Mikie2times Posted September 21, 2006 Posted September 21, 2006 I’ve been working on these ratings for awhile and will continue posting weekly updates. Theory- I come to my ratings by taking the total yards a team has accumulated, then dividing that number by the number of potential yards a team could gave gained. So if a team’s starts a drive on their own 30 yard line they could conceivably gain 70 yards on that drive. If they only gain 35 yards they would get a 50% rating, 70 would equal a 100% rating. Once I get these numbers I take the offensive and defensive ratings, and pit them against strength of schedule. The original net difference is never used, instead I take the strength of the defenses a team has faced, and adjust their original offensive rating based on the defensive SOS. I do the same process to the defense, the combined result is the teams overall power ranking. So as a whole the rankings are very accurate, but more importantly the offensive and defensive ratings are very accurate individually because they are also adjusted to there opponents. If any teams seem out of place remember it’s still very early in the year. I compared mine to Football Outsiders and we seem to have many of the same teams out of place. TEAM OFF IND 29.7% CHI 22.1% ATL 18.6% SD 13.2% PHI 12.8% SF 8.7% SEA 6.1% NYG 4.3% JAC 3.9% CIN 3.8% NYJ 3.6% NE 3.2% NO 2.0% HOU 0.8% DAL 0.2% MIN -0.4% DEN -2.4% ARI -3.9% GB -4.1% PIT -6.2% STL -6.7% WAS -7.0% MIA -7.4% BAL -8.4% KC -8.9% TB -9.1% CLE -9.3% TEN -11.7% BUF -14.7% CAR -18.4% OAK -19.3% DET -21.9% TEAM DEF BAL 20.5% SD 19.6% JAC 15.0% DET 12.4% DAL 10.6% NO 8.6% BUF 6.2% SF 4.7% CIN 3.6% ATL 3.1% PIT 2.7% MIN 2.5% CHI 2.1% KC 1.9% CAR 1.8% OAK 1.7% STL 1.3% PHI -0.2% SEA -1.0% NYJ -1.4% ARI -2.0% WAS -3.6% DEN -4.0% MIA -4.6% NE -5.4% GB -6.0% TB -6.6% IND -7.6% NYG -7.7% CLE -8.3% TEN -15.0% HOU -30.8% TEAM POWER SD 32.7% CHI 24.1% IND 22.1% ATL 21.7% JAC 18.9% SF 13.4% PHI 12.7% BAL 12.1% DAL 10.8% NO 10.5% CIN 7.4% SEA 5.1% NYJ 2.2% MIN 2.1% NE -2.2% NYG -3.4% PIT -3.5% STL -5.4% ARI -5.9% DEN -6.4% KC -7.0% BUF -8.6% DET -9.5% GB -10.1% WAS -10.6% MIA -12.0% TB -15.7% CAR -16.6% CLE -17.6% OAK -17.6% TEN -26.7% HOU -30.1%
Alaska Darin Posted September 21, 2006 Posted September 21, 2006 I'd certainly like to know how Miami is above us. Did we not just beat them in their stadium? Are they not sporting a winless record? Note to "journalists": It's OK to rank one team ahead of another - because POTENTIAL means you're not worth a damn YET!
Mikie2times Posted September 21, 2006 Author Posted September 21, 2006 I'd certainly like to know how Miami is above us. Did we not just beat them in their stadium? Are they not sporting a winless record? Note to "journalists": It's OK to rank one team ahead of another - because POTENTIAL means you're not worth a damn YET! 781272[/snapback] I got Miami slightly below us, New England slightly above us. The Jets seem way overrated by my ratings, that should correct itself in a couple weeks. What I like this early on is not many teams have lost in these rankings to teams worse then them. For only 2 weeks of stats, and very little true SOS, I'm very happy with the way these look.
Alaska Darin Posted September 21, 2006 Posted September 21, 2006 I got Miami slightly below us, New England slightly above us. The Jets seem way overrated by my ratings, that should correct itself in a couple weeks. What I like this early on is not many teams have lost in these rankings to teams worse then them. For only 2 weeks of stats, and very little true SOS, I'm very happy with the way these look. 781280[/snapback] I didn't mean you personally, sorry for the mistake. The PP I saw today somewhere in my surfing had them above us. Again, sorry for my wording.
Mikie2times Posted September 21, 2006 Author Posted September 21, 2006 I didn't mean you personally, sorry for the mistake. The PP I saw today somewhere in my surfing had them above us. Again, sorry for my wording. 781284[/snapback] I misunderstood somebody once, ONCE!
bartshan-83 Posted September 21, 2006 Posted September 21, 2006 I misunderstood somebody once, ONCE! 781287[/snapback] You shouldn't hang me on a hook Johnny....
BuffOrange Posted September 21, 2006 Posted September 21, 2006 I like your idea. Only flaw is as you pointed out, SOS isn't worth a damn at this point and probably won't until around Thanksgiving.
