Investigating the internal and external load output across different drill classifications in semi-elite male Australian football players

Abstract

Player and training load is constantly monitored during most team sports training, especially in intense sports such as Australian Football (AF) to ensure players are not over or under training. However, limited research has explored how drill types can influence the internal and external load placed on athletes. This observational study explored how semi-elite male AF players’ (n = 17) internal and external load output is influenced by different drill types. Participants completed preseason training sessions wearing Global Positioning Systems (GPS) and Heart Rate (HR) monitors and gave differential Ratings of Perceived Exertion (RPE) in relation to breathlessness (RPE-B) and leg-muscle exertion (RPE-M) at the conclusion of every drill. Data were analysed using linear mixed effects regression models, and pairwise comparisons between each drill type were conducted using Sidak's post hoc correction. Fundamental and ball movement drills elicited the lowest absolute and relative internal and external load values when compared to conditioning, small-sided games (SSG) and match simulation drills. Match simulation elicited the highest absolute external loads, excluding high speed running. Conditioning elicited the highest relative external and internal load outputs, excluding decelerations. Coaches should consider the type of drills they implement across the training week to ensure athletes are achieving the desired physiological adaptations, while also optimising preparation for competition.

Keywords

Global positioning system heart rate monitoring high-speed running match simulation rating of perceived exertion,small-sided games

Introduction

Australian Football (AF) is an intermittent game with short, repeated bouts of high intensity activity interspersed with bouts of lower intensity activity.¹ AF players require well-developed physical qualities to cope with the demands of competition, including aerobic fitness, strength, power, speed and agility.² Precise manipulation of training modality, intensity, and volume to optimise training prescription is vital to produce sufficient overload and facilitate improvements in athletic performance.³

Commonly, training load is captured using objective and subjective data to quantify on-field training demands in the form of external and internal load.³ Objective data is quantifiable from the athletes through wearable microtechnology, and measures of external load represent the physical output of an individual during a session.^4,5 Microtechnology devices including Global Positioning System (GPS) are a commonly utilised objective external load monitoring device, deriving variables of distance, speed, velocity, acceleration, deceleration, and movement repetition counts for individual athletes during trainings and games.^3,4,6 In contrast, subjective data relies on athletes reporting their perceptions of training intensity, and are considered important measures of internal load in conjunction with objective internal load measures such as heart rate (HR), blood lactate, oxygen consumption, rating of perceived exertion (RPE), and session-RPE (sRPE).^4,5 All objective and subjective measures of external and internal load offer a complementary insight into an athlete's well-being and performance.

In most team sport environments, external and internal load metrics are collected at the end of trainings or games to evaluate the demands of a whole session.⁴ While this may accurately represent overall load,^4,6,7 it is likely internal and external training loads in a single session differ across their various components. In AF, trainings are typically constructed using different drills.⁸ While these drills are likely to differ slightly between clubs, they often share similar foci and characteristics that allow them to be categorised into drill typologies. It is important that the impact on external and internal load associated with such drill types is understood as drill selection is a critical part of structuring training to ensure intensity and volume can be precisely manipulated to facilitate adaptation and appropriate load management.

While no previous research has examined the effects of drill classification on measures of external and internal load in AF, some studies have investigated this in other team sports. Research in professional soccer athletes found that HR values obtained during sessions comprised of skill circuit training and small sided games, and ball-possession games, were substantially greater than those sessions focused solely on tactical training.⁹ Similarly, all internal training load markers were substantially greater in skill circuit training and small sided games than any other sessions.⁹ However, this study only examined internal load through HR and RPE and did not account for measures of external load. This study also categorised whole training sessions based on their content, rather than into the specific drills within a training session.⁹ Additional research in professional rugby league compared GPS metrics between playing positions within different drill types, and found that hit-up forwards completed less very-high-speed running than outside backs and adjustables.³ Hit-up forwards also experienced greater relative 2-dimensional bodyload demands than outside backs.³ Internal load was not considered in this study. Although the author outlined different drill types, only four training drill types were identified: speed and agility, conditioning, generic skills and positional skills. The load within each drill was compared between positions, rather than differences in internal and external load being compared between drills.³

No research has compared drill types in AF. Therefore, this research aims to explore how training drill classification impacts the external (GPS) and internal (HR and RPE) loading of AF athletes, providing practitioners with the information necessary to effectively monitor, prescribe and manipulate training demands through training drills to influence player loading. It is likely that small-sided games and match simulation drills would lead to the highest measures of internal load due to their highly contested nature and extensive change of direction demands. Conversely, one would expect conditioning drills to involve the most high-speed running due to their focus on the development of aerobic fitness. Additionally, given the technical aspect of fundamental and ball movement drills, it is likely that the internal and external demand of both these drill types would be lower than that of other drill types. Based on previous literature we expect that drills that are more highly contested will elicit greater internal loading when compared to drills that are not as contested. In comparison, we anticipate that conditioning-based drills will elicit the most high-speed running.

