I would say those two bids were competitive, and in duplicate there is incentive to compete more keenly at lower levels, so maybe light opening is a misnomer. 

'Forget...' is 100% correct in saying duplicate is about avoiding mistakes, and one follows the system precisely in 1st and 2nd seats. However, passing a suitable bid is also a mistake! 11 points with 5-4 in first seat would meet 'rule of 20' as does a 10 point 6-4 hand. An unacceptable light is 11 points balanced; Some 12 point hands shouldn't open, while in 3rd and 4th Seats one effectively adopts a different bidding system that's even lighter.

The top players get good scores from bidding marginal situations well. They recognise subtle differences and bid according to the context. It is not bidding light per se but rather bidding to the situation with well-defined exceptions to 'basic' rules, so partner can still read the hand EXACTLY, and make the marginal decision needed. An opaque 'error-range' around the normal bid is always bad.

In duplicate there's more to gain from competing at a lower level, less to lose on doubled contracts, and an incentive to sacrifice; going down -1 to prevent opponents making a contract is usually better, but this comes from precise bidding at low levels and not the light bidding.

In the long term, down -1 saving 10 points 9 times out of 10 outweighs a -50 once in 10, and getting 60% 9 times and 0% once is good bridge, while 60% 8 times and 0% twice is bad bridge! Margins are thin so one's generally conservative to avoid mistakes, and 'Light' is bad where partner may misinterpret and place the contract wrongly.

But it's not a grey area, I would conclude the principle is: If you don't know which 'Light' exception you're following then don't bid it, but when you do know, then do it! And improving is about learning which marginal condition you and your partner should or shouldn't bid.