Does the Pentagon Need an Acquisition Chief??

Title of an article in Aviation Week.  They have one now.  The incumbent, Frank Kendall, claims that cost overruns were 51% before his time and he has reduced them to 5%.  His job is on the line, latest Senate defense authorization bill would remove it and replace it with two lower ranking slots, one for R&D and one for "management and support"  what ever that might be.  Pure paperwork perhaps?
  Acquisition is a serious problem at the Pentagon.  Look at the F35 program, a decade late and zillions over budget.  There was a new Marine One helicopter program that got so far out of line that Obama had it canceled.  The KC-46 tanker is years late and under attack by nit pickers.  I don't follow the new programs as closely as I used to back when I was a serving Air Force officer.  So there has got to be more grief out there.
   Success or failure (cost overruns and delays) rest with program management.  Take F-35 for example.  It's problems can be laid at the feet of F35 program management.  Extra layers of Pentagon paper pushers have nothing to do with it.  
   Every military officer in program management needs to know that his Officer Efficiency Report (his future promotion chances)  rest upon program success.  Bring the program in on time and under budget and you get ranked at the top.  If the program is late or overbudget, you get ranked at the bottom.
   Program management needs to have input to the specification writing.  Many program disasters result from ridiculous specifications, spec that called for unobtainium, or faster than light, or other things impossible to actually make.  Or, gold plating the project with nice-to-have but not really necessary expensive gadgets.  I'm thinking of the Tactical Situation Display in the old F106.  It never worked, and the plane flew and fought successfully without it. Or the C-5 program which sank under the weight of impossible to make requirements.  Or the F35 burdened with an airborne digital networking system, and Intelligence Surveillance Reconnaissance (ISR) systems neither of which are needed in a fighter.  Fighter planes are expensive and should concentrate on air superiority, shooting down enemy aircraft and attacking enemy ground troops.  We have recon aircraft, drones, and satellites for ISR.
   Then program management has to iron out the myriad boggles and whoopsies that come up during the program. Specifications almost but not quite met.  Subsystems that just don't work.  Program management must be prepared to accept small shortcomings when the cost of fixing them is high.  And be prepared to just dump subsystems that aren't working.  And accept cost reduction suggestions from the contractor. 
    Trump needs a good, intelligent defense secretary to sort this stuff out.  The current secdef, Ash Carter isn't bad.  John McCain would be good, he at least knows the issues and knows which end is up.