Translating SP WAR to "#1/#2/ect"

This isn't exactly something easy to google so I figured I'd ask here :)

Hitter is generally easy.  ~2 is an average player, ~4-5 is all star territory, ~7+ is MVP territory.

SP though have the whole "#1, #2, #3, #4, #5" deal.  So, my question is, what's the WAR of a #1?  What's the WAR of a #2?

From playing around with calculating WAR (From Fangraph's primer), here's roughly what I came up with for a NL SP this season (off of 180 IP, which would be 30 starts at 6 IP per start.  I believe this is the equiv of 150 games played for position players, correct me if wrong, I tried to figure it out myself):

#5 = 4.60-5.10 ERA/FIP (0-1 WAR) (aka, bad bench player)

#4 = 4.60-4.10 ERA/FIP (1-2 WAR) (aka, good bench player)

#3 = 3.70-4.10 ERA/FIP (2-3 WAR) (aka, average starter)

#2 = 3.30-3.70 ERA/FIP (3-4 WAR) (aka, above average starter)

#1 = 2.50-3.30 ERA/FIP (4-6 WAR) (aka, all star)

Cy Young/ACE/whatever = Sub-2.50 ERA/FIP (6+ WAR) (MVP candidate)

That look about right?

My main reason for asking actually has to do with prospect grades.  I like John Sickel's system, but using projected WAR to give grades.  So for instance, for SP, it's roughly

#5 = C

#4 = C+

#3 = B-

#2 = B

#1 = B+

ACE = A

I guess the main that is it takes a 4.10 ERA to be the equiv of a average position player is just kinda hard to take.  For grading system, maybe something closer to

5/C = -0.5 - 0.5

4/C+ = 0.5 - 1.5

3/B- = 1.5 - 3

2/B = 3 - 4.5

1/B+ = 4.5 - 6

ACE/A = 6+

?

That'd instead give

5/C = 4.85-5.30

4/C+ = 4.35-4.85

3/B- = 3.70-4.35

2/B = 3.10-3.70

1/B+ = 2.50-3.10

ACE/A = Sub-2.50

Does that look better?

Obviously this is really objective, but I figured this would be a nice place for open discussion regarding it.