When Andy Murray recently won Wimbledon for the second time in his career, three years after his first win, in many ways it was unexpected. To many sports fans, this was because in that time his coaching and support team had undergone many changes, he had, dealt with surgery on a chronic back injury, and also been in the shadow of Novak Djokovic’s incredible dominance of the sport. However from an evolutionary view, two other events in Andy’s life since his first Wimbledon win in 2013 made his win in 2016 even more surprising; his marriage to Kim Sears in 2015 and the birth of his daughter, Sophia, in February this year. More on Andy’s specific case later, as well of those of the other two players currently in the top three in the world, Roger Federer and Novak Djokovic, but first more on why we should consider this unusual.
To begin with, it’s important to recognize that a great deal of the success of sportsmen (and women) is down to their extremely high levels of competitiveness, a characteristic that works in tandem with their natural ability to drive and compel them to the very top of their profession. So what affects competitiveness, and why is this important? The answer to this is partly lies in our biology, and relates to differing levels of testosterone, a hormone that circulates within all our bodies. There are clear links between levels of testosterone and measures of competitiveness in many mammal and avian species, and it is also the case that testosterone is found in far higher levels in males than females. The evolutionary explanation for this latter finding is that, due to differences in parental investment, females invest more time and effort into offspring care than males do, and therefore need to be more choosy in who they mate with (a finding that Charles Darwin famously called ‘female choice’) and as a result males need to compete amongst themselves (what Darwin called ‘intra-sexual competition’) to be chosen as a mate. Therefore, the drive to compete should be stronger in males, and higher testosterone levels facilitate this.
So far so good, but things get really interesting when we examine how this can vary over a man’s lifetime. In the early 2000’s, the anthropologist Peter B. Gray and colleagues found, across a number of studies, that heterosexual men who were married or in committed, long-term relationships had lower levels of testosterone than their single counterparts. This makes sense from an adaptive perspective when one considers how costly it is to invest large amounts of time and effort into competing with other males. Furthermore, high levels of competition puts an individual at high risk of potential physical injury, not to mention that testosterone is quite a nasty chemical to have flowing at too high volumes as it can suppress the immune system. When a male is single, such costs are outweighed by the benefits that can be gained from out-competing others for access to potential mates. However, when a male is no longer single and is in a committed relationship, a better use of his time and energy is to invest in his current partner. Therefore lower testosterone levels in paired males leads to an adaptive shift in their motivation from seeking new partners to maintaining and developing their relationship with their long term partner or wife (and consequently any future offspring they may share). Again, this is not unique to humans, and the same phenomenon has already been observed in different bird species, where males have higher testosterone levels only during the breeding season
If there is an adaptive change in testosterone levels in men when they enter a long-term, committed relationship, then we should also expect to see similar changes in competitive behaviour. This is something that slowly dawned on me as I continually heard tennis commentators remark that different players had not been the same since they got married. Although anecdotal, the famous examples were quite striking; John McEnroe married Tatum O’Neal in 1986, and subsequently took two long breaks from playing and never reached his previous career highs. Pete Sampras won his 13th grand slam title in 2000, later that year married Bridgette Wilson, and then only one won more grand slam title (the US open in 2002). Sampras has also been quoted as saying the following in 2001, “There were times, five years ago, where tennis was my life. I was consumed with being No. 1. You know, just being on top for so long, I think I kind of had enough. Getting married and having a child on the way gives me balance.”
With these examples in mind, I examined in more detail what, if any, effect marriage had on the competitive performance of male professional tennis players. Along with my co-author Daniel Nettle, we looked at the end of year Association of Tennis Professionals (ATP) ranking points of the top 100 male players (a nice, explicit measure of competitive performance) over a ten year period between 1996-2005, and compared the points achieved by married players in the year before they got married with the year after. However, as players tended to get married later in their careers it was expected that performance would drop anyway regardless of marital status, so we also used as a sample of age-matched single players as a control group. We found that although both married and single players experience a drop in performance in the later year, the drop in performance was significantly sharper among the married players. As well as looking at overall ranking points gained, we also looked at their winning percentage (to control for the possibility that married players were just playing in less tournaments). Again, the winning percentage decreased at a greater rate for married players than it did for their single counterparts
Although we were unable to measure players’ testosterone levels as part of this research (The likelihood of ever getting funding to follow the professional tennis circuit around the world all year to go into locker rooms to get samples of saliva from the top players is pretty slim!), there appears to be a clear parallel between what we found from behavior and what had previously been found from hormones. In fact, other findings about how testosterone varies in males over their reproductive life have famous examples from sport to illustrate them. For example, further research has found that divorce leads to men having an increase in testosterone levels, due to them once again entering the mating market and needing to compete with other males. This may explain why Andre Agassi rose from outside the top 100 to number one in the world after his divorce from Brooke Shields, and also how the six-times divorced Pancho Gonzalez was still able to play at the top of the game at the age of 44. There’s also the finding that paired males who still seek extra-marital affairs have testosterone levels similar to those of single males, as despite their relationship status the former are still competing to attract further partners. And with that, the example of Tiger Woods springs very prominently to mind…
Two things we couldn’t examine in our paper are also noteworthy. Firstly, we only looked at male tennis players. We did aim to look at what effect marriage may have on female players, but unfortunately it seemed that female players rarely married during their careers, making it impossible to get sufficient data to analyze this. However, I would predict a similar finding (although not as strong), as in monogamous species such as humans there is still some degree of competition for partners among females, and there is some evidence that the links between relationship status and testosterone in men are similar in women. Secondly we purposefully eliminated any players who were fathers from the analysis, as we wanted to concentrate solely on the psychological effects marriage may have, rather than compounding that with other tangible effects that fatherhood brings (e.g. sleepless nights). However fatherhood too is linked with lowering testosterone levels so if we would have been able to include fathers as a separate group I predict that these players’ performance would have dropped even further then their married-but-not-a-parent colleagues.
