Welcome. I didn't read the article, so I am going by your post.
Blind defense is one area that is not covered well in poker literature (that I have seen).
In the case described, 30-60, the blinds will be 20 and 30, which is a factor for two reasons. First it means the big blind gets more odds on his money. Second, it generally means the the button will be looser in attacking the blinds (and properly so).
Against most typical button hand ranges calling with K3off in the big blind here is correct. Yes, correct.
This is not a case of making money. As is often the case with blind defense, you are trying to lose less money than you would if you had folded.
Just for reference, I ran three simulations, each for 100,000 hands. This included rake and a 1 dollar toke. These simulations do play the hands out to the end.
Button raiser is tight, you lose $22.65 per hand
Button raiser is average, you lose 23.15 per hand.
Button raiser is loose, you lose 20.52 per hand.
Now, what you see is that you are losing less than what you would lose if you folded your 30 dollar blind. Those amounts are significant, and are enough, probably, to overcome whatever errors might exist in the simulations.
Keep in mind, the profiles I used have a relatively "loose" starting range, against a tighter (like someone who does not adjust to the situation of being on button against two blinds), your expectation decreases. Tight enough, and you would begin to lose significantly. One thing that tends to counterbalance this, is often such overly tight players (for the situation), also don't play headsup properly either. This is speculation, but is generally probably true.
So, specifically, you should fold the K3 when the opponents hand range is about 66 and up, any two cards ten and up, A8o up, A5s up, K8o up, K9s up, 89s, J9s, Q9. This hand range is generic, and is not specifically designed to beat K3 (as in does not include the weaker kings that would dominate K3). If the opponent hand range is close to this or tighter, or the opponent plays well, fold. I included the specific hand ranges because our ideas of what are tight and loose etc, are likely very different.
One way to look at this situation is from the point of veiw of the button. If you are the button, do you want the big blind to call? Not usually. Only with your best hands would you like action. You are quite happy with most of your hands to take down the 50 dollars in blinds.
My personal (of course anecdotal, no one person's results are enough), online hands, indicate that these simulations actually tend to underestimate the profit of K3 here. But this might have something about how I would position myself in a game where blind stealing is important.
You are correct in thinking that less than expert players should consider folding, because these simulations were run with a profile who is competent post flop. I folded in such situations for a long time. In full ring games, you will not "lose" much. This is not true of shorter handed games. There is also complicated "poker eco system" arguments that do affect things a few dollars here and there.
So if you have doubts about your abilities post flop, I agree that you should fold until you don't have to ask. I know this sounds harsh, but losing a little (theorectically) is better than losing more in situations that are hard to figure out.
Strangely, when I first read your post, I thought it said that the small blind had cold called. This was right in line with something I had been emailing someone about.
Now guess what happens when the small blind cold calls and you have K3 in the big blind. Simple question. Is is better or worse situation than the one described in your post?
Notes: I used Turbo Texas Holdem to run those sims. Sims are not the be all and end all, and have problems and limitations. Don't put all your faith in what they tell you. Use your mind, think for yourself, don't just copy what someone has in a book, or what Abdul has on his website.