Methods

Pre-registration

This study was prospectively pre-registered on the Open Science Framework (https://osf.io/2gwyu).

Participants

Seventeen semi-elite adult male AF athletes (mean ± SD age 24 ± 2.7 years, height 187 ± 7 cm, and body mass 85 ± 7 kg) participating in the South Australian National Football League (SANFL) competition from one club were recruited. This study was approved by the institutions Human Research Ethics Committees (protocol number 206765). All participants gave written informed consent before participating. To be included, all athletes were required to be currently listed as a SANFL league level athlete above the age of 18 years. Exclusion criteria from this study included: (1) Athletes who were ruled out of participating in training during the 8-week preseason data collection period due to injury (either sustained prior or during the preseason period), and (2) having any underlying medical diagnosis (i.e., heart arrhythmia, etc) that would alter internal measures of load. There were no participant dropouts over the duration of the study.

With the expectation that there would be at least 16 observations per training drill per participant during the preseason period, and assuming a within-participant correlation of 0.7, it was estimated a minimum of 10 participants would be required to detect a medium effect (f – 0.25) with 80% power and an alpha of 0.05. A total of 17 players were included to account for fluctuations in training attendance.

Physical training

Preseason training was undertaken from November 2024 until the start of the competitive season in March 2025, with a four-week break between 13th of December and 10th of January. Therefore, from the start of January, eight weeks of preseason training data was collected. During this time, athletes completed three on-field training sessions per week (Table 1).

Table 1.

Typical weekly training schedule during preseason.

	Monday	Tuesday	Wednesday	Thursday	Friday	Saturday	Sunday
AM		Full body resistance training		Full body resistance training	Optional upper body resistance training	Occasional conditioning based on-field training	Rest day
PM	Fundamentals and speed work On-field training		Main session on-field training		Game awareness on-field training		Rest day

Training drills, were based on previous literature but were defined in training by the head coach and were based on the focus and prescription of the drill.³ All drills were allocated to one of five following training-drill categories. Examples of these drills are provided in supplementary digital content 1:

Fundamentals: Drills designed specifically for skill development and skill repetition were defined as fundamental drills. These fundamental drills allow the performance of the skills without excessive fatigue and commonly involve fewer players and limited contact.

Ball movement: These drills incorporate skills with more game specific movement patterns and situations, while still involving limited opposition pressure and/or limited contact.¹⁸ These drills typically take place either over the full field or half the field and are repeated quickly to minimise stationary time. These drills involved guided ball and player movement, with minimal decision making.

Small-sided games (SSG): These drills allow for the practice of team movement patterns in competitive conditions while under pressure, stress and fatigue.¹⁷ These drills are performed on a reduced field size (e.g., within the 50 m arch, or within a 20 × 10 m rectangle) to increase player density (player numbers ranging between six to twelve per team), thus differing from match simulation. SSG's are considered important as they incorporate physical, tactical, and technical elements of the game that can be practised at training to be used in a game.

Match simulation: Drills that replicate the demands of gameplay, including field size and player numbers. These drills involve decision making opportunities for athletes as they do not have predetermined ball movement patterns and include the addition of opposition players.¹⁸

Conditioning: Drills that only focused on developing the physiological characteristics desirable in AF, including aerobic and anaerobic endurance performance were named conditioning drills.²⁰ Throughout this study, conditioning drills were defined as either long-or-short-interval training drills. Long interval drills had a working period of two or more minutes with a passive rest period of no more than two minutes, and an intensity of less than 95% of Maximum Aerobic Speed (MAS).¹⁰ Short interval drills had a working period ranging from 15 s to 2 min, with a passive rest period between 15 and 90 s, and performed at a higher intensity between 100–120% MAS.¹⁰

Outcome measures

Global positioning system measures

All athletes involved in this study were issued a GPS monitoring unit (STATSports pro series v3.2, STATSports, Ireland), that were placed inside a separate garment designed specifically to fit the GPS unit and positioned so the unit sat between the athlete's scapulae.¹¹ The average satellite count for sessions measured in this study was 14.52 ± 1.56. The average Horizontal Dilution of Precision (HDOP) in this study was 0.58 ± 0.43. Drill periods were captured each time a drill began and ended. If a coach interrupted a drill to provide feedback and/or an educational piece that resulted in athletes being stationary for an extended period (∼>60 s), the drill was excluded from analysis. In the instance where a drill was stopped for ∼<60 s, the data was retained. During these occurrences, the GPS units were paused during drill cessation to ensure stationary periods were not captured.

