If you like betting on basketball matches from NBA or US college basketball you might be interested in knowing how the bookmakers create their lines for under and over betting on total number of scored points by both teams. This question has many different answers, but mine is related to the average performance of both teams.
According to my researches, the basis for the selection of lines for under and over betting is made by some very easy to get betting data. It is the average number of scored points made by the hosts in their home games, the average number of points allowed by the hosts in their matches at home and respectively the same points, but for the visiting team.
This applies to more than 80% of the NBA matches where, the differences between the results of the above pointed method and the bookmakers’ lines for under and over is within 2 or 3 points.
In those 20% of the matches in which we can see a larger difference between the stats I pointed above and the offered by the betting houses lines and it may be associated with various other factors, such as injuries to key players or big bets in some other directions.
It is important to know that exactly in the deviation we can search for betting profits, especially if we properly understand the reasons behind the bookmakers’ decision to set higher or lower lines.
This is actually the biggest problem. I really know how they set the points, but how can we use it?
The options are few. One of them is to find the relationship between the standard deviation of the way the line was made and the number of matches for each team, finished with unders or overs. The idea is that if we have a team which has finished more than 60% of its matches in one direction under or over it is possible the betting house to artificially lower the lines to avoid bets for the easy to guess outcome.
Of course, these are simply some assumptions without any evidence, but anyway, what is clear is that there is something out there. Something we can use. Something which we should find first.