Thanks to all who participated. The final projection uses 76 entries (after I removed 4-5 crazy-to-impossible entries), and came out much more realistic that I originally thought. Full spreadsheet of the data is available here.

Obviously, we have a bit of a bias when making these so I thought I'd do some PECOTA-style adjustments to our average to come up with a more well rounded picture. For an unbiased projection I simply subtracted 5% from each category (10% for SLG). Then I gave you a High and a LOW projection by adding or subtracting the standard deviation.