The goal of everyone involved in sport should be to enhance the performance of the athlete during competition. To that end, we undertake a variety of approaches to place load and stress on the athlete, with the understanding that the provision of this load, in the short term, acts as a stressor. This stressor requires a response from the body, allowing it to adapt and overcome, with the end result being an increase in physiological capacity. If we’re lucky, this will lead to an improvement in performance (although it is worth pointing out that physical improvements don’t always correlate with performance improvements).
There is a fine balance between load and adaptation: too much load can lead to maladaptation, where the athlete becomes over-fatigued and under-recovered, reducing their performance and increasing their injury risk. We don’t want this. On the other hand, if the load is too low, there will be too little stimulus for adaptation and, as a result, the athlete will not elicit any physical improvements from training. We don’t want this either, although arguably it is better than too much load, as at least in this state the athlete has the physical capacity to still compete while fresh.
The Measures of Monitoring
As the standard of sport performance increases, the importance of efficient and well-balanced training becomes paramount. To that end, over the last five years or so, and aided by technological advances, coaches and support staff have become more focused on monitoring the training process. Overall, this monitoring has two main goals:
- Determine when the athlete is able to tolerate load.
- Determine when the athlete is unable to tolerate load (and may therefore get injured).
As such, we can think of athletes as being in an optimally adaptive or sub-optimally adaptive state. A host of factors affect this—genetics, nutrition, stress—such that no athlete is likely in the identical adaptive state as their teammate. Even if we had two identical twins, it’s likely that one had slightly higher quality sleep the night before, or had a slightly easier training session a week ago. For a coach, being able to measure this is important, as it determines how much load will elicit the required stimulus: For twin A, a heavy lifting session might be ideal for that point in time, but for twin B, that same lifting session might be slightly too much.
This isn’t news to coaches: In a recent study, 67% of high-level rugby coaches rated monitoring the training load as “very important,” and 29% rated it as “important” —leaving only 4% who didn’t recognize a need for it. There are many different ways to monitor training load, with the optimal one changing based on available equipment and the needs of the performance staff. Common methods include just the quantification of training load, which we can achieve by monitoring the amount of time each player trains for (in total minutes, number of sessions, or number of drills), which is a very crude measure.
We can enhance this by monitoring something referred to as session rating of perceived exertion (sRPE), whereby the players score how hard they found the session (usually out of 10), which we can then multiply by the length of the session in minutes to get a workload score. This is a really useful metric, because now it allows us to take into consideration athlete fitness and fatigue. For example, the more fatigued the athlete, the harder they will find the session, leading to an increased sRPE score. The fitter (and fresher) the athlete, the easier they will find the session, leading to a decreased sRPE score. As such, this process allows individual variation to become apparent, which can be incredibly useful.
Fortunately, subjective measures, such as sRPE, have been shown to be as effective at monitoring athlete wellness as more objective (and usually expensive) measures. There is a caveat to this, however, and that is that sRPE is only valid if athletes give accurate and reliable scores. This isn’t always the case; if they have to give their scores in front of teammates, they might attempt to act macho and lie. Similarly, if the last portion of the session is harder than the rest, this can bias the athlete’s perception of how hard the session was, inflating the scores (one way around this is to wait until 15-20 minutes after the session to collect the RPE data).
We can then monitor training load metrics, including sRPE, over a period of time. Tim Gabbett is one of the eminent researchers in this area. Gabbett popularized the acute:chronic workload score, with the idea being that training load data (primarily sRPE x training time) could be collected to give a four-week rolling average picture of the “chronic” workload, and this can be compared to the weekly (or shorter) “acute” workloads. Large increases or decreases in this acute workload relative to the chronic, standard load the athlete is used to, are believed to increase the risk of injury, something which is again supported by Gabbett’s research.
Subsequent to this, there has been some discussion regarding the best method to calculate the acute:chronic workload, with some researchers pointing out a potential issue with Gabbett’s methods. Because fitness isn’t static, but can increase and/or decrease over time, the use of chronic workloads can be slightly misleading. As an example, I might be in a training taper, which means that if the taper goes on for long enough, I will become less fit. As a result, any training I do acutely will be more taxing than the acute:chronic workload score might suggest—in this case, the researchers recommend using an exponentially weighted moving average, which the research has shown to be effective.
