wildbill is dead on in saying that regression to the mean is one of the most valuable concepts in handicapping.
simply put, bettors are very fickle. they make way too much of a win and overreact to much to a loss.
even the most inexperienced gambler is familiar with the 'letdown' angle.
but the good bettors are the ones who are able to identify teams that are UNDERperforming and catching more points than they would normally be, or laying less than they would normally be. just because a team suffered an inexplicable bad loss (or even string of bad losses) doesnt mean that the talent suddenly disappeared from that team. northern illinois was a good example of that in football this past week. they had underperformed for a handful of weeks in a row, but the talent on that team has been there all along. it didnt go anywhere. the final result of the cmu/niu game is inconsequential to this point. the most important thing is that the line of cmu (-4) was just a gross overreaction to some poor play by northern. even if cmich has won by 14 and covered the spread, it still wouldnt have been the correct number.
a lot of people think an ATS result is the end-all, be-all. just because a team wins for you doesnt mean the spread was right or wrong. if you bet into overinflated lines enough, you will eventually pay for it. and if you consistently take advantage of generous lines that have been overadjusted, you will be a winner over time. but people let a few early results dictate whether or not an approach was correct. for example, its easy to abandon this approach when you back a team like the buffalo bulls in college football and they take it on the chin by 30 one week and you feel like you threw money down the drain. but if you are committed to backing teams that present true value, you will be successful.
this is obviously one of those topics that is so broad that its tough to attack it on an internet forum. but wild bill is dead on with his original point.
