There is some technical support for the results Steve is seeing. While this probably won't be as true for recurved tips as for longbows, the basic geometry of the limb-string relationship would indicate that the long string (up to a point) would generate a force-draw curve that would be slightly shallower than the braced bow.
To explain: Longbows typically use a string about 2-3" shorter than the NTN dimension, and when drawn, an idealized limb bends in a circular arc. Now, only the force that is perpendicular to the limb tip bends the bow limb. That force directly related to the tension in the string and the angle of the string to the limb. Then, the draw force is directly related to the angle of the string to the dirction of the draw.
A long string that is only a few inches longer than the braced string, doesn't change the geometry all that much, so the force-draw curve is close to the same. As the long string gets longer, this becomes less true.
Also, what is different between the long and full brace string, is the compression that is put down the limb from the braced string. This would be the parrallel force that is the result of the tension in the string at brace and the angle the string makes with the limb. This initial compression diminishes as the bow is drawn and can get down to zero if the string goes to a 90 deg angle to the tip at full draw.
Ken