Appreciate the love guys- the way I see it is whether you do it or not, the decisions of all players with edges at the short stack all come down to the math. Some players do it by feel and experience, but I find my way in the numbers. Humans are horrible at recognizing patterns, though we really like to think otherwise. You don't necessarily need to have this phd level of understanding to know what's going on, but running through the algebra off the table can help you realize where you net chips a lot faster than just playing and playing and playing.

If you have 2 hours to dedicate, I think I'd recommend looking at situations where effective stacks are between 12 and 20bb, in the small or big blind, and the button or cut off opens for a raise. Guess what their range would be, and enumerate it. (22-AA AT-AK, for example, is 13 x 6 + 16 x 4 = 142 hands out of 1326 total hands = 10.71%) Then say out loud to yourself, “With what hands can they call a shove?” For a villain opening 30%, even if they called all pairs and AT+, they're still folding (30-10.71/30) 64.3% of the time.

To find out if the situation is profitable (and then ~how~ profitable), run the expected value of the situation.

The pot would be (Villain's raise + blinds + antes + your shove)- the first part is multiplying that sum by the % of the time that you get folds. If in the big blind, and villain folds 64% of the time, you would multiply that sum by 64%. Next, add in the times you get called. Multiply 36% x (villains stack + your stack + blinds + antes) x your hand's equity vs his range. If your total number is greater than the number you started with (your stack size), the difference between the two is straight profit. 

Like all things in life, you can make this as simple or complicated as you want/feel comfortable. A follow up would be if youre in a different position, estimate how often you get through all of the villains. No villain will fold say JJ-AA, so you'd take one extra villain remaining as like 98% x 64% to know the % you get folds. Another follow up is to not only see if something is profitable or not profitable (a discrete, binary measure, either 0 or 1, profit or no profit, etc) but to compare how profitable it is and how often you lose what you risk, against other forms of making chips. Hope that helps! I feel like doing 1 or 2 hands could potentially take you that long. So play whenever you can, and simply note some hands under the criteria I listed. Take care!