We examine the ability of teleseismic P waves to provide a timely image of the rupture history for large earthquakes using a simple, 2D finite‐fault source parameterization. We analyze the broadband displacement waveforms recorded for the 2010 M_w∼7 Darfield (New Zealand) and El Mayor‐Cucapah (Baja California) earthquakes using a single planar fault with a fixed rake. Both of these earthquakes were observed to have complicated fault geometries following detailed source studies conducted by other investigators using various data types. Our kinematic, finite‐fault analysis of the events yields rupture models that similarly identify the principal areas of large coseismic slip along the fault. The results also indicate that the amount of stabilization required to spatially smooth the slip across the fault and minimize the seismic moment is related to the amplitudes of the observed P waveforms and can be estimated from the absolute values of the elements of the coefficient matrix. This empirical relationship persists for earthquakes of different magnitudes and is consistent with the stabilization constraint obtained from the L‐curve in Tikhonov regularization. We use the relation to estimate the smoothing parameters for the 2011 M_w 7.1 East Turkey, 2012 M_w 8.6 Northern Sumatra, and 2011 M_w 9.0 Tohoku, Japan, earthquakes and invert the teleseismic P waves in a single step to recover timely, preliminary slip models that identify the principal source features observed in finite‐fault solutions obtained by the U.S. Geological Survey National Earthquake Information Center (USGS/NEIC) from the analysis of body‐ and surface‐wave data. These results indicate that smoothing constraints can be estimated a priori to derive a preliminary, first‐order image of the coseismic slip using teleseismic records.