BTW, conceptually, the same can be accomplished by orienting the Blumlein microphone in Mid/Side orientation instead of L/R orientation + the coincident cardioid. This arrangement, with two Mid microphones pointing the same direction, may make the Mid/Side ratio adjustment stuff in the post above easier to visualize.
Then, if one were to use an omni instead of a cardioid, the result is horizontal-plane native-array ambisonics, with an even wider range of "variable Mid" pattern, extending beyond cardioid to subcard and omni Mid options.
But in non-theoretical real life, having all three capsules pointed more "on-axis" to the sources of interest (the stage and PA to either side of it) might be better, with the bi-directionals facing +/- 45 degrees and the cardioid facing directly forward.