Nominals incorrect in FCF

In the code, are you aligned to -B-? Based on what I am seeing, -B- is a cylinder? If so, are you leveled and zeroed to B? Geometric Tolerancing will create an alignment based on the datums you pick. What happens when you change the fitting condition from DEFAULT to LEAST SQUARES in Geo Tol?  

The base alignment is leveled to a plane and rotated about lin2. The print I am using calls out position from the width to the cylinder. so I put it in as it was on the print. I am a novice at best when it comes to using the FCF function. My company has yet to send me for training so I am using what little I know. I changed the Datum math to LSQ and it stayed the same

try removing the (M) for the B in the dimension, see if it changes back to zero.  I'm guessing that the MMB is doing a datum shift. 

Datum shift won't affect the nominal.  It will be due to the difference between the alignment (levelling to a plane, rotating to a line and origin on the cylinder) and the DRF that gets created by the geometric tolerance command (primary datum cylinder B only).

Any ideas on how to fix it? Should I add in the other alignment datums even though the print doesn't call them out?

This is basically what im working with the position of the keyway to the pilot diameter at mmc on both.

No.  If the print only shows datum B then the geometric tolerance command is correct, it's just that your active alignment (which is what the edit window numbers relate to) is different.  Rather than add datums that aren't shown on the print to the dimension, you should try to create an alignment that approximates the DRF behaviour.  As bfire85 said, the alignment should really level to the datum B cylinder and also origin on it.  You would rotate to the width.  That should at least get your nominal values closer.

made no difference unless I did something wrong.

You would need to measure your width AFTER you've levelled and origin to the cylinder, then rotate to the width to complete the alignment.  Just adding an alignment after you've measured everything will have no effect.

I noticed most of the vectors are slightly off in your earlier screenshot too so they would need to be cleaned up.  The sequence should go something like this...

Thank you so much everyone for all your help. Hopefully soon my company will send me for training. Neils solution worked.