enough to make better decisions without too much burden and paralysis by analysis.

With the new technology embedded testing is more easily doable, but it is still a pipe dream. For some performance indices, we can use proxies from day-to-day training and we should use the MVP concept to test more frequently (as a source of information, as well as accountability) at the end of each sprint for things we cannot approximate using the embedded testing idea.

Here are some examples of embedded testing for certain training components.

 
This is indeed a concept I need to expand further in the weeks to come. Please make sure to read THIS for more info about the topic.

Minimum Viable Product/Performance

MVP means we need to create a demonstrable performance at the end of each sprint. There are multiple reasons for this.

The first one is that we are held accountable for it and represents a gut check. When we know we need to deliver, it will keep us on the plan and we are less likely to f*ck around.

Second, it represents a tremendous source of information and evidence. Why wait till the end of pre-season? Then we cannot utilize this information to correct our planning.

Third, it represents some of the most intense stimuli, especially if it is done in a competitive environment and it should be. I recently wrote an article about decentralizing training program and splitting teams in sub-teams to make a competitive environment even in team sports.

This doesn’t mean by any means we are neglecting long-term development for the sake of short-term demonstrable performance. That’s why we have training blocks and that’s why we call this performance minimum viable.

And yes, this goes a bit against Verkhoshansky LTDTE (Long Term Delayed Training Effect). In contemporary sports, we don’t have time for the long planning cycles and due job insecurities, and we need to constantly deliver demonstrable performance to the fans. The nature of team sport also negates this approach due short preparatory period and longer and longer competition period. Risks associated with LTDTE and functional-overreaching approaches also need to be taken into account. There is more and more research questioning this type of approach (for example THIS one, as well as great presentation by Yann Le Meur pp74-84).

Managing training load for sport performance [le meur madrid 2014] from Yann LE MEUR (@YLMSportScience)
 

Please make note that this does NOT mean avoiding concentrated and overload periods – it just means avoiding functional over-reaching and putting emphasis on (almost) always being able to demonstrate MVP.

Another interesting example I like to use is Boxing. One can spend weeks and months perfecting technical elements of striking and then stepping into the ring to spar and realizing the perceptual/decision making components are completely underdeveloped. This is the opposite of MVP approach. MVP approach would be having working technique and embracing sparring early on – mostly very directed and constrained sparring and flow work. This also serves as a tremendous source of info in identifying limiting factors. Risking by waiting too long to demonstrate performance is not the best way. As the saying goes, “If you are going to fail, fail early and cheap”.

Demonstrating MVP at the end of each sprint does not mean being in crazy competitive situations and burning out – it just means being accountable in demonstrating something (minimum viable). Injecting meets, friendly games, sparring and other easier competitions early on is a great way to create accountability to performance, gain valuable and actionable information and providing the most specific stimuli. And failing early, rather than too late.

For example, MVP in strength training could be doing AMRAP sets (open sets) or OTS (Off The Script by Joe Kenn) or Joker Sets (Jim Wendler) besides utilizing embedded testing with LPTs and/or RPEs. The athletes should not go balls to the wall, but they need to demonstrate something that could be used as a gut check for the next sprint (to help with planning by providing information and evidence). Here is an example of embedded testing and MVP with hex bar squats.

Here is another article I wrote a year back on the topic of Horizontal vs. Vertical Variation in strength training that resonates with this topic as well.

I hope this short “manifesto” got you some ideas and principles that can make your planning more “antifragile”. This is work in progress.

Feel free to leave your thoughts in the comments section at the bottom of the page.