At the conclusion of each training session, the GPS data was uploaded to Sonra (version 5.0.14, Santiago, Chile) via the GPS devices. The training session data were then exported to Microsoft^® Excel^® (Microsoft 365 MSO [Version 2408 Build 16.0.17928.20336] 64-bit) for analysis, including the variables total distance (km), high-speed running distance (23 + km·h-1), number of accelerations (>2.78 m.s⁻²),^12,13 decelerations (←2.7 m.s⁻²),^12,13 and impacts (>3 G). GPS units of 10 Hertz, show excellent validity (r = 0.91) and reliability (CV = 0.7%) in team sport athletes^11,14

Acceleration and deceleration thresholds were chosen to be consistent with prior research in AF.¹⁵ Research has found that 10 Hz GPS units are able to detect the smallest change during constant velocity and acceleration phase for 1–3 m/s⁻².¹³ To determine number of efforts for accelerations and decelerations, a dwell time of 1 s was used. The impact threshold was selected based on prior research indicating that AF impacts typically correspond to a low-intensity category (compared to heavy collisions observed in rugby league).¹⁶ Thus, impacts in AF have previously been reported to correspond to forces of greater than 3G being recognised as a forceful impact to an athlete.^16,17

Ratings of perceived exertion measures

RPE was collected at the conclusion of each drill within a training session, using the Borg Category-Ratio (CR10) scale,¹⁸ whereby performance staff asked each athlete involved in this study to rate their perceived exertion for that training drill. RPE was collected utilising differential RPE constructs with ratings recorded for, breathlessness (RPE-B) “how puffed do you feel from that drill?”, and leg muscle exertion (RPE-M) “how heavy did that drill make your legs feel?”, players were also supplied with a visual RPE scale to refer to at the conclusion of each drill.^18,19 RPE has been shown to be reliable and have good validity (r = 0.83) when compared to objective measures of load.^20,21

Heart rate measures

HR monitoring device (STATSports HRM 210 s, STATSports, Ireland) were also issued to each player. HR monitoring straps were placed at approximately the fifth intercostal space in the midaxillary line, just below the pectoral muscle, centring the HR unit directly to the sternum.²² All athletes wore the same devices for the duration of the study to limit the possibility of inter-unit variability influencing the results.

HR and Training Impulse (TRIMP) measures were used for analysis. HR data was classified as time spent in each HR zone. The HR zones utilised in this study are expressed as a percentage of max HR (as the peak HR measure obtained during training during the preceding 12 months), and were used to quantify TRIMP in intermittent team sport players using the weighting factors described by Stagno et al., (2007).²³ To calculate TRIMP, time spent in each HR zone was multiplied by the weighting factors described by Stagno et al., (2007) and summed. The HR Zones were categorised as follows: zone 1: 65–71%, zone 2: 72–78%, zone 3: 79–85%, zone 4: 86–92% and zone 5: 93–100%.²³ HR chest strap worn monitors have displayed both excellent validity (r = 0.99) and reliability (CV = 3%).^22,24 There were some instances where HR units dropped out during a specific drill. Therefore, HR data were excluded from analysis if the recorded duration differed from the corresponding GPS data by one minute or more.

Statistical analysis

All GPS and HR outcomes are reported in relative (per minute) format, while all absolute values are provided in supplementary digital content 3. A linear mixed effects regression model was conducted to examine the influence of training drill type on external and internal load measures, whereby a separate model was conducted for each outcome measure. For all analyses, participant ID was included as a random factor to account for within participant clustering. Pairwise comparisons between each drill type were conducted using Sidak's post hoc correction to control for the familywise error rate. A total of 80 pairwise comparisons were completed, with a further 60 comparisons completed on absolute data. With 10 comparisons per outcome variable, the significance threshold set for all comparisons was p < .0052. All de-identified data and code have been made available via the Open Science Framework (https://osf.io/ytpg8/files/osfstorage). All statistical analyses were performed using R studio (version 3.4.1).

Deviations from pre-planned protocol

The initial protocol had proposed to also observe “contested” drills, but this was combined with the “SSG” drill category due to the high similarity in the drill's structures.

Results

A total 2065 player observations were collected across 21 training sessions, of which 1800 had usable HR data and 2065 had usable GPS data. Table 2 displays the number of observations of each metric per drill type and the mean ± standard deviation. All pairwise comparisons and effect size estimates are reported in supplementary digital content 2. Additionally, all absolute results are reported in supplementary digital content 3.

Table 2.

N of observation per drill type of each metric, mean +/- sd.

