Power Ratings: NFL Week 2

Week two of the NFL season is a difficult week for bettors and bookmakers alike. Each year there is a six-month football hiatus between February and September which culminates in an unloading of finances and emotion on the opening Sunday. This football betting overload almost always results in over analysis and reaction. Bettors over react to what they saw in the opening week and bookmakers over react in judging what they think the bettors saw. Because of this games in week two become hard to price and lines move quickly and often.

This is true again this year as just one game this week has remained static with its price. The average line move per game is a very high 1.5 points. This is understandable considering that last week 11 of the 16 games were decided by 6 points or less – an NFL record. Also, the fact that more than 1/3 of the entire league has quarterback uncertainty (first time starter, new team, injury, suspension) is not doing anyone any favours either.

If you are looking to avoid over reacting and keeping your ratings steady throughout the season, it is important to remember that every adjustment must stay within the boundaries of the season. For most bettors, rating teams from week to week is focused solely on the match-up at hand, and not taking into account the bigger picture. Moving teams within the bigger picture was a lesson I learned very quickly in my time working at Paradise Games. The first time I created a futures market, I priced each team in the NFL based solely on how I thought about how each would perform. I ended up extremely top heavy and priced the market to 175%!

The easy way for bettors to adjust teams accurately and fairly is to stay in line with the seasons win market. This was something I went over in detail last week. The main point in the article was that half a win is worth one point on a neutral field. The reason for this is that an average team without margin priced in would be priced at EV each game, which has an implied probability of 50%. Over the course of 16 games, a team with a 50% chance in each would be expected to win eight games. One point in a margin free price would move the line to -115 (1.87) and leave an implied probability of 53.44%. A team with a 53.44% chance in each would be expected to win 8.5 games (16 * 0.5344 = 8.5).

With that in mind, unless there is an injury or significant change within a team, no adjustment in power ratings from week to week should be more than 1 to 1.5 points at the maximum. It is very rare that a case can be made that a team becomes a full win or a full loss worse based on the performance of one game, especially early in the season.

I weigh week one very lightly. If a team performs poorly in their first game in over six months, it is not time to raise any flags. From week one to week two, I have adjusted the ratings of just five teams in the league, moving each a half win up or down.

The two teams I adjusted down are Kansas City and Seattle. This may be a surprise to many considering that both won their games in week one. However, just because a team won does not mean their rating must go up, likewise, just because a team lost does not mean their rating must go down.

Seattle entered week one as the highest projected team when combining season win totals with the Pythagorean expectation at 12.5 wins. As a double digit favourite they needed a last second score to beat the Dolphins. While their defence looked strong, it was their offence, particularly their offensive line which made me sure this team was not a full win and a half better than second highest rated Pittsburgh. With five starters on the offensive line playing in new positions, the Seahawks dynamic offence will be severely limited for weeks to come until continuity is developed.

The worry in Kansas City is on the other side of the ball. KC is rated tied for the fourth highest projection in the league combining win totals with the Pythagorean expectation at 9.5 wins, the Chiefs were listed as 6.5 point favourites at home to San Diego. While the Chargers do have a strong offensive line, the absence of Justin Houston and Sean Smith was glaring. The Chiefs, who rely on their pass rush and shut down secondary suffered, failing to generate any QB pressure and allowing the Chargers to put up 27 points. KC still has to find someone to fill the spot of Smith and wont have Houston back until Week 8. Tough times ahead for this typically strong defence.

The three teams I adjusted up were Houston, Pittsburgh and Tampa Bay. All three have their quarterback situation sorted and played the three best all around games of the opening week. Pittsburgh made their case to be the highest rated team in the league with a big win on the road Monday Night. Tampa Bay showed that they have a ton of continuity on offence and should be at worst considered a league average team. Houston met their above average expectations and showed that their defence is as strong as ever and eliminated any uncertainty about their quarterback and running back.

My ratings in the opening week went 10-5-1 ATS with teams listed at a difference of 3 points or more going 3-2 ATS. This week there are four teams with a difference of more than a field goal, Tennessee, Pittsburgh, Dallas and Indianapolis. I have bet two of these games myself and advised clients to do the same. If you are not a client, you can become one here.

Untitled.png

Advertisements

Author: Adam Chernoff

Adam went from small town Saskatchewan to being an Oddsmaker in Medellin, Colombia. On his blog, adamchernoff.com you can read stories about his wild decade of gambling.

2 thoughts on “Power Ratings: NFL Week 2”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s