Mind The Gap: Internal Training Load During Practice vs. Competition

This article is written by Domenik Theodorou, a Firstbeat Sports client and Performance Coach at RASTA Vechta, a professional basketball team competing in the Basketball Bundesliga (BBL). Prior to his current role, Domenik also worked at Bamberg Baskets and BG Göttingen. In this article, Domenik shares practical advice and tips on using Firstbeat to help identify gaps of internal training load during practice and competition.

In this blog post, I’ll dive into the differences in how internal training load accumulates during training versus competition, sharing some of my own anecdotal observations. I’ve lately noticed a significant gap that I hadn’t been fully aware of before.

Internal training load during practice

In a previous blog post, I explained how I use the Firstbeat system to track fitness trends in our players over time. When comparing team practices from the early weeks of pre-season to the start of the regular season (6-7 weeks later), it’s common to see lower TRIMP scores with similar Movement Load (ML) values, assuming athletes remain healthy during this period.

The gap between TRIMP & ML increases, which is a highly desirable outcome. Additionally, the time athletes spend above 90% of their maximum heart rate during training decreases, and EPOC-Peak no longer reaches previous levels.

From an internal training load perspective, everything seems to level out a bit—basketball practice comes with a smaller “price tag” as athletes have improved their fitness. However, no two practices are ever the same. If a session features a significant amount of full-court play, the internal training load will be higher compared to predominantly half-court live drills. Still, it will rarely reach levels seen at the beginning of pre-season.

Internal training load during competition

Before addressing internal load during competition, it’s important to note that our players rarely use the Firstbeat system during games. We prioritize letting athletes focus fully on the match without any potential distractions, in line with the club’s policy. However, exceptions are made when necessary or when players voluntarily choose to wear their sensors during games.

At our club, all players from the first team consistently wear the Firstbeat sensor during every basketball-related activity, except for games and lifting sessions. This includes every team practice, individual workouts, return-to-play sessions, and even casual shooting on off-days.

By doing so, I can track all on-court activity and gain a comprehensive understanding of each athlete’s Acute Training Load. This continuous monitoring allows for more informed decisions and better management of their workload. For games and lifting sessions I add manual inputs to keep the load accurate.

For our second professional team, only the top prospects are equipped with a Firstbeat sensor, which they are required to wear just like the first team. All the training data is overseen by me.

The following example is about one of our young prospects who played for three different teams during season 2023/24. He was part of our second professional team in the Pro A (second highest level in Germany), the NBBL team (under 19 years old, highest level in Germany) and sometimes he even played for our third team (2. Regionalliga, fifth highest level in Germany).

Most of the times he took part in two games over the weekend. The demands on the body were quite high as you can probably imagine. He had a small role on the Pro A team but a very large role on the NBBL team. If he didn’t see a lot of minutes during a Pro A game or wasn’t even on the roster we decided to give him more playing time on our third team if the schedule allowed it.

Internal training load during practice vs. competition

Table 1 provides an overview of approximately 1.5 months of training load during the competitive season, specifically covering weeks 19 to 25. The pre-season spanned 6 weeks, with competition beginning in week 7.

All entries with a white background represent regular team practices, while those with an orange background indicate games, and a green background show off-days. The key metrics I track include TRIMP, Movement Load (ML), Movement Efficiency (MI), EPOC Peak, minutes spent above 90% of HRmax (HIT in MIN), Acute Training Load (ATL), Acute to Chronic Workload Ratio (ATWR), and the total TRIMP (TRIMP T) for each day.

If you study table 1, and only look at the practice loads of that particular player you can get a pretty good idea of his physiological fingerprint. The average loads at the bottom of the table for game day -4 (GD-4), game day -3 (GD-3), game day -2 (GD-2), game day -1 (GD-1) and games are only for the timeframe shown on the table. These examples reflect how we are trying to set up our microcycle to execute a training week. GD-4 and GD-3 are meant to be our most intense training sessions, while GD-2 and GD-1 show a clear a reduction in training load.

By closely analyzing the athlete’s movement efficiency scores, it’s evident that he is in excellent physical condition. This is clearly shown by the significant gap (>2) between his TRIMP and ML values. For more insights into my personal experience with movement efficiency, feel free to explore one of my former blog posts.

I would like to focus particularly on the training data from the games on January 6th, January 7th, and January 28th, as these were the only games listed in Table 1 where the athlete was wearing his sensor. For the other games, I manually estimated his TRIMP scores based on his playing time in each match. While this method isn’t perfect, it still provides a valuable overview and maximizes the accuracy of long-term training data.

I did not apply a specific formula to determine the TRIMP scores after games but rather relied on my personal insights into how each player accumulates training load during practice. When two players have the same playing time, I avoid assigning identical TRIMP scores because each player may accumulate TRIMP at a different rate. I believe it’s essential to trust your instincts as a coach in these situations, even if the method isn’t flawless—it’s still effective for understanding overall training loads.

On January 6th, the player wasn’t selected for the Pro A team, so we decided to have him play with our third team to maximize his playing time. I asked if he’d be willing to wear the sensor during the game, as I wanted to analyze how the demands of games compare to practices. The athlete was eager to cooperate, as he was also curious about his training load during competition.

