I’ve been looking at winning distances, mainly because there are spreadbetting markets in them. As far as I can tell the spread companies know what they are doing so there isn’t an obvious profit opportunity. However, one thing puzzles me: the average distance between second and third place is consistently higher (both in the prediction markets and in reality) than the difference between first and second. One might expect that in say a 6 runner handicap the runners would be bunched in the middle so the distance from third to fourth would be the smallest and the distance from first to second or fifth to sixth would tend to be largest. So second to third ought to be closer than first to second.
Can anyone explain this phenomenon?