- I'd like to start by asking a quick favor of readers -- before casting a critical eye toward this week's Extended Standings take a moment to determine who would make the playoffs today according to this week’s region playoffs and Class A power ratings. I’ll wait :-)

The point of that exercise is to remind ourselves that the Extended Standings use only the current season's results to date and the usefulness of early season results can be limited, just as in the region standings and Class A power ratings. However, I'm confident we’ll see in the coming weeks the Extended Standings reach a better conclusion sooner than either of the two models currently in use.

- After week 1, the rankings follow two general patterns – teams with larger margins of victory are rated higher (in line with the second principle of equability in the “Getting it Right” series) and teams are grouped around their classification.

- Perhaps a surprising result is that Class AAAAA fairs so well, but consider that last week Class AAAAA teams went 22-5 against all other classes, including 6-2 against Class AAAAAA. Although this highlights the model has no prior notions about the strength of specific teams or classifications, we can reasonably expect the classifications to correct to their intuitive order as the season progresses.

- Some of the terms used below have not been explained yet in the “Getting it Right” series. I’ll publish Part III and Part IV of the series later today that will explain adjusted wins (adjW) and adjusted losses (adjL).

- Only nerds would be interested in the "Sum of the negative log-likelihood" and "Solution converged in X iterations", but I'll eventually explain those in the "Getting it Right" series as well.

Extended Standings

All data is courtesy of the Georgia High School Football Historians Association.

Please submit errors or omissions through their forums.

Each team is shown with it's overall rank, classification rank, classification, actual record, adjusted wins and losses (adjW and adjL), rating, extended wins and losses (xWin and xLoss), and extended winning percentage (xWin%).

Sum of the negative log-likelihood: 327.8424

Solution converged in 950 iterations