After the game, I reviewed his data and was shocked by the significant difference in internal load compared to what he typically generates during practices. While I anticipated a disparity, the extent of it was far beyond my expectations. What made it even more surprising was the context: he played just 22 minutes in a game we won by 40 points, with only about 30 people in the audience.

The game on January 7th was an away match, which we lost by 13 points. In this game, the player played 26min and his internal peak values were even higher than the previous day. This could likely be attributed to the fact that he had to play back-to-back games, adding to the overall physical strain.

The last game I tracked on January 28th was a home game, and the internal load surpassed the previous two examples. It was a rivalry match against a direct contender, played in front of 100 spectators, which we narrowly lost. The player played 34 minutes.

I understand this is a small sample size, but whenever I’ve selectively equipped a professional player with a heart rate strap during competition, the results have been consistent: a significant disparity between internal training load during games versus practices.

Why is internal training load so much higher in competition?

When I first reviewed the internal training data from January 6th, I was surprised by the significant gap between training and competition load. The margin was larger than expected, and I found it hard to believe at first. Curious about the potential causes, I reached out to several colleagues to see if they had encountered similar findings. I also dove into various pieces of literature and scientific papers, which largely confirmed that competition loads tend to elicit heart rate (HR) responses exceeding 85% of HRmax—something we might expect. Firstbeat’s data also confirms that basketball and rugby lead all sports in the amount of time spent in high-intensity training zones during competition

However, none of the sources provided a clear explanation for the striking gap I observed. In the following, I’ll offer my own perspective on this issue, which likely falls into the “common sense” category.

Environmental factors

The competitive environment presents numerous challenges that are difficult to replicate in team practices. Factors like audience, loud music, unfamiliar arenas, hostile atmospheres, elevated temperatures, visual distractions or unknown opponents all heighten the players’ stress and demands. This, in turn, leads to increased arousal and higher heart rates. I firmly believe that each athlete responds to these factors in their own unique way—some likely managing them better than others.

Psychological aspects of competition

In professional sports, the outcome of a game is shaped by a complex interplay of internal and external expectations. Internal expectations stem from the pressure athletes place on themselves to excel, while external expectations arise from coaches, teammates, fans, and even family. Every athlete is driven to perform at their best, which inherently increases the pressure.

This intensity escalates during competition, where mistakes carry far greater consequences. A turnover in practice is just a learning opportunity—a mistake to correct and move on from. But in a game, a turnover has a much heavier toll, especially mentally, as it unfolds in front of coaches, teammates, opponents, and spectators, leaving the player exposed in a way practice never does. The stakes feel higher, amplifying both the pressure and the emotional impact. At the same time, we cannot neglect the impact of positive emotions, like celebration, cheering, hope and team cohesion which will also lead to increased heart rates.

Today, everything in sports is under a global spotlight. Games are recorded, broadcasted live, and shared across the world. Statistics are readily available, leaving no room to hide. Every performance is on full display, scrutinized from every angle, and it either opens the door to your next opportunity—or it closes it. Athletes are constantly under the microscope, with each game contributing to the narrative that shapes their future.

The anticipation of competition, combined with the physical demands of basketball, along with the environmental and psychological factors, will also elevate adrenaline levels and contribute to higher heart rates.

Are these factors sufficient to explain the notable gap between internal training load  during practice and competition? At least, I would argue that it’s impossible to simulate these aspects to the same extent during team practices.

How does basketball practice differ from competition?

Practice sessions often face interruptions for technical and tactical coaching. Players are generally given opportunities to request substitutions, while scheduled water breaks help maintain structure. The intensity demands vary for each player based on their position and the specific requirements of the drill.  Shooting and half-court drills don’t even come close to game demands.

In my experience, 5v5, 4v4 and 3v3 full-court play tend to generate the highest internal load, but this is only achieved without constant disruptions; otherwise, the breaks between plays become too significant.

It raises the question of whether training can truly match the internal demands of competition. While practice can replicate many of the external demands of basketball—though perhaps not perfectly—mirroring the internal load is more challenging. The internal load challenges of competition often present demands that are difficult to replicate in a controlled training environment.

What are the consequences for practice? How hard do we really need to train? If the goal is to replicate the demands of a game, the practice scenarios in Figure 1 suggest that our internal training load—particularly EPOC Peak and the minutes spent above 90% of HRmax—may be insufficient. However, a key question arises: are we even capable of consistently training at the levels that our competition data demonstrates? Increasing the amount of full-court play while minimizing coaching interruptions could be an effective way to elevate internal load levels during training. This approach creates a more continuous, game-like environment, which can help better simulate the physiological demands of competition.

In my view, the unique environmental and psychological factors present only in games elevate internal demands to extraordinary heights, creating peaks that are difficult to replicate in training settings. I believe it is essential for us as coaches to be mindful of this. Nevertheless, it remains important to expose athletes to well-structured, intense training sessions that aim to maximize their preparedness as much as possible.

In light of this blog post, think about the importance of getting game repetitions one more time.

I am always happy to connect and chat about S&C work. Please send me a message on Instagram @domceps if you would like to connect.

If you’re interested in learning more about how Domenik uses Firstbeat, check out our free webinar recording – Mastering Data Driven Decision Making in Pro Basketball