	Fundamentals		Ball movement		Conditioning		Small-Sided games		Match simulation
	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD	N	Mean ± SD
Relative Distance (m/min)	761	79.1 ± 28.2 ^△ ^{★ ●} ^▪	274	98.3 ± 23.4 ^# ^★	445	152.0 ± 40.8 ^# ^△ ^{● ▪}	382	94.3 ± 29.2 ^# ^★ ^▪	203	108.1 ± 22.9 ^# ^{★ ●}
Relative Accelerations (n/min)	761	1.0 ± 0.6 ^△ ^★ ^▪	274	0.6 ± 0.4 ^# ^★ ^●	445	1.5 ± 1.2 ^# ^△ ^● ^▪	382	1.0 ± 1.3 ^△ ^★ ^▪	203	0.6 ± 0.3 ^# ^★ ^●
Relative Decelerations (n/min)	761	0.4 ± 0.4 ^●	274	0.4 ± 0.3 ^●	445	0.5 ± 0.9 ^●	382	1.1 ± 1.4 ^# ^△ ^★ ^▪	203	0.6 ± 0.3 ^●
Relative Impacts (n/min)	761	1.9 ± 1.7 ^{★ ● ▪}	274	2.9 ± 2.5 ^★	445	7.3 ± 7.4 ^# ^{△ ● ▪}	382	3.6 ± 2.8 ^{# ★}	203	3.5 ± 2.3 ^{# ★}
Relative HSR (m/min)	761	0.5 ± 1.4 ^★	274	3.4 ± 4.0 ^★	445	29.7 ± 29.4 ^{# △ ● ▪}	382	2.5 ± 3.0 ^★	203	4.2 ± 3.2 ^★
RPE_B (au)	759	4.0 ± 1.7 ^△ ^{★ ● ▪}	273	5.3 ± 1.5 ^{# ★ ● ▪}	442	6.8 ± 1.8 ^{# △ ●}	380	6.0 ± 1.7 ^{# △} ^{★ ▪}	202	6.8 ± 1.1 ^{# △ ●}
RPE_M (au)	759	4.3 ± 1.8 ^△ ^{★ ● ▪}	273	5.5 ± 1.6 ^{# ★ ● ▪}	442	7.0 ± 1.8 ^{# △ ●}	380	6.2 ± 1.8 ^{# △} ^{★ ▪}	202	6.9 ± 1.2 ^{# △ ●}
Relative TRIMP (au)	724	1.5 ± 0.4 ^{★ ● ▪}	265	1.7 ± 0.5 ^{★ ●}	428	2.3 ± 0.7 ^{# △ ●}	198	3.7 ± 3.1 ^{# △} ^{★ ▪}	185	2.0 ± 0.4 ^{# ●}

Abbreviation: N = Number of observations, SD = Standard Deviation, RPE_B = Rating of Perceived Exertion Breathlessness, RPE_M = Rating of Perceived Exertion Leg Muscles, HSR = High Speed Running. TRIMP = Training Impulse. Denotation: # = significantly statistically different to fundamentals, △ = significantly statistically different to ball movement, ★ = significantly statistically different to conditioning, ● = significantly statistically different to SSG, ▪ = significantly statistically different to match sim.

Distance

There was a main effect of drill type on relative distance (424.96, p < 0.001). Pairwise comparisons indicated all drill types were significantly different to one another (all p < 0.005; d = 0.1 to 2.4), other than ball movement and SSG (p = 0.57; d = 0.1). Conditioning had the highest relative distance (152.00 ± 40.82 m/min), while fundamentals had the lowest relative distance (79.12 ± 28.22 m/min).

Accelerations & decelerations

There was a main effect of drill type on relative accelerations (p < 0.001) and decelerations (p < 0.001). Pairwise comparisons indicated all drill types for relative accelerations were significantly different to one another (all p < 0.001; d = 0.01 to 1.1), other than fundamentals and SSG (p = 1.00; d = 0.1) and ball movement and match simulation (p = 1.00; d = 0.01). Conditioning had the highest relative accelerations (1.52 ± 1.24 efforts/min). Match simulation (0.56 ± 0.25 efforts/min) and ball movement (0.57 ± 0.35 efforts/min) had the lowest relative accelerations. Conversely, pairwise comparisons for relative deceleration showed that SSGs were significantly higher than all other drill types (all p < 0.001; d = 0.6 to 0.8) (fundamentals [d = 0.8], ball movement [d = 0.8], conditioning [d = 0.8] and match sim [d = 0.6]), all other drill types were similar to each other.

Impacts

There was a main effect of drill type on relative impacts (p < 0.001). Pairwise comparisons indicated all drill types were significantly different to one another (all p < 0.005; d = 0.01 to 1.4), other than ball movements and SSG (p = 0.13; d = 0.2), ball movement and match simulation (p = 0.34; d = 0.2) and SSG and match simulation (p = 1.00; d = 0.01). Conditioning had the highest relative impact count (7.30 ± 7.37 impacts/min), while fundamentals had the lowest (1.91 ± 1.68 impacts/min).

High speed running

There was a main effect of drill type on relative HSR (p < 0.001). Pairwise comparisons again indicated that conditioning resulted in greater HSR than all other drill types (all p < 0.001; d = 1.9 to 2.2) (fundamentals [d = 2.2], ball movement [d = 2.0], SSG [d = 2.0] and match sim [d = 1.9]), with no differences observed between any other drill types. Conditioning had the highest relative HSR (29.74 ± 29.36 m/min), while fundamentals had the lowest (0.47 ± 1.39 m/min).

