Usually an oscillator is best analysed as a negative resistance oscillator when the feedback path isn't obvious. This is generally the case for oscillators up at UHF. Have a look at the JFET circuit below. This can be made to oscillate up near 1GHz with suitable inductor values but the circuit below should oscillate around 500MHz.
Looking back into the drain you should see a negative resistance in series with maybe 1.5 to 2pF across maybe 250-700MHz. So adding the 50nH in the drain makes this resonant up at 500MHz and the negative resistance should be a lot bigger than the loss resistance in the 50nH inductor and the decoupling cap.... so it should oscillate strongly
But you need to work out what is happening inside the JFET and where the feedback is. This then holds the clue as to why it can generate the required negative resistance in series with a small capacitance.
A crude equivalent of the circuit is given in the second image. This is a tank circuit resonant up around 500MHz where the (-20 ohm) negative resistance in pink overcomes the real loss resistance in the inductor. So it should oscillate