The author of the blog post considered bookmakers as the ones laying the outcomes. What if one considers the layers as other bettors in a betting exchange?

This scenario would be more interesting than trying to arbitrage bookies. Many bettors do bet based on emotion or some voodoo magic. Betting exchanges like BetFair are not laying anything, so they don't care. If some smart bettor is arbitraging the other bettors, the betting exchange is more than happy to collect a fee off the arbitrageur's profits.

Betting exchanges are also even more watched by people looking for arbitrage opportunities than the major bookmakers and more volatile in the short run (often they'll be used to form one side of an arbitrage bet). You're dealing with a more knowledgeable betting crowd and rates that are corrected more quickly by market forces (and a spread of unmatched bets/lay options that have already been placed). As the betting exchange fees/commissions still drive a wedge between effective market rates on the betting and laying side you have to be even quicker to snag the profit (or correctly predict the direction of the movement of odds between placing the bet and the lay)

If it were novel research, the author would have been an idiot to blog it instead of publishing it in some peer-reviewed journal. No one claimed that it was novel. I thought it was interesting, that's why I submitted it to HN.

As I said, perhaps it was just me. But how hard is a footnote?