RPE-M & RPE-B

Pairwise comparisons indicated all drill types were significantly different to one another (all p < 0.001, [RPE-M; d = 0.07 to 1.7] and [RPE-B; d = 0.01 to 1.8]), other than conditioning and match simulation which were not different (RPE-M, p = 1.00; d = 0.1; RPE-B, p = 1.00; d = 0.01). Conditioning (RPE-M = 6.97 ± 1.76 au: RPE-B = 6.80 ± 1.84 au) and match sim (RPE-M = 6.86 ± 1.18 au: RPE-B = 6.78 ± 1.14 au) had the highest RPE values, while fundamentals had the lowest RPE-M (4.28 ± 1.76 au) and RPE-B (3.97 ± 1.70 au) values.

TRIMP

There was a main effect of drill type on relative TRIMP (p < 0.001). Pairwise comparisons again indicated all drill types were significantly different to one another (all p < 0.001; d = 0.2 to 2.0)), other than fundamentals and ball movement (p = 0.09; d = 0.2), ball movement and match simulation (p = 0.01; d = 0.3) and conditioning and match simulation (p = 0.10; d = 0.2). SSG had the highest relative TRIMP (3.68 ± 3.12 au/min).

Discussion

To our knowledge, this research is the first to investigate the external and internal load outputs across different drill classifications in male AF players. Based on previous literature we hypothesised that drills that are more highly contested will elicit greater internal loading when compared to drills that are not as contested. In comparison, we anticipated that conditioning-based drills would elicit the most high-speed running. Our findings partially supported our hypotheses. SSG drills are often designed to elicit higher contested work performed by players, which may explain why they elicited the greatest internal relative TRIMP responses in this study. However, conditioning drills, which involved little to no contested elements, produced the highest RPE responses, along with match-simulation drills which involved lower player density and subsequently contested play. The results of this study provide insight into how different drill typologies impact measures of internal and external load. This is important to ensure player loads are appropriately manipulated to avoid under- or over-training, and to guide training decisions to ensure desirable training adaptations are achieved.

Fundamentals & ball movement

Fundamentals resulted in the lowest relative external and internal demands (across six of the eight captured metrics) in comparison to all other drill types, with ball movement also recording the lowest demands (across two of the eight metrics) compared to all other drill types. The difference between these drill types is likely a result of their purpose and implementation. Fundamentals are typically performed with minimal movement and in smaller space, while ball movement drills, despite being low intensity, often encourage movement over a large area. It is therefore likely that fundamental drills are best utilised for pure skill development and acquisition. Whereas ball movement drills allow athletes to repeatedly execute skills and accumulate external load, while maintaining overall low exercise intensity. Therefore, if athletes require a lower-intensity training session, more training time could be allocated to fundamental and ball movement drills.

Small sided games

As expected, SSG drills elicited the highest relative TRIMP responses. However, despite this, SSG only led to a moderate ranking of RPE-M and RPE-B values when compared to other drills. The high internal loading but moderate perceptual exertion could be partially due to the external load demands of SSG, which produced the highest number of relative decelerations. This drill type was also associated with low HSR/min which may be responsible for the reduced RPE values. It may be that the frequent accelerations and decelerations, were enough to elicit an elevated HR response without causing a subsequent increase in RPE, which may be more tightly linked to HSR. Previous research by Marynowicz et al., (2020) in youth soccer demonstrated that sRPE increased with increasing HSR demands,²⁵ which supports this finding. Similarly, prior research in youth basketball athletes has indicated that the addition of COD into conditioning drills can cause a significant increase in HR without a change in RPE,²⁶ offering further support for this suggestion. SSG may require the need for carefully planned exposures as, given the high deceleration count, this could introduce higher eccentric training load and thus greater muscle damage.²⁷ Therefore, there should be a gradual progression of SSG durations during the pre-season to minimise muscle damage and allow players to gradually adapt to the demands. There may also be merit in limiting SSG exposure during the in-season periods. For example, Sparkes et al. (2018) found significantly elevated creatine kinase levels 24 h after a SSG session amongst professional soccer players, suggesting muscle damage and impaired recovery.²⁸ In-season, if coaches choose not to limit their inclusion, they should be scheduled earlier in the week as to avoid fatigue carrying over into game day. Additionally, the high relative TRIMP load indicate that SSG may be suitable for the development of aerobic adaptations. As such, SSGs might be a suitable alternative to conditioning drills during periods where high decelerations are not a concern. Additionally, they may be able to be used interchangeably with conditioning drills during the pre-season where building an aerobic base is priority, while also providing additional opportunities for skill development.

