I'm trying to apply a load on a hole which a shaft connects through with a close enough fit to assume they're same diameter. In the image below you can see the results.

The bearing load is applied, putting the bar into tension. I applied it across the whole of the bored surface, is that correct? Or should it have been applied across only the half which the force would likely act upon?

Also, I find it difficult to believe that the weakest area of the part is where it connects to the shaft. Where the highest stresses are highlighted.

I would expect it to experience the most stress where the neck of the bar decreases, seen as the vertical stress line. (I know it is highlighted as a concentrated stress area, but its still not as large as the stresses in the shaft hole.)

If anyone could tell me where I may be going wrong or has any tips on how to model this situation it would be great. I've already had much trouble with stress singularities around the outer circumference of the hole.