jayjay, I hear you, that's why I formulated my initial statement as an opinion (I think) rather than a fact - I didn't categorically say that it can't be done over a prolonged period of time but I have serious doubts it can be done..
The starting premise is that you're trying to find value pre-match betting in, say, a sport like tennis where the favourites are usually extremely low on the odds. The fastest way of securing profit from this type of betting is by picking winners consistently. If you follow a strategy of picking, say, all favourites you might end up getting a 10%-20% increase of your bank throughout the course of a season; which is hardly a profit you can use as a regular income. Then the issue of discipline could play an enormous part - what happens when you have a period of "drought" where picking favourites is simply not cost-effective - and you end up losing money? ( a period we're witnessing since the French Open and which is continuing until now ). What happens if that period lasts for months and months? How do you maintain your discipline waiting for the better days that will inevitably come?
Then of course backing underdogs is surely a losing strategy because the underdogs win at the most unexpected of times. Who would have thought that Fish would beat Federer or that Henin would lose to Bartoli for example? I don't think even 1% of bettors could have predicted these results prior to the event, hence my statement that either strategy is unsuccesful.
However, I'd be interested to hear the "grey" version of a successful bettor - which is always the way things work.