Sorry for the late response. Work gets in the way sometimes.
Let me try and explain this spreadsheet. The original intent of this spreadsheet was for my use only, and wasn't intended for others to understand easily.
Since roller followers are used on may other things (machinery, etc.), a designer needs to understand the loads at the roller follower/bore interface. The designer also needs to understand at what pressure angle the follower will "lock" in the bore, so as to avoid having a catastrophic failure. That is the intent of the spreadsheet. What that being said, MOST roller cams in race engines are designed with a pressure angle at or below 30 degrees. I doubt that Mike Jones veered from that standard when designing your camshaft (providing you told him you had a 0.904" lifter and what roller wheel diameter you were using). So, using the actual data that you provided and a 30 degree pressure angle, the spreadsheet shows the loads at the top and bottom of the lifter bore. The values in red (pressure angle that will cause the lifter to "lock" in the bore) are not the values that you are likely to have. IF the lifter bore has wear and the lifter is allowed to "cock" in the bore, the pressure angle increases to a value higher than the design angle and you begin to have highly localized contact of the lifter to the lifter bore. When that happens, the coefficient of friction (I used typical values in my spreadsheet for you application) increases significantly.
With an 1149 lb. side load on the lifter the next step would be to calculate the allowable load at the maximum lifter velocity using a steel to bronze interface. I will perform that calculation for you a little later and post it here. I will use typical cam lobe maximum velocity (maybe 0.007 in/deg2) and 7500 engine RPM in my calculations. I seriously doubt that you combination is near the limit, since there are many engines with higher valve spring loads and more aggressive camshafts that aren't experiencing this problem.