Mile High Posted September 21, 2006 Posted September 21, 2006 I get your idea. But why strength of schedule? In todays NFL the parity of the teams alone will kill this theory. For example: Take for instance Miami this year . They were 9-7 last year. And this year they'll be lucky if they win 6 games. To me strength of schedule shouldn't be a factor in how you are ranking your teams. Looking at your list you have N0 and SF in your top 10. Is that due to the division of total/potential? If you ask me rankings should be part gut and record of that team with some emphasis on offensive and defensive stats. Just my opinion though.
crazyDingo Posted September 21, 2006 Posted September 21, 2006 Bills are ranked number one in my rankings. I take the animal on the helmet, divide it into tasty sections and grill it to perfection. I then factor in whether or not the helmet's color might make a good dipping sauce. Bills Bears Broncos Cardinals Dolphins Patriots (Human flesh with a mercury dipping sauce hurts their ranking) EDIT: This will be the last week of my rankings since the Flaming Redhead just informed me the Bears actually have the letter C on their helmets, not a bear. Imagine the thought of all that black bean dipping sauce going to waste...sigh.
Shamrock Posted September 21, 2006 Posted September 21, 2006 I’ve been working on these ratings for awhile and will continue posting weekly updates. Theory- I come to my ratings by taking the total yards a team has accumulated, then dividing that number by the number of potential yards a team could gave gained. So if a team’s starts a drive on their own 30 yard line they could conceivably gain 70 yards on that drive. If they only gain 35 yards they would get a 50% rating, 70 would equal a 100% rating. Once I get these numbers I take the offensive and defensive ratings, and pit them against strength of schedule. The original net difference is never used, instead I take the strength of the defenses a team has faced, and adjust their original offensive rating based on the defensive SOS. I do the same process to the defense, the combined result is the teams overall power ranking. So as a whole the rankings are very accurate, but more importantly the offensive and defensive ratings are very accurate individually because they are also adjusted to there opponents. If any teams seem out of place remember it’s still very early in the year. I compared mine to Football Outsiders and we seem to have many of the same teams out of place. TEAM OFF IND 29.7% CHI 22.1% ATL 18.6% SD 13.2% PHI 12.8% SF 8.7% SEA 6.1% NYG 4.3% JAC 3.9% CIN 3.8% NYJ 3.6% NE 3.2% NO 2.0% HOU 0.8% DAL 0.2% MIN -0.4% DEN -2.4% ARI -3.9% GB -4.1% PIT -6.2% STL -6.7% WAS -7.0% MIA -7.4% BAL -8.4% KC -8.9% TB -9.1% CLE -9.3% TEN -11.7% BUF -14.7% CAR -18.4% OAK -19.3% DET -21.9% TEAM DEF BAL 20.5% SD 19.6% JAC 15.0% DET 12.4% DAL 10.6% NO 8.6% BUF 6.2% SF 4.7% CIN 3.6% ATL 3.1% PIT 2.7% MIN 2.5% CHI 2.1% KC 1.9% CAR 1.8% OAK 1.7% STL 1.3% PHI -0.2% SEA -1.0% NYJ -1.4% ARI -2.0% WAS -3.6% DEN -4.0% MIA -4.6% NE -5.4% GB -6.0% TB -6.6% IND -7.6% NYG -7.7% CLE -8.3% TEN -15.0% HOU -30.8% TEAM POWER SD 32.7% CHI 24.1% IND 22.1% ATL 21.7% JAC 18.9% SF 13.4% PHI 12.7% BAL 12.1% DAL 10.8% NO 10.5% CIN 7.4% SEA 5.1% NYJ 2.2% MIN 2.1% NE -2.2% NYG -3.4% PIT -3.5% STL -5.4% ARI -5.9% DEN -6.4% KC -7.0% BUF -8.6% DET -9.5% GB -10.1% WAS -10.6% MIA -12.0% TB -15.7% CAR -16.6% CLE -17.6% OAK -17.6% TEN -26.7% HOU -30.1% 781267[/snapback] Question without notice Mike. I appreciate the yardage situation. What happens to the yardage from a field goal- is that counted as making the 100% of yardage required- just wasn't sure. With a TD that's obvious. Thanks.
Mikie2times Posted September 21, 2006 Author Posted September 21, 2006 Some Questions, I will do my best to answer them. Mile High- This is a difficult concept to convey on a message board. I spent a great deal of time explaining the SOS, but the role it plays in the final number is only 25%. This is actually less then most statistic based rating systems. The disparity you see with some teams in the rankings does not mean they haven't performed well, all it means is the NFL isn't deep enough in it's schedule for a more accurate SOS to be accessed. Without great raw numbers you won't be making it very high in these rankings anyway. The overall record of teams that have played those ranked lower then them is 59-5 after 2 weeks, So like I said, despite the lack of a true SOS early on these rankings are still representing what has happened in the NFL very accurately. If you want a more detailed explanation of how I apply SOS read the next paragraph. Lets say Buffalo scores a a 10% offensive rating, and a 15% defensive rating (just an example) The SOS for offense would be made up of two parts. 50% would be the the combined opponents defense rating, with the other 50% being the offensive rating those opponents have faced. So if we played defenses who average 10% better then the rest of the league, and those teams have played offenses that average 5% better then the rest of the league, the defensive SOS we've faced would be 15%. Our raw offensive rating was 10%, so just take 25% of the SOS number and add it to the offensive rating and that's the final offensive rating. The same is done of the defense, and the combined number ends up being the power rating. Shamrock- This doesn't factor in points in anyway. It might be a little over the top for some but I don't feel points scored and allowed is the best statistic to use for future predictions. Points get effected by way to many outside factors that may or may not happen again down