Introduction
Several studies have reported that geographical location and extent of game day travel impact win-loss records of collegiate and professional American football teams. For example, analyses of the 1996 National Collegiate Athletic Association regular season revealed that football teams playing 2 or more time zones away from home were more statistically likely to lose (16 14 13
Daily endogenous peaks and troughs in alertness and fatigue are well characterized in humans. Subjective and objective measures of alertness are positively aligned with daily fluctuations in core body temperature (CBT) and plasma cortisol and norepinephrine levels. As CBT wanes, fatigue increases (4–6 9,11 1 2 3 8 10 7
In this study, we aimed to identify underlying factors related to mental and physical fatigue that could contribute to injuries and known disparities in winning percentages of west vs. east coast teams. We focused our attention on the number of turnovers as a determinant of mental fatigue and cumulative weeks of injury reserve for each position in 2 time zones (Pacific time [PT] vs. Central time [CT]) across the regular season; we chose these 2 time zones to account for the disproportionate number of teams and travel in each respective time zone. We examined if the direction and extent of time zone travel had weight on the number of turnovers and weeks on injury reserve. We also examined how closely aligned game time was with biological time (i.e., the biological time of a PT team playing a CT team at home at 1300 is 1100). We were most interested if rates of mental fatigue (measured by errors resulting in turnovers) and physical fatigue (measured by injuries and resulting weeks on injury reserve) decreased when games were played closer to teams' biological peaks in alertness, namely midmorning and late evening. Furthermore, we sought to understand if this relationship was independent of the direction and extent of game day travel.
These game day predictions of mental and physical fatigue and injury risk are critically important for strength and conditioning communities. Modifications may need to occur throughout the week of strength and conditioning training for specific teams and specific field positions to help ameliorate endogenous fatigue on game day that is beyond control of coaches and trainers: predetermined regular season schedules. Further determinations of injury type (i.e., lower vs. upper extremity) and how these risks correspond with regular season schedules can also help strength and conditioning coaches know what to program to effectively reduce mental and physical fatigue and injury risk.
Methods
Experimental Approach to the Problem
Data for analyses were publicly available from Pro-Football-Reference (www.pro-football-reference.com
Subjects
Across the 17 weeks of regular season play in 2013, the average number of fumbles and interceptions (i.e., turnovers) was obtained from Pro-Football-Reference (www.pro-football-reference.com
Statistical Analyses & Practical Applications
Statistical significance was derived from parametric and nonparametric tests, where appropriate, using SPSS 20.0 (Chicago, IL, USA).
Results
West Coast Teams Win Twice as Often Despite Traveling Four Times as Often as East Coast Teams
Across the 17 weeks of regular season play, the 32 teams of the NFL traveled across 236 time zones. There were 129 instances of westward travel and 107 instances of eastward travel (p = 0.432; Pearson's Chi-Square, direction and week of season). Teams traveling eastward won twice as many games as teams traveling westward; 34 wins were accounted for traveling westward, whereas 69 wins were accounted for traveling eastward (Figure 1 ; p = 0.001; Pearson's Chi-Square, direction and outcome). West coast teams traveled 4 times as often across time zones compared with east coast teams (p = 0.026; Mann-Whitney U ; post-hoc analyses). Total number of time zones traveled averaged 4.6 ± 0.3 time zones for the 17 teams in the Eastern Time zone (cumulative total, 74 ± 4), 6.2 ± 0.1 time zones for the 9 teams in the CT zone (cumulative total, 62 ± 1), 9.5 ± 1.0 time zones for the 2 teams in the Mountain Time zone (cumulative total, 19 ± 2), and 15.8 ± 0.3 time zones for the 4 teams in the PT zone (cumulative total, 63 ± 1).
Figure 1.: Teams traveling eastward during the 2013 regular season won more games. Left panel: total amount of westward compared with eastward travel across the 17 weeks of regular season play was fairly consistent. Right insert: teams traveling eastward won more games (p = 0.001; Chi-Square) **p ≤ 0.05.
Rates of Turnovers Unaffected by Direction and Extent of Travel
In general, losing teams had one more turnover compared with winning teams (1.1 ± 0.1, winning team; 2.1 ± 0.1, losing team; p < 0.001; Student's t -test). This statistic was independent of whether teams were traveling eastward (1.0 ± 0.1, winning team; 2.1 ± 0.1, losing team; p < 0.001; Student's t -test), westward (1.2 ± 0.1, winning team; 2.0 ± 0.1, losing team; p < 0.001; Student's t -test), or were playing in the same time zone (1.0 ± 0.1, winning team; 2.1 ± 0.1, losing team; p < 0.001; Student's t -test). There was no statistically significant relationship between the number of game day turnovers and the direction of travel (p = 0.846, Pearson's correlation). There was also no statistically significant relationship between number of game day turnovers and the extent of travel (p = 0.911, Pearson's correlation).
