You can only be successful in them if you are consistent with the winning but if you are not, a single loss will ruin the party. Imagine, if you bet three matches with 1.3 odds, it will take a single loss to clear all your 3 initial winnings, and it will take you 3 winnings to meet up with a single loss, so how convenient is that? The right risk management is not included in this kind of low-odd betting so I try as much as possible to avoid them.
This is the point of the experiment, can the top team beat the odds or not!?
This is still the same thing but in a different format. The top team will beat the odds more times but luck is also needed here because no matter how better a team is against the other, there is always a time when the underdogs win which will ruin the party. A good example was yesterday when Nottm Forest beat Liverpool in their home, Nottm Forest of all teams? This rarely happens. However, some underdogs' wins are even more rampant than this, so it's not a matter of certainty win for the top team even if their chance of winning is high.
Bookies offer odds, but they don't offer really chances for it more like risks for gains percentages, if those were really chances you could go by all results and turn the whole thing into a predetermined outcome, bookies have no way of dealing with you a blow if Madrid for example goes on a rampage and beats every single team in the championship, and if they fail once in such a streak the odds for the next match after a failure would definitely pick up.
Just like anyone who knows how to analyse matches and the possible outcome very well, they would have known the possible odds of bookies on matches depending on the distinction of strength/weakness of the meeting teams. For this, it is an automatic way of knowing the chance of winning or losing of either of the teams, so invariable, those with low odds have the higher chance of winning especially if your analysis agrees with the odds of bookies, so technically they (bookies) has already hinted the chance of winning with the odds.