Yes, and it appears that none of the remedial classes, holidays with the princess (inflated or deflated), nor any other interventions have worked.
A question though for you and any of the other clever people here. I note that in the James Willoughby articles that I posted there, in the female allowance one he has used Impact value to adjust expected chance for number or runners in each race analysed. Exactly what you and I did when we examined the NZ handicapping data pre and post the rating changes for NZTR. Except that we were able to go a step further by using current ratings as a proxy for ability as well. Something that he wasn't able to do leading to considerable caution in his conclusions.
My question though is that I see in the earlier study on the 3yo allowance, he has used beaten runners to achieve the same thing as the IV. In your opinions is that as accurate a measure/correction for expected chance due to number of runners per race?