37. Simply the best (30-4-2020)
Simply the best (30-4-2020)
Written by: Jan Luiten van Zanden
Twice a day, our neighbour Leo releases his racing pigeons, and they fly in circles around our house, rest a bit on the rooftop, harassed sometimes by one of the magpies (who know that they are tame softies, not used to the dangers of ‘streetlife’, so easy prey), and then called in by Leo with the sound of their food in a tin can. He has kept pigeons for many years, and even set up a website for the trade-in them, from which he profits handsomely. It is big business these days, since pigeon racing has become popular in China, also and perhaps in particular as an object of gambling. The best Dutch and Belgian pigeons are exported for incredible sums of money.
In pigeon racing the idea is that your best birds are released at great distance – 500 to more than 1000 km, usually in France – and then have to fly back home. The first to arrive home is the champion, and often directly sold abroad. The pigeon keepers (in Dutch we have the nice word ‘duivenmelkers’ for them) do everything to bring them in good shape (for example, they select the best food and do everything to guarantee good health), but on top of that they are incentivized by manipulating their sexual desires. The doffer (male pigeon, I am afraid they are the stars of the competition), once selected for a release in France, is not allowed to join the daily flight, but kept in his loft for weeks. At the right moment the pigeon keeper will locate an attractive female pigeon next to him, to make him even more eager. In the language of pigeon keepers, on top of ‘condition’ you need ‘form’ to perform at the limit of your capabilities (a perhaps bad joke would be ‘don’t try this at home’, but I will not make it). I am not sure what happens when the doffer returns home, but all mating activity is highly planned and regulated, so I am not optimistic about it. In a way, the flight from Chateauroux to the Netherlands is just a test to select the ‘fittest’ birds for the next round of the selective breeding process, aimed at pigeons with even better condition and form.
Another friend, who is both an academic and has kept pigeons many years ago, has been involved with a big debate among pigeon keepers about how to measure performance. The problem is that the pigeons that are released in France are owned by duivenmelkers from all parts of the Netherlands. They are registered when they arrive home (electronically, these days, they carry a kind of tracing device), but for obvious reasons pigeons from Breda have an advantage over those from Groningen. So it is not the pigeon who touches base first who is simply the best, but they calculate the average speed of all pigeons who register, and the first prize goes to the fastest, with the highest average speed. So far so good, but the problem is that pigeons become tired, and therefore can be expected to fly slower on the last leg – from Breda to Groningen. The existing ‘model’ assumes that the average speed of the Chateauroux – Breda leg can be extrapolated to this second leg, but that is not correct. How big a correction is necessary, is also dependent on weather conditions between Breda and Groningen: when the weather is bad, or the wind blows from the north, the ‘northern’ pigeons are more disadvantaged than with sunny weather and a south-western wind. Large datasets are now available to study this carefully, and advanced statistics make it possible to perform the necessary corrections. However, most pigeon keepers are not academics, and find it difficult to trust the complex manipulations of data that are required.
Moreover, this critique implies that duivenmelkers from the south have been systematically favored by the old system, and that northern pigeons and keepers for decades did not have the same chances to win. The geography of pigeon keeping in the Low Countries resembles that of professional cycling, with a heavy concentration in Flanders and Brabant. The southern pigeon keepers dominate the organization and have by far the most valuable pigeons – so there is a lot at stake, especially since Chinese customers have driven up prices. The message that this biased distribution is due to a wrongly specified ‘model’ is not a guarantee for popularity – it even implies that the breeding strategies of the past were at least to some extent based on imperfect information. So ‘big data’ are even shaking up the world of racing pigeons.
Continue reading: The wonders of the Orkneys (1-5-2020)