Defensive and Offensive Lines of East Coast Teams are Injured Four Times as Often
We then developed our own injury metric termed environmental circadian adjusted injury (ECAI). This metric compared cumulative weeks that starters and special teams were placed on injury reserve with cumulative weeks of regular season travel across time zones. The metric was calculated for each position. We focused on depth charts of teams in PT vs. CT zones. One reason was to have better control for the disproportionate number of teams represented in each time zone (4 teams for PT; 9 teams for CT). A second reason was that the extent of regular season travel was statistically different for teams represented in each time zone (4.6 ± 0.3 for PT; 6.2 ± 0.1 for CT; p = 0.040; Mann-Whitney U ; post-hoc analyses).
Before applying our ECAI metric, we initially discovered that the defensive lines in PT and CT were placed on injury reserve for similar amounts of time across the 2013 regular season; 7.4 weeks on injury reserve for PT vs. 7.7 weeks for CT (Figure 2 ). In contrast, offensive lines in CT were on injury reserve twice as often as offensive lines of the PT; 4.2 weeks for PT vs. 9.2 weeks for CT (Figure 2 ). For both time zones, defensive tackles and left and right guards were injured most often compared with the other field positions.
Figure 2.: Teams on Central Time (CT) zone had more defensive and offensive players placed on injury reserve compared with teams on Pacific Time (PT). Gray values denote the average number of cumulative weeks of injury reserve during the regular season for each defensive position and offensive position. There were 9 teams on CT and 4 teams on PT. FS = free safety; SS = strong safety; WLB = weak-side linebacker; MLB = middle line backer; SLB = strong-side linebacker; CB = center back; DE = defensive end; DT = defensive tackle; FB = full back; LT = left tackle; LG = left guard; C = center; RG = right guard; RT = right tackle; WR = wide receiver; QB = quarterback; TE = tight end; RB = running back.
We found even more striking disparities in rates of injuries using the ECAI metric. First, defensive lines in CT were injured 4 times as often as defensive lines in PT (1.42 ± 0.56 for PT; 4.40 ± 1.06 for CT; p = 0.03; Mann-Whitney U ). Second, offensive lines in CT were injured 5 times as often as offensive lines in CT (0.96 ± 0.57 for PT; 5.72 ± 1.50 for CT; p = 0.03; Mann-Whitney U ).
West Coast Teams Play Games Closer to Endogenous Peaks in Alertness
Next, we examined the temporal distribution of injury reserve for offensive and defensive lines in CT vs. PT. We found a disparity midseason (weeks 4–8). During this time, nearly every team in CT had players on injury reserve, whereas teams in PT had few (p = 0.001; Pearson's Chi-Square, time zone and injury). The extent of travel alone did not account the prevalence of injury because teams in PT were traveling more often than teams in CT (p = 0.001, both; Pearson's Chi-Square, time zone and extent and direction of travel). Also, several teams in PT and CT had their annual “bye” (off) week during this time.
Given this time window of injury prevalence for CT vs. PT, we plotted actual game (exogenous) time against each team's hypothetical biological (endogenous) time. For example, we discovered that teams in PT played games closer to endogenous, biological peaks in alertness (attenuated fatigue): (a) midmorning (28% for PST vs. 0% for CST; p = 0.001, Pearson's Chi-Square, time zone and biological game time); and (b) early evening (39% for PST vs. 24% for CST; p = 0.001, Pearson's Chi-Square, time zone and biological game time). In contrast, teams in CT played games closer to endogenous, biological troughs in alertness (accentuated fatigue; Figure 3 ): afternoon (76% for CT vs. 33% for PT; p = 0.001, Pearson's Chi-Square, time zone and biological game time). Thus, the higher injury prevalence for CT vs. PT could possibly be explained by game times that correspond with the biological trough in alertness.
Figure 3.: Games played in (biological) afternoon correspond with more midseason time spent on injury reserve. Top panel: weekly distribution that starters and special teams on Central (CT; black) and Pacific (PT; gray) times. Lower left: biological time of midseason game derived from actual game time and location [lower right] extent and direction of midseason travel. BYE = rest week; EA = early afternoon (1300–1600); LA = late afternoon (1600–1800); M = morning (1000); N = night (1900–).
Discussion
Our analyses revealed that west coast teams of the NFL were more likely to play at a time that corresponded with typical biological peaks in alertness across the 2013 regular season. This relationship of game time and performance outcome was surprisingly stronger than a relationship with travel time (number of time zones traveled) and performance outcome. First, we learned that teams traveling eastward (i.e., west coast teams) won significantly more often than teams traveling westward (i.e., east coast teams). The extent or duration of game day travel alone could not account for more wins, as west coast teams traveled 4 times as often as east coast teams. The number of turnovers could also not account for more wins for west coast teams.
