Still using the classic 0.45 translation factor to predict how well AHL players will perform in the NHL? While that simple approach will steer you right most of the time, we've recently been taking a deeper look at more multivariate models for projecting NHL scoring based on AHL performance.
We've divided the data into three groups of players, each of which are in the AHL for entirely different reasons, and translate differently to the NHL. Last week, we looked at the young crowd, aged 19-22, and this week we'll look at the middle group aged 23-26, for which the historical 0.45 translation factor is most appropriate. We'll wrap up with the oldest group later this week.
A word about our methodologywe've been looking at all the players who played at least 20 games in the AHL, followed by 20 games in the NHL, since the Lockout. Players going the other direction aren't included, because that's a different beast, and one in which we're typically less interested.
The first thing we've learned about the group of 23- to 26-year-old AHLers is that the average translation factor of 0.45 will typically work just fine, as long as it's understood that it will work best for players with average scoring, but not those at the high-scoring or low-scoring extremes.