Dear Readers,

Asteroids and comets are in the news: On February 15, a meteor hit the Chelyabinsk region in Russia that, according to NBC news, “triggered an atomic bomb–sized shock wave that injured more than a thousand people, blew out windows, and caused some Russians to fear the end of the world.” Only a few days later, a new scare was prompted by a near miss of an asteroid (Asteroid 2012 DA14) that, according to NASA, was at a “distance ranging from 74,000 miles to 195,000 miles away from planet Earth.” The latest headline, at the time of writing this letter in late March, is about a newly discovered comet called C/2013 A1 (Siding Spring) that presumably will get exceedingly close to Mars, even with a remote likelihood it may hit the red planet.

Nick Horton, a fellow editor at CHANCE, referred me to a passage in a February 23, 2013, article in The Economist about the 1908 meteor that exploded over Siberia: “[B]ut the land in question was Siberia, so few people noticed and those who did had little influence. Suppose, though, that it had devastated a city in Europe or North America. The history of the 20th century would have been different, as the best scientific and engineering brains were brought to bear on the question of how to stop it happening again.”

We then found it quite befitting to devote the lead article of this issue to a highly informative piece by Germano D’Abramo, a leading researcher in space astrophysics and planetary sciences on what goes into the calculation of the impact probability of a newly discovered asteroid. As D’Abramo argues, the recent hype is partly due to the significant advancement of planetary observational tools, as well as the employment of “sophisticated mathematical and numerical tools” for the propagation of measurement error. In other words, asteroids have been around for a very long time, but they receive extra attention these days, possibly because they can be tracked far more efficiently. On the technical side, D’Abramo points out that to gauge the likelihood of an asteroid hit, one would be better off calculating what he labels as “a posteriori conditional probability of approaching unity,” a somewhat intricate representation of the almost certain probability of an asteroid colliding with Earth.

Also in this issue of CHANCE, relying on a successful career both in academia and the private sector, Stephen Blyth gives an insightful set of recommendations for applying statistical modeling of complex phenomena, particularly that of financial data. The main message of this timely and important article is that mechanical employment of statistical techniques, which is often manifested through ignoring modeling assumptions and restrictions, could result in catastrophic consequences, the case in point is the great financial collapse of 2008. According to Blyth, “applied statistical reasoning” is only achieved when sharp statistical knowledge is coupled with a deep understanding of the subject matter to which a statistician is contributing.

Jay Kadane was on the frontlines, defending the principles of Bayesian statistics, long before their increasing popularity due to recent strides in computational speed and affordability. More importantly, in my view, Jay is an embodiment of a pragmatic-academic, an individual with the conviction that combining theory and application with conscientious action can realistically contribute to social justice. His long-term involvement with the statistical dimensions of legal matters would only reinforce the above claim. Michelle Dunn and I interviewed Jay for CHANCE, discussing his latest book, Principles of Uncertainty, his views on statistics, and his advocacy for human rights.

Following their widely debated earlier article, Donald A. Redelmeier and Christopher J. Yarnell have written yet another provocative piece for CHANCE, suggesting that the unusually sharp elevation of the number of road accidents around April 15 may have something to do with, well, the anxieties and burdens of the tax deadline.

In his latest column, Howard Wainer, in collaboration with Robert Cook, looks into the extremely sensitive issue of gun violence in America, in conjunction with state-by-state life expectancy, gun registration, and economic and educational status. In the same vein as Joseph Fletcher’s maps of moral statistics (the topic of a previous column), Wainer and Cook’s graphical juxtaposition of the above parameters on the U.S. map leads to a striking, if not necessarily shocking, display that the number of reported gun deaths is in concordance with availability of guns, lower life expectancy, and lower economic standing, itself an indicator of poor quality of education.

In this issue of Ethics and Statistics, Andrew Gelman argues that there can be ethical breaches when top-ranking journals incentivize publication of papers that are thematically formed to “assert-and-defend” hypotheses, as opposed to “explore-and-study” them. The column is developed around a much-discussed article in a highly respected economics journal, asserting that genetic diversity of populations would be a strong indicator of their economic development.

Finally, Nicole Lazar advocates a more direct presence of statisticians in the Big Data movement; the editors of Taking a Chance in the Classroom use the end-of-the-semester student evaluation data to explore if there is any association between instructors’ good looks with more agreeable student feedback; and book editor Christian Robert analyzes six books, including Magical Mathematics, the subject of our winter interview with the great Persi Diaconis.