Here we lay out our one day ahead predictions for the next day’s racing. The basis is our overall estimates of crew speeds, but they’re adjusted based on previous days’ performances. For example, if a crew has gone up more than we’d expect, we’ll adjust our prediction to reflect that. We don’t incorporate additional information beyond the results, for example crashes.
Please note that the methodology behind these predictions is lower quality than our overall estimates, so take them with an even larger pinch of salt. The predictions on the boundaries of decisions will not always align, as each division is modelled independently here.
The columns of each table give the probability of each crew moving up by more than one place (>+1), moving up by one place (+1), staying in the same place (0), moving down by one place (-1), and moving down by more than one place (<-1). Crews can move more than one place in a day either by being involved in an overbump or being the sandwich boat.
On Saturday morning, we adjusted the methodology behind these predictions due to technical issues we encountered later in the week. This means that the historical predictions on this page changed. They reflect what we would have predicted at the time using the updated methodology. The original predictions are available here.
Please note that the M3 division predictions for this day are especially uncertain, and should be viewed with extreme caution.
Please note that the M3 division predictions for this day are especially uncertain, and should be viewed with extreme caution.