SI,
Here is my basic handicapping protocol:
Preseason -- Before the season starts, I wrap my head around all the changes each team has made during the off-season. I analyze the schedule and make NFL futures selections for win totals. This work will get me ready for the regular season.
1. I have a personally selected saved trend file with a few hundred systems. This is really the MEAT of my handicapping. They get run automatically every day and I get a file of all the ones that are active for the upcoming week of action. On the average 10-20 of them are active. Here are the records of the first ten systems in the database: 20-54, 46-11, 62-33, 48-4, 93-41, 19-0, 6-48-2, 25-2, 129-69-2 and 663-479-29. These systems have been consistent winners over the years and are not over-fitted. These systems give me a base of 5-10 teams that I'm going to look at further for possible plays.
2. I also have a computer generated file that; searches through 250,000 handicapping situations each week, filters the ones that are active for the upcoming games and then groups them by game, ranking with the best on top. Once I know what games I'm looking at, I look at the team trends in past performance, focusing on average margin, timeliness, and whether the trend makes handicapping sense.
3. I research the individual teams to ascertain whether the conditions under which the trends and systems were developed still exist. For example, once Peyton Manning moved to the Broncos, we threw out most of Denver's trends in past performance -- especially trends that involved the passing game.
4. Next I look for intangibles that could outweigh the trends and systems. For example, a key injury, a coach has cancer or other tragedy, extreme weather, or simply an extreme change in the character of the team.
5. When everything comes together, I write the game up and post it on pregame.
Over my 30-plus years of handicapping experience, I have found that more and more I'm not handicapping the teams themselves, I'm handicapping the bettors. That is, my systems uncover spots in which the bettors vastly over or underestimate the relevance of recent events. For example, the week one performance of a team that got knocked out of the playoffs in the first round last season.
Dr M.