Solving this kind of problem is fundamental to structural engineering of wood construction. The guiding theory is called dowel bearing strength. The equations for this are given in section 11.3 of the National Design Specification for wood construction.
If we approximate the "dumbell" as an infinitely strong bolt being loaded in double shear by infinitely strong side members, then the equation 11.3-7 governs.
The allowable compression parallel to the grain for this scenario is given as: Z = D * lm * Fem / Rd
where D is the diameter of the bolt (5/16) lm is the thickness of the main member (1.5 inches) Fem is the dowel bearing strength for the wood species in question. If we assume Douglas Fir with a specific gravity of 0.5, the dowel bearing strength is 5600 PSI. Rd is a reduction factor. In this case of perfectly parallel to the grain loading, Rd = 4
so Z = .3125 * 1.5 * 5600 / 4 Z = 656
The wood can be relied upon to carry a weight of 656 pounds for 10 years. This can be increased by a factor of up to 1.6 for "short duration" events (approximating seismic loads or wind gusts) to 1050 pounds.
There is approximately a 3:1 safety factor in this number, so we would not expect it to fail at less than 3000 pounds in a quick test. A particularly good specimen might go to as much as 7:1 or 7350 pounds.
The proposed additional drilled holes are considered of no consequence from an engineering design perspective, since the spacing is greater than 5 times the diameter of the fastener. The wood has sufficient shear strength to transfer the compressive load to the neighboring fibers and thus around the holes below the one carrying the load. After all, this kind of loading is exactly what the trunk of a tree has to support as it is growing -- carrying the weight of the crown down past branches and woodpecker holes.