However, what could have possibly accounted for more wins for west coast teams was the discovery that the prevalence of injury to defensive and offensive lines of teams in CT was 4 times greater compared with teams in PT. In general, defensive tackles and right and left guards in CT and PT spent the most amount of time on injury reserve. However, there was significant disparity in time spent on injury reserve for teams in CT vs. PT midseason. Midseason, we discovered that teams in PT play more games closer to typical, endogenous peaks in alertness (attenuated fatigue) (4–6 4–6
Similar to the previous studies of Smith et al., 1997 (14 13
In agreement with our present findings, there are several studies that have demonstrated optimal or improved performance near endogenous peaks in alertness in the midmorning and late evening. In fact, evidence for biological “alerting” and “fatigue” signals influencing mental and physical performances is strong. These data have significant experimental reliability and ecological validity (4,9,11,12 1 2 3 8 10 7 Ref. 15
Our data support the theory of endogenous, circadian “alerting” signal can have significant dictation on game day errors, likelihood of injury, and subsequent win-loss records. Future studies relevant to circadian misalignment and injury metrics ought to focus on injury type (e.g., neurological vs. orthopedic) as well as injury prevention and risk (e.g., risk for traumatic brain injury, changes in bone density, and effect on neuromuscular response times). This information is particularly imperative for players on the front defensive and offensive lines as our data demonstrate that these players more likely to be injured during regular season play compared with other positions. These game day predictions of mental and physical fatigue and injury risk are also critically important for strength and conditioning communities. Modifications in how and what to program for specific teams and specific field positions may need to occur throughout the week of strength and conditioning training to effectively reduce injury risk. In addition, these fatigue and injury metrics are also operationally relevant to mission preparation and unit readiness for the United States Armed Forces.
Acknowledgments
This was not an industry-supported study. The authors have indicated no financial conflicts of interest.
References
1. Arnett MG. Effects of prolonged and reduced warm-ups on diurnal variation in body temperature and swim performance. J Strength Cond Res 16: 256–261, 2002.
2. Atkinson G, Todd C, Reilly T, Waterhouse J. Diurnal variation in cycling performance: Influence of warm-up. J Sport Sci 23: 321–329, 2005.
3. Bessot N, Moussay S, Clarys JP, Gauthier A, Sesboüü B, Davenne D. The influence of circadian rhythm on muscle activity and efficient force production during cycling at different pedal rates. J Electromyogr Kinesiol 17: 176–183, 2007.
4. Czeisler CA, Duffy JF, Shanahan TL, Brown EN, Mitchell JF, Rimmer DW, Ronda JM, Silva EJ, Allan JS, Emens JS, Dijk DJ, Kronauer RE. Stability, precision, and near-24-hour period of the human circadian pacemaker. Science 284: 2177–2181, 1999.
5. Dijk DJ, Duffy JF, Riel E, Shanahan TL, Cziesler CA. Ageing and the circadian and homeostatic regulation of human sleep during forced desynchrony of rest, melatonin, and temperature rhythms. J Physiol 516: 611–627, 1999.
6. Duffy JF, Dijk DJ, Hall EF, Cziesler CA. Relationship of endogenous circadian melatonin and temperature rhythms to self-reported preference for morning and evening activity in young and older people. J Investig Med 47: 141–150, 1999.
7. Facer-Childs E, Brandstaetter R. Circadian phenotype composition is a major predictor of diurnal physical performance in teams. Front Neurol 6: 208–221, 2015.
8. Kline CE, Durstine JL, Davis JM, Moore TA, Devlin TM, Zielinski MR, Youngstedt SD. Circadian variation in swim performance. J Appl Physiol 102: 641–649, 2007.
9. McCauley P, Kalachev LV, Mollicone DJ, Banks S, Dinges DF, Van Dongen HP. Dynamic circadian modulation in a biomathematical model for the effects of sleep and sleep loss on waking neurobehavioral performance. Sleep 36: 1987–1997, 2013.
10. Reilly T, Atkinson G, Edwards B, Waterhouse J, Farrelly K, Fairhust E. Diurnal variation in temperature, mental and physical performance, and tasks specifically related to football (soccer). Chronobiol Int 24: 507–519, 2007.
11. Shekleton JA, Rajaratnam SM, Gooley JJ, Van Reen E, Czeisler CA, Lockley SW. Improved neurobehavioral performance during the wake maintenance zone. J Clin Sleep Med 9: 353–362, 2013.
12. Sletten TL, Segal AY, Flynn-Evans EE, Lockley SW, Rajaratnam SM. Inter-individual differences in neurobehavioral impairment following sleep restriction are associated with circadian rhythm phase. PLoS One 10: e01282730–e01282741, 2015.
13. Smith RS, Efron B, Mah CD, Malhotra A. The impact of circadian misalignment on athletic performance in professional football players. Sleep 36: 1999–2001, 2013.
14. Smith RS, Guilleminault C, Efron B. Circadian rhythms and enhanced athletic performance in the National Football League. Sleep 20: 362–365, 1997.
15. Teo W, Newton MJ, McGuigan MR. Circadian rhythms in exercise performance: Implications for hormonal and muscular adaptation. J Sports Sci Med 10: 600–606, 2011.
16. Worthen JB, Wade CE. Direction of travel and visiting team athletic performance: Support for a circadian dysrhythmia hypothesis. J Sport Behav 22: 279–283, 1999.
