Target: approximately 2 mA from a 12 V supply with enough VDS for signal swing. The divider gives VG=4.00 V.
Solving the square-law/source-feedback equation gives ID about 2.08 mA, VS about 0.98 V and VD about 5.75 V, so VDS is about 4.77 V.
The exact simulated current shifts because LAMBDA adds finite output conductance. That discrepancy is useful: hand equations choose the topology and values; the SPICE model then refines the result.