Developing a Monitoring Program
So, what does the ideal monitoring protocol look like? Again, this depends on each person’s unique situation, but the best protocol is one that athletes can follow and that gives usable information. There is no point having the best protocol possible if it takes an hour to complete and you don’t get the results for two weeks. Athletes won’t want to waste time submitting the data, and the information garnered won’t inform the training process, due to the processing delay.The immediacy of feedback is the most important characteristic of any monitoring protocol. Click To Tweet
Returning to the study on rugby coaches, it determined that a maximum of 10 minutes spent monitoring fatigue and recovery per session was appropriate. The immediacy of feedback was deemed to be the most important characteristic of any monitoring protocol, for reasons mentioned previously. The coaches also felt that the monitoring protocol should be both inexpensive—no fancy equipment here—and easy to administer, potentially ruling out blood tests. Other important aspects were that any test should be able to have the entire team complete it at the same time, and the test should be non-fatiguing.
Alongside wellness data and training load data, it makes sense to also collect some performance data. The reason is perhaps obvious: Performance is the metric that matters, and yet we have the potential to lose sight of that if we just focus on wellness and training load metrics. By collecting performance data, we get an idea of how the athlete is responding to training—if their scores are improving, they are tolerating the load well and getting fitter; if their scores aren’t improving, then they are either in an intensified training phase (which is fine, as long as it is planned), or they are not tolerating the training load adequately, and as such are in a maladaptive state, increasing the risk of injury.
Performance is the metric that matters, and yet we have the potential to lose sight of that if we just focus on wellness and training load metrics.
Most coaches conduct performance testing at regular, but widely spaced, intervals; perhaps every month or two. While this allows for the collection of useful performance data for comparison to older data, it doesn’t allow for the making of rapid adjustments. For this, we need more frequent testing data. This creates a problem; while we need more frequent performance data in order to better adjust the training load, we can’t put in maximal testing sessions on a weekly basis because they interfere with training too much. A really good way around this is the use of sub-maximal tests on a semi-regular basis, usually in the warm-up. This approach is perhaps ideal, because the sub-maximal aspect of it means that the test doesn’t interfere with training too much, and shouldn’t be affected to a large extent by motivation.
Sub-Maximal Tests for Performance Monitoring
The sub-maximal tests you might want to use depend on the athletes you’re working with. For team sport or aerobic athletes, a useful test is the 5-5 football test, well studied by Martin Buchheit. In this test, they run for five minutes at 9 km/h with a heart rate monitor, and you collect their average heart rate for the final 30 seconds of the test and after 60 seconds of recovery (to give a score of heart rate recovery). If their sub-maximal heart rate lowers with training, then they’re getting improvements in aerobic fitness; if it increases, they’re perhaps ill or excessively fatigued. The same is true for heart rate recovery; if that goes down, they’re getting fitter. The best part about this test is that it doesn’t require a warm-up beforehand, and so it can function as a warm-up itself.
For monitoring of speed-power metrics, there are a number of different options. One is a six-second maximum cycle sprint, followed by one-minute recovery, followed by an additional six-second max sprint; in this test, you monitor peak power output. This has been shown to be a valid measure of neuromuscular fatigue in Australian Rules footballers.
A second potential measure is that of counter-movement jump height (CMJ). While this often requires expensive equipment (such as an Optojump), you can make a poor-man’s version by just having a measuring tape by the wall and seeing how high the athletes jump. There is also the option of an iPhone/iPad app called myJump, which the scientific literature has shown to be valid. When it comes to CMJ, average height is more sensitive to neuromuscular fatigue than greatest height, so it makes sense to do three to six CMJs during the latter stages of a warm-up.
A final option is that of bar speed during lifting movements, which many coaches utilize. Again, this could be sub-maximal during the warm-up—for example, a set of six hang snatches at 50% 1RM—or could just take place naturally during training, as athletes tend to lift similar loads session to session. If an athlete’s velocity at a given load is significantly lower on a given day, then they are potentially struggling with the overall training load.
