About 10 years ago, I did something I could never have imagined doing. I started writing a weekly football column for The Times. Admittedly, it is rather a quirky one. The Fink Tank, as it is called, is based on a statistical model of the game and casts light on both matters of immediate interest (who, say, is the more likely winner of the FA Cup final) and of longer term curiosity (does the form book fly out of the window for derby games, for instance).
It has been surprisingly popular. And there's been another surprise. The football column has changed my way of looking at the world.
I realise that the mistakes that I saw other football pundits making - seeing isolated pieces of data and thinking them representative - I was probably making myself when it came to giving my opinion on politics and history.
In fact, I realised, a great deal of what passes for history is the result of examining individual events, when they should be seen as part of a series.
Anyway, my interest in such things has led my friends to send me all sorts of fascinating material about the application of statistics to political life. But rarely has anybody alerted me to a study as arresting as the one I have just been sent by my good friend, the economist, Professor Jonathan Haskel.
What he sent me, was a paper called Persecution Perpetuated: The Medieval Origins of Anti-Semitic Violence in Germany. And I think even the idea that its authors - Nico Voigtlander and Hans Joachim Voth - had for the study is original.
What these academics were interested in was the persistence of cultural traits. But they chose to study it in a way that answered another question, or more than one.
Was the Holocaust the result of deep-seated German hatred of Jews? Or was it primarily a modern response to economic collapse? Is antisemitism something that breaks out in random places when people face other problems? Or is it there in certain cultures over a long period?
The study begins with the Black Death. When this hit Europe between 1348 and 1350, the suffering was immense but, at that time, the cause was obscure. So, guess what many people did. Yes, that's right, they blamed the Jews. And all over Germany (but not merely there, others joined in quite happily) there were mass killings of the Jewish population.
Some communities were attacked and others spared. The incidence of violence against Jews wasn't geographically concentrated and there isn't much explanation of why it happened in some areas and not in others.
Next, they looked at the Holocaust. Where were the anti-Jewish pogroms in the 1920s? Which areas voted for the National Socialists in the early years before they were a mass movement? Where were the synagogues attacked and when? And what was the timing and number of deportations in different parts of the country?
The last step was to try and relate these two sets of data - antisemitic activity during the period of the Black Death, and antisemitic activity in Nazi and pre-Nazi Germany.
And, guess what. "Pogroms during the Black Death are a strong and robust predictor of violence against Jews in the 1920s, and of votes for the Nazi Party. In addition, cities that saw medieval antisemitic violence also had higher deportation rates for Jews after 1933, were more likely to see synagogues damaged or destroyed in the Night of Broken Glass in 1933, and their inhabitants wrote more anti-Jewish letters to the editor of the Nazi newspaper Der Sturmer."
This result was not explained by their being more, or indeed fewer, Jews in the areas under examination. The most plausible theory is that cultural traits - in this case virulent antisemtisim - persisted over a very long period and that some local cultures were more susceptible to it than others.
The study strongly supports the idea that hatred for Jews was deep-seated, and not a modern creation. The role played by economic collapse emerges as the same as the role played by the Black Death. It was a social disaster that led people to turn to violence.
The idea that people turned on Jews because of the specifically economic nature of the crisis is certainly knocked (if not knocked out) by these findings.
A fascinating piece of work, I hope you agree.