Geoff emphasized the creating of small value studies in order to work out lights, mediums, and darks before starting a painting. It helps balance the composition as you link the medium-value areas for unity, and, as importantly, determine where to save the white/light areas. (Since watercolors are mostly transparent, you can't use lighter colors over darker ones; you have to save the light areas all throughout the painting process.)
Awhile back I took a workshop from Midwestern artist Andy Evansen, who also emphasizes value studies. I did them for awhile but fell away from the practice. But I'm older and (theoretically) smarter now, and I see the need more clearly.
I'm having a great time with these studies in my 7x10" spiral-bound Stillman & Birns sketchbook, Zeta Series. Typically you use just one color of paint, but sometimes it's tempting to add another color or two, for emphasis/interest or to test out foundational colors for the painting.
Of course, sometimes the value study is better than the ensuing "real" painting. (That is the case re: the top study; I've tried two versions that have bombed.) Or sometimes the study doesn't grab you enough to want to make a painting from it. Whatever the case, I'm having fun trying, and I think this is an important skill to develop within the watercolor learning curve.