A key consideration is athlete compliance. Research indicates that it is crucial that any monitoring of wellness doesn’t take a long time, and is easy for athletes to carry out. The more barriers that an athlete faces in the delivery of wellness data, the less likely you are to get consistent information. If using a questionnaire, athletes should ideally be able to complete it in around 60 seconds. An additional consideration is the ease with which you can log and record this data; if you have paper questionnaires for 30 squad members, you need to input that data into a spreadsheet. This may or may not be a worthwhile use of your time, so other options are apps that record to a centralized database that you can access, although cost becomes an issue here.
Putting the Pieces Together
Perhaps the best practical paper on athlete monitoring comes from Gabbett himself, along with other high-level authors. They published their ideas in an editorial from mid-2017 in the British Journal of Sports Medicine. Their guidelines are:
- Determine what you want to achieve through the monitoring process
- As already discussed, this is likely going to the monitoring of improvements in training, the effects of fatigue from training, and a reduction in performance; all of which contribute to the end goal of enhancing athlete performance.
- When it comes to monitoring performance, select fitness tests that are relevant to your sport. You wouldn’t give sprinters an aerobic fitness test, for example.
- For external training load, decide on relevant metrics. If you’re a runner, this might be total training distance multiplied by intensity.
- This can be as in-depth as you require; you might want to collect blood after training to determine hormonal status, although this is likely overkill for most.
- Make sure you collect the data in a reliable and valid format; attempt to keep conditions the same.
- There is no point in collecting data if you don’t use it to inform your decisions, so this is the crucial step. Without wanting to get too deep into statistics, for wellness data, you likely want to use standard deviations of z-scores. For fitness testing data, you have a number of options, but the smallest worthwhile change metric is perhaps the most important. If you’re not sure what this is, Anthony Turner has the best video I’ve seen on it, and this website is also an excellent resource.
Let’s examine this through the hypothetical situation of a sprinter I’m coaching. I’ve decided that I want to collect daily wellness data before the training session, to assess the athlete’s “readiness to train.” I do this via a questionnaire that takes about 60 seconds to complete, and allows me to understand how well she has slept, how tired and sore she is, and whether she is ill. Any large deviations from her “normal” scores (and I can determine what normal is by calculating monthly averages, for example) acts as a red flag to me; I can speak to the athlete to see what’s going on, and possibly change the training session accordingly.
A few times per week I could also implement some sub-maximal tests, the type of which we discussed earlier, to allow me to get some objective data on how well the athlete is tolerating load; if her scores indicate fatigue, and this level of fatigue is unplanned (i.e., isn’t a deliberate training variation such as functional overreaching), then, again, I can modify training. (Once more, the more data I collect, the more robust my baseline data becomes, so I can better detect deviations from normal.)
After sprint-based sessions, I could create my own modified sRPE metric by asking the athlete to rate how hard the session was out of 10, and multiplying this by the distance covered. (In the gym, I could just multiply sets x reps x weight for a volume score). Over time, I can set this up in my spreadsheet to determine the acute:chronic workload, which then allows me to tell if the athlete is under- or over-cooked, or just right.
Finally, at set points during the year, perhaps every six weeks or so, I can insert some specific performance tests; perhaps a speed test (60m from blocks), a strength test (back squat 1RM), and a power test (standing long jump). Overall, this package allows me to make small daily variations in training volume and intensity where required, guard against injury, and then determine whether the athlete is responding to training by showing improvements in testing.
Make Monitoring Part of the Training Process
In summary, athlete wellness, load, and performance monitoring is a worthwhile addition to the training process, possibly protecting against overtraining syndrome, illness, and injury, and allowing for on-the-fly modifications to the training program. While this can be expensive and time-consuming, it doesn’t have to be: Simple 60-second questionnaires, logging of total session workloads, and occasional sub-maximal performance tests are likely to be sufficient for most training program’s goals. Ensuring athlete buy-in is also crucial, as the monitoring program is only as good as the data you can collect, and if this data is unreliable, then the system falls down.
- Buchheit, Martin. “Monitoring Training Status with HR Measures: Do All Roads Lead to Rome?” Frontiers in Physiology. 2014: 5(73).
- Halson, Shona L. “Monitoring Training Load to Understand Fatigue in Athletes,” Sports Medicine. November 2014: 44(2); 139-147.
- Turner, Anthony N.; Bishop, Chris; Springham, Matt; and Stewart, Perry F. “Identifying readiness to train: when to push and when to pull.” Professional Strength & Conditioning. 2016: (42); 9-14.