Match simulation

As mentioned above, match simulation and conditioning elicited the highest RPE values compared to all other drill types. Considering the high HSR observed during conditioning, this is a logical finding. However, despite high RPE, match simulation drills did not elicit high external load responses. This could be an indication that metrics that are not readily available for monitoring in athletes, such as impacts of tackles, cognitive demands of game play, other psychological stressors, and positional requirements, may influence perceptual measures. Therefore, match simulation might not be the ideal drill type to develop the physiological qualities of AF athletes, despite the high perceptual load associated with these drills. The relative external load values captured during match sim (m/min & HSR/min) in this study do however align with previous studies describing the match day demands of senior elite AFL games.^1,29 With this understanding, these drills may need to be carefully monitored, and in some cases limited, due to their high tackle counts and contested demands, which may increase the risk of contact injuries occurring. For example, in the preseason, match simulation plays an important role allowing players to reach competition-like demands in training, thereby targeting the physiological, tactical and cognitive systems required for sport-specific performance. While in-season they may need to be utilised less as players experience these competition demands through match days, and to reduce the risk of contact injuries during training. Our findings also demonstrate that while correlated, perceptual measures (i.e., RPE) are not a substitute for objective load metrics and therefore, a combination of objective and subjective measures should be routinely used when assessing training and game demands.

Conditioning

Conditioning led to some of the highest measures of relative distance, HSR distance and accelerations. This likely reflects the design of the drills, which are implemented to elicit high physiological responses and improve fitness.³⁰ The high HSR distances are noteworthy since chronic HSR exposure is a critical aspect of athlete preparation, as it prepares athletes for match demands. However, acute spikes in HSR have been proposed as a risk for injury.³¹ As such, exposure to HSR should gradually build in pre-season, like relative decelerations in SSG. This may also suggest that conditioning may need to be carefully applied during the in-season period to ensure that chronic HSR exposure is maintained with consideration for individual athlete's match loads to avoid acute spikes in HSR.

Additionally, relative accelerations were also highest in conditioning. While conditioning drills were not separated by type in this study, it is likely that the acceleration counts would increase during short interval sessions compared to long interval sessions due to their greater number of absolute efforts. Similarly, as they are generally performed at higher running velocities than long intervals,³² they may also result in higher HSR distances. Therefore, long intervals may be a viable option when seeking to reduce accelerations and HSR distance and still elicit aerobic/anaerobic adaptations during the season. However, conditioning, along with match simulation, also resulted in the highest RPE-M and RPE-B values compared to other drill types. This finding likely reinforces the link between RPE and HSR, and suggests that conditioning drills should be most heavily applied during the preseason to induce the physiological adaptation required for in-season performance.

Practical applications

In line with these findings, it may be that conditioning drills and SSGs can be used to elicit an aerobic and anaerobic training stimulus based on training goals and focus. SSGs may be used to limit HSR exposure while increasing relative decelerations, whereas conditioning drills may be chosen to generate high HSR distances with fewer decelerations. However, SSGs may need to be implemented sparingly throughout the in-season period, due to the high number of relative decelerations and high relative TRIMP. This combination of high physiological internal load and external load makes them ideal for sport-specific preparation but may result in greater muscle damage and longer recovery times.³³

Additionally, ball-movement drills may be preferred over match-simulation drills when the goal is to develop skills while minimising athletes perceived exertion. While both these drill types elicited similar external and TRIMP responses, Match simulation elicited a much greater RPE response than ball movement drills. This may be due to the addition of contact, and by extension the greater number of recorded impacts, in match simulation drills. It may also be that the predetermined movement patterns, during ball movement drills, compared to the free-flowing play of match simulation, eliminates the need for decision making, which may contribute to the perceived difficulty of a drill.³⁴ Therefore, although ball movement and match simulation drills elicited similar TRIMP and external load metrics, it is plausible that match simulation drills are better suited for preparing athletes for match demands due to the aforementioned factors that are not captured by load monitoring tools. Conseqently, prescribing ball movement drills alone to achieve the desired external training volumes would likely leave athletes underprepared for competition.

Additionally, the large difference between external load and TRIMP in SSG compared to other drill types may suggest that there is an aspect to SSG's that is placing a greater physiological load on athletes that common GPS metrics are not currently capable of detecting. This highlights the importance of internal load monitoring in conjunction with external load monitoring for practitioners, and indicates that further development of GPS and inertial sensors is required for more holistic workload monitoring of athletes. Across all drill types observed in this study, RPE-M and RPE-B aligned quite closely with each other. This might suggest that a measure of global RPE is sufficient to capture perceived internal load during a training situation, rather than capturing muscle or breathlessness individually.

