There's no perfect model in existence (it was, but times and demands were different then).

We have too many conflicts:

(1) officers who leave after x years - officers who serve till retirement
(2) officers to lead men in combat - offers who lead support people - officers who do very important jobs without many subordinates
(3) effectiveness - costs
(4) uncertainty about potential - desirability of having officers qualified all the time, instead of adding learning-intensive basic qualifications after many years in service
(5) costs of long education and training - benefits of the same
(6) specialization on military skills - ability to get a job after life in uniform, ability to grasp and respect civilian stuff
(7) need for high quality - need for great quantity (in case of mobilization)



I suspect different career / education / training patterns are justified for different groups of army officers

(a) young officers and former NCOs (promising NCOs get a second NCO course (Feldwebellehrgang) and as a test the opportunity to lead a platoon for minimum 1 year or an equivalent staff task)
(b) civilian experts joining with full civilian qualification
(c) proved officers, selected for staff service (i.G.)
(d) proved officers, selected for leading army field units
(e) useful officers, selected for leading other units

example:

(a) Quick (6 months) lieutenant course (~ "70% solution" for up to company command)

(b) Quick (3 months) officership training (use of personal equipment, survival, regulations).

(c) staff courses and academic studies (military history, logistics, intel, art of war, technology of war) that span years, but with interruptions by normal staff assignments - and much autodidactic learning effort is expected

(d) command courses and academic studies that span years, but mostly command assignments. These officers should not be branch officers, but become combined arms officers.

(e) Branch-related courses to add competence in management and the area of occupation (logistics, for example)


JMA's concerns about wasting too much training on duds should be addressed by this. The young officers would either be proved NCOs or wouldn't get much training till being commissioned.
Young recruits on the direct path to officer would not sign up for many, many years - instead, they would sign up for two years and becoming lieutenant of the reserves. The army could then invite as many as it needs to further active service (signing up for a few years with expectation to move up to captain and probably get a company command minimum for a year, then in a third step become career officers).