I recently listened to most of an experimental homebrewing podcast. The topic was expected ibu's vs measured. I have not listened to the lead up cast but the experiment was to provide listeners with an APA, IPA and IIPA recipe. Listeners brewed these and submitted them for testing. Across the board there were discrepancies of up to 30%. Glenn Tinseth was on hand to discuss the results. What I found really interesting is that Glenn said that his formulas are based on whole leaf hops and not pellets. 
I find it interesting that even what we consider to be the go to formula really isn't all that accurate. 
Podcast

Wish you hadn't posted this Lumpy; a topic I could write on all day :). (I will look forward to listening to the podcast though when next driving or on the tools.)

Also glad you brought it up though as I have 25 minutes up my sleeve right now and am going to use that to dig up at least some old posts I remember that relate exactly to this. Linking them here will be useful for me, maybe no one else though?

35 Different Results from 35 Brewers given the Same Recipe
A Pic That Shows Why Brewers Should Be Wary of Quoted IBU's
What Should We Do With IBU's? 
Laboratory, Perceived and Estimated IBU's

Really need to consolidate all that info - the above is just a bit of it 

The BIABacus adjusts the hop amount weight between the "FL" flower setting and the blank setting.  His formula does align up with BIABacus "FL".

I realize there is already a ton of documentation on the inaccuracies of ibu calculations. What I found most interesting is the fact that Glenn used whole hops in his testing and believes that once the hops were covered in the sugars in the wort that hop oil extraction dropped and that pellets are probably more efficient.