The use of contact sand to achieve good hydraulic connection between the tension infiltrometer (TI) membrane and the soil is known to introduce an offset between the pressure head set on the bubble tower (h0) and the pressure head applied to the soil surface (hs). The nature and importance of the offset are poorly understood, however. Hence, the objectives of this study were to characterize the offset and to demonstrate its impacts on TI determinations of near-saturated hydraulic conductivity, K(h), sorptive number, α*(h), flow-weighted mean pore diameter, D(h), and number of flow-weighted mean pores per unit area, N(h). The offset, Δh = hs − h0, consists of a constant elevation component and a variable head-loss component. The elevation component increases hs relative to h0, and comprises most of the offset for low TI flux density, q(h0), and large contact sand hydraulic conductivity, Kcs. The head-loss component decreases hs relative to h0, and becomes more important as q(h0) increases or Kcs decreases. The offset has little effect on the accuracy of K(h), α*(h), D(h), and N(h) when these relationships are insensitive to changes in h0. When the relationships are sensitive to changing h0, the offset can change the shapes of the relationships; cause systematic overestimates of the K(h), α*(h), and D(h) values; and cause systematic underestimates of the N(h) values. The amount of overestimate and underestimate increases with increasing offset and should be corrected using a form of Darcy's law to prevent the introduction of systematic biases in TI results.