Limitations

These findings should be interpreted with certain limitations. Firstly, it only included one semi-elite AF team and may not generalise to other groups as the results likely reflect the playing style and training philosophy of the individual players and coaching staff. It would be beneficial to explore this topic in other AF clubs and levels to evaluate whether the observed responses persist more broadly. Secondly, data collection was limited to the preseason period to minimise variability in the data due to the uncontrolled workload that occurs during matches. However, it is possible that responses to training drills differ within the in-season period, which was not explored in this study and should be considered in future research. Furthermore, the wide SDs observed across several measures captured during certain drills likely reflects the natural variation in drill types. While this does provide insight into the loading demands of different drill types more broadly, the results are unlikely to perfectly reflect all drills within a certain drill type category. Lastly, due to software restrictions, the dwell time for acceleration and deceleration data was fixed at 1.0 s. As such, accelerations and decelerations that lasted less than 1 s were not captured.

Conclusions

In semi-elite male Australian Football athletes, fundamental and ball movement drills elicited the lowest relative internal and external loads, whereas conditioning drills produced the highest HSR, accelerations, and RPE. Small-sided games generated the highest relative TRIMP and deceleration counts, highlighting their potential for aerobic development, but also the need for careful exposure due to their potentially high eccentric loads. Match-simulation drills, while eliciting high perceived exertion (RPE), produced moderate external loads and provide opportunities to practice sport-specific skills under competitive conditions. These findings suggest that coaches can strategically manipulate drill type and duration to target specific physiological and perceptual outcomes, balancing skill development, conditioning, and match preparation within a training session.

Supplemental Material

sj-docx-1-spo-10.1177_17479541261445255 - Supplemental material for Investigating the internal and external load output across different drill classifications in semi-elite male Australian football players

Supplemental material, sj-docx-1-spo-10.1177_17479541261445255 for Investigating the internal and external load output across different drill classifications in semi-elite male Australian football players by Keely Cannizzaro, Jordan Fox, Grace Greenham, Samuel Chalmers and Hunter Bennett in International Journal of Sports Science & Coaching

Supplemental Material

sj-docx-2-spo-10.1177_17479541261445255 - Supplemental material for Investigating the internal and external load output across different drill classifications in semi-elite male Australian football players

Supplemental material, sj-docx-2-spo-10.1177_17479541261445255 for Investigating the internal and external load output across different drill classifications in semi-elite male Australian football players by Keely Cannizzaro, Jordan Fox, Grace Greenham, Samuel Chalmers and Hunter Bennett in International Journal of Sports Science & Coaching

Supplemental Material

sj-docx-3-spo-10.1177_17479541261445255 - Supplemental material for Investigating the internal and external load output across different drill classifications in semi-elite male Australian football players

Supplemental material, sj-docx-3-spo-10.1177_17479541261445255 for Investigating the internal and external load output across different drill classifications in semi-elite male Australian football players by Keely Cannizzaro, Jordan Fox, Grace Greenham, Samuel Chalmers and Hunter Bennett in International Journal of Sports Science & Coaching

Footnotes

Acknowledgements

The primary author was supported by a Research Training Program Domestic Fee Offset administered by the University of South Australia (now Adelaide University) on behalf of the Australian government. We would also like to thank the Central District Football Club for facilitating this research.

ORCID iDs

Keely Cannizzaro

Jordan Fox

Grace Greenham

Samuel Chalmers

Hunter Bennett

Ethical considerations

This study was approved by the institutes Human Research Ethics Committees (protocol number 206765). All participants gave written informed consent before participating.

Consent to participate

All participants gave written informed consent before participating.

Consent for publication

All participants gave written informed consent before participating.

Author contributions

Keely Cannizzaro: Conceptualisation, Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Visualization, Writing – original draft, Writing – review & editing.

Jordan Fox: Project administration, Supervision, Visualization, Writing – review & editing.

Grace Greenham: Conceptualisation, Data curation, Formal analysis, Methodology, Project administration, Resources, Supervision, Visualization, Writing – review & editing.

Samuel Chalmers: Conceptualisation, Project administration, Supervision, Visualization, Writing – review & editing.

Hunter Bennett: Conceptualisation, Data curation, Formal analysis, Methodology, Project administration, Resources, Software, Supervision, Visualization, Writing – review & editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

All data and code can be found on the Open Science Framework project linked to this study ().

Code availability statement

All data and code can be found on the Open Science Framework project linked to this study ().

Data deposition

All data and code can be found on the Open Science Framework project linked to this study ().

Supplemental material

Supplemental material for this article is available online.

References

Wisbey

, et al. Quantifying movement demands of AFL football using GPS tracking. J Sci Med Sport 2010; 13: 531–536.

Sullivan

, et al. The efficacy of talent selection criteria in the Australian football league. J Sport Sci 2020; 38: 773–779.

Cummins

, et al. Positional differences in external on-field load during specific drill classifications over a professional rugby league preseason. Int J Sport Physiol 2017; 12: 764–776.

Inoue

, et al. Internal training load perceived by athletes and planned by coaches: a systematic review and meta-analysis. Sports Med 2022; 8: 35.

Hauer

, et al. Lacrosse athletes load and recovery monitoring: comparison between objective and subjective methods. Int J Env Res Pub He 2020; 17: 3329.

