Lucas: Green has ramp/fixing, Blue has card draw/manipulation to find lands. R/W generally wants to curve out as fast as possible, so it can't afford to wait to draw the colour it needs; B/W suffers because black and white both tend to have stringent mana requirements.
If I remove stuff like painlands/Scarslands, I get this configuration:
2 Celestial Colonnade, 2 Flooded Strand, 2 Hallowed Fountain, Azorius Chancery, (Mystic Gate?)
2 Creeping Tar Pit, 2 Polluted Delta, 2 Watery Grave, Dimir Aqueduct, Sunken Ruins
2 Bloodstained Mire, 2 Blood Crypt, Graven Cairns, (? more Mires/Crypts? another Cairns maybe?)
2 Wooded Foothills, 2 Stomping Ground, Grove of the Burnwillows
1 Stirring Wildwood, 2 Windswept Heath, 2 Temple Garden, Horizon Canopy
2 Marsh Flats, 2 Godless Shrine, Fetid Heath
2 Verdant Catacombs, 2 Overgrown Tomb, Twilight Mire
2 Misty Rainforest, 2 Breeding Pool, Flooded Grove
2 Scalding Tarn, 2 Steam Vents, Cascade Bluffs, Izzet Boilerworks
2 Arid Mesa, 2 Sacred Foundry
so UW (7/8), UB (8), BR (5), RG (5), GW (6), WB (5), BG (5), GU (5), UR (6), RW (4)
This is clearly uneven, and the colour combinations with the lowest amount of manafixing are the ones that need it most. I guess my question is, is there a compelling reason to play e.g. Battlefield Forge over another Sacred Foundry, given that Foundry is arguably a better land than Forge by itself and lets the RW deck find more fixing in the form of any other fetchland (and also works with Eternal Dragon, KotR, etc); and, if I'm adding more shocks, which colour pairs would want an added fetchland instead?