This brings me back to the strange case we are currently experiencing in the tennis world whereby the current top 3 players in the world (Novak Djokovic, Andy Murray, and Roger Federer) are all married and also fathers (and in Federer’s situation, the father to two sets of twins). Does this suggest that our original finding from 2007 is no longer relevant? Well, it’s certainly the case that the link between marriage and competitive performance is more complex when we look at other sports. In basketball, there is evidence that married players perform better than unmarried players, and famously Sir Alex Ferguson said he liked his Manchester United players to ‘settle down’ (i.e. get married) in order for them to be more focused and therefore successful. However these are both team sports, where the competitive edge an individual player possesses is often subsumed or even stunted for the good of the team. If one looks at other individual sports then we see similar sex differences in competition, such as in distance running. Therefore, it does seem valid that we can view modern sporting contests as the perfect cultural equivalents of the competitive arenas that our ancestors once graced.
But this still does not address what is special about Andy, Novak and Roger that makes them contradict both theory and previous evidence. To do this, one speculative theory I suggest is that any effect of a decrease in motivation to compete following marriage and/or fatherhood will vary from player to player. As a result, any effect of this on actual performance will be more pronounced in players for whom a competitive will and desire to win is more central to their performance. In other words, this could be characterized as more fiery and emotionally-charged players such as John McEnroe or Pancho Gonzalez. This is because any effect of decreased competitiveness caused by changes in testosterone levels will be a very subtle one and will manifest indirectly, (i.e. not, for example, from the fact that wearing a wedding ring weighs you down a bit or affects the toss-up of the ball). We’re talking instead here about small things like slightly poorer preparation for tournaments, coping slightly less well with pain and fatigue, or stopping training and practice sessions slightly earlier than before. Without this, top players will lose that vital 1%-2% they need to win, and this is more damaging for those for whom 100% is more necessary for them to win. So, the very good and naturally gifted players such Roger Federer can cope with a tiny reduction in motivation and get by pretty well on their supreme natural ability. The same too may go for Novak Djokovic and Andy Murray who have spent many years developing and strengthening their technical skills and how they cope psychologically on the court, so that any drops in competitiveness can be accommodated well.
However that’s just my theory, and not only is it difficult to test but also may already be unnecessary – after his recent surprise early defeat at Wimbledon this year, Novak Djokovic immediately spoke about how important his new family were to him and of his life away from the sport, which perhaps suggesting a change in motivation or focus? Time of course will tell, and as the world’s top players prepare for next month’s Olympic Games it will be interesting to see if any further light is shed on the complex interplay between professional sportsmen’s professional and private lives.
 Darwin, Charles (1871). The Descent of Man, and Selection in Relation to Sex. London: John Murray.
 Gray, P. B., Chapman, J. F., Burnham, T. C., McIntyre, M. H., Lipson, S. F., and Ellison, P. T. (2004). Human male pair bonding and testosterone. Human Nature, 15, 119–131.
Gray, P. B., Kahlenberg, S. M., Barrett, E. S., Lipson, S. F., and Ellison, P. T. (2002). Marriage and fatherhood are associated with lower testosterone in males. Evolution and Human Behavior, 23, 193–201.
 Wingfield, J. C., Hegner, R. E., Dufty, A. M., Jr., and Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. American Naturalist, 136, 829–846.
 Farrelly, D., & Nettle, D. (2007). Marriage affects competitive performance in male tennis players. Journal of Evolutionary Psychology, 5, 141-148.
 Mazur, A., & Michalek, J. (1998). Marriage, divorce, and male testosterone. Social Forces, 77, 315-330.
 McIntyre, M., Gangestad, S. W., Gray, P. B., Chapman, J. F., Burnham, T. C., O’Rourke, M. T., & Thornhill, R. (2006). Romantic involvement often reduces men’s testosterone levels–but not always: the moderating role of extrapair sexual interest. Journal of personality and social psychology, 91, 642.
 van Anders, S. M., Hamilton, L. D., & Watson, N. V. (2007). Multiple partners are associated with higher testosterone in North American men and women. Hormones and Behavior, 51, 454.
 Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2011). Longitudinal evidence that fatherhood decreases testosterone in human males. Proceedings of the National Academy of Sciences, 108, 16194-16199.
 Ze’Ev, S. (2015). The marriage premium and productivity: the case of NBA players. International Journal of Economic Sciences, 4, 53-65.
 Deaner, R.O., (2013). Distance running as an ideal domain for showing a sex difference in competitiveness. Archives of Sexual Behaviour, 42, 413-428.