Cummins

, et al. Global positioning systems (GPS) and microtechnology sensors in team sports: a systematic review. Sports Med 2013; 43: 1025–1042.

Inoue

, et al. Internal training load perceived by athletes and planned by coaches: a systematic review and meta-analysis. Sports Med 2022; 8: 35.

Corbett

, et al. Development of physical and skill training drill prescription systems for elite Australian rules football. Sci Med Football 2018; 2: 51–57.

Campos-Vazquez

, et al. Relationships between rating-of-perceived-exertion-and heart-rate-derived internal training load in professional soccer players: a comparison of on-field integrated training sessions. Int J Sport Physiol 2015; 10: 587–592.

10.

Buchheit

Laursen

. High-Intensity interval training, solutions to the programming puzzle. Sports Med 2013; 43: 927–954.

11.

Johnston

, et al. Validity and interunit reliability of 10 hz and 15 hz GPS units for assessing athlete movement demands. J Strength Cond Res 2014; 28: 1649–1655.

12.

Aughey

. Australian Football player work rate: evidence of fatigue and pacing? Int J Sport Physiol 2010; 5: 394–405.

13.

Varley

Fairweather

Aughey1

. Validity and reliability of GPS for measuring instantaneous velocity during acceleration, deceleration, and constant motion. J Sport Sci 2012; 30: 121–127.

14.

Castellano

, et al. Reliability and accuracy of 10 hz GPS devices for short-distance exercise. J Sport Sci Med 2011; 10: 233.

15.

Bauer

, et al. GPS Variables most related to match performance in an elite Australian football team. Int J Per Anal Spor 2015; 15: 187–202.

16.

Gastin

, et al. Tackle and impact detection in elite Australian football using wearable microsensor technology. J Sport Sci 2014; 32: 947–953.

17.

Gastin

, et al. Quantification of tackling demands in professional Australian football using integrated wearable athlete tracking technology. J Sci Med Sport 2013; 16: 589–593.

18.

Borg

, et al. An index for breathlessness and leg fatigue. Scand J Med Sci Spor 2010; 20: 644–650.

19.

McLaren

, et al. A detailed quantification of differential ratings of perceived exertion during team-sport training. J Sci Med Sport 2017; 20: 290–295.

20.

Foster

, et al. 25 Years of session rating of perceived exertion: historical perspective and development. Int J Sport Physiol 2021; 16: 612–621.

21.

Scott

, et al. Validity and reliability of the session-RPE method for quantifying training in Australian football: a comparison of the CR10 and CR100 scales. J Strength Cond Res 2013; 27: 270–276.

22.

Plews

, et al. Comparison of heart-rate-variability recording with smartphone photoplethysmography, polar H7 chest strap, and electrocardiography. Int J Sport Physiol 2017; 12: 1324–1328.

23.

Stagno

Thatcher

Van Someren

. A modified TRIMP to quantify the in-season training load of team sport players. J Sport Sci 2007; 25: 629–634.

24.

Buchheit

. Monitoring training status with HR measures: do all roads lead to Rome? Front Physiol 2014; 5: 73.

25.

Marynowicz

, et al. Relationship between the session-RPE and external measures of training load in youth soccer training. J Strength Cond Res 2020; 34: 2800–2804.

26.

Grimal

Lorenzo

. The effects of including changes of direction (COD) in the high intensity interval training (HIIT) on heart rate and the rate of perceived exertion in young basketball players. Rev Psicol Deporte 2018; 27: 165–172.

27.

Gastin

, et al. Deceleration, acceleration, and impacts are strong contributors to muscle damage in professional Australian football. J Strength Cond Res 2019; 33: 3374–3383.

28.

Sparkes

, et al. Neuromuscular, biochemical, endocrine, and mood responses to small-sided games’ training in professional soccer. J Strength Cond Res 2018; 32: 2569–2576.

29.

Johnston

, et al. Movement demands and metabolic power comparisons between elite and subelite Australian footballers. J Strength Cond Res 2015; 29: 2738–2744.

30.

Lisenchuk

, et al. Fitness training in functional preparedness of highly qualified football players. J Phys Educ Sport 2023; 23: 502–509.

31.

Stares

, et al. Identifying high risk loading conditions for in-season injury in elite Australian football players. J Sci Med Sport 2018; 21: 46–51.

32.

Valstad

, et al. Comparison of long and short high-intensity interval exercise bouts on running performance, physiological and perceptual responses. Sports Med Int Open 2018; 2: E20–E27.

33.

Johnston

, et al. Influence of physical contact on neuromuscular fatigue and markers of muscle damage following small-sided games. J Sci Med Sport 2014; 17: 535–540.

34.

Gorman

. 12 Using constraints to enhance decision-making in team sports. In Motor learning in practice: a constraints-led approach. Routledge 2010: 144.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.05 MB

0.23 MB

0.00 MB