One reason: Launch costs.

DEvelopemnt costs are already done for both, as we also have the design for the rovers in hand.

The launch cost is the same, so which returns the highest scientific yield? Though the MPL instrument suite is different, the rovers are mobile.

Second reason: opportunity costs

We could launch MPL in the next cycle when we have a better idea of exactly where to land it. There was a long exmination made of where to land the rovers, and to my knowledge a similar analysis has not yet been performed for MPL.

The first MPL attempt was rushed, as it was designed to remove barriers to manned Mars flight by 2020. [Goldin intended to convert the Shuttle to Shuttle C for manned Mars flight after the RLV became operational in 2005.] Given that manned Mars flights are now scheduled to occur at least 5 years later, [Shuttle conversion to Shuttle C after CEV operation in 2010] there is no timing restraint providing a reason to repeat that mistake.

MPL is going to require very high precision navigation, and to my knowledge even the atmospheric entry models have not yet been updated with the new data, little less adjusted for gravity.