Tip of the hat to Jeremymiles over at BadScience.net.
I like these quotes from the (short) article:
The omission of information on survivors is a particular problem if there are fluctuations in the proportions of left and right handed people in the population. In fact, more tolerant attitudes have resulted in a gradual increase in the proportion of left handed people during this century.7 There are therefore relatively few left handed elderly people (as many were forced to switch hands), but more left handed people among younger groups. Any comparison using the average age of death in 1994 is likely to come to the spurious conclusion that left handers die younger because it is weighted by the preponderance of elderly right handed people.7 14 We therefore re-examined the potentially important claim of a difference in the mortality of left and right handed people by using analytical techniques that avoid these problems.
Bowling (or throwing) a ball is a good predictor of handedness,16 17 and the precision required makes it unlikely that proficient bowlers would learn to use their non-preferred hand. Indeed, throwing hand is
relatively insensitive to cultural pressure,18 and the strategic value of left handed bowlers is well recognised, making it less likely that left handers might be forced into learning to bowl right
Our study of 6173 adult men provides a rigorous and extensive analysis of archival data on survival. It highlights the problems that can arise if mean age at death is used as the sole measure of longevity. The regression analyses found no overall difference in the survival curves of the left and right handed cricketers despite apparently different mean ages at death.
Our findings leave unanswered the intriguing issue of why left handed players should have been more susceptible to unnatural deaths and, in particular, deaths during warfare. Though the proportion of left handed players increased slightly before the war years (fig 1), the increase is not sufficient to explain the results. The difference remained significant after year of birth was adjusted for. It seems, therefore, that left handed people may face particular disadvantages during warfare, perhaps because equipment and training are designed for right handed people. Our findings for other forms of accidental death are less clear as they are based on small sample sizes. Although it has been suggested that left handed people are more prone to serious accidents,6 this claim has been challenged.11
[emphasis added by me]
I have a wacky conspiracy-type "theory" for why left-handed British cricket players die more frequently during warfare: all the subjects in this study were professional cricket players in UK leagues. Perhaps, given the strategic (actually tactical) value of left-handed bowlers, and the fact that any individual cricket team will have many more than one opposing team in its league, any fan(atic) of an opposing cricket team could decide to eliminate a team's valuable bowler during the confusion of a battle - shoot that damned lefty in the back, and blame it on ze Germans. The fact that these professional players are more likely to be officers just serves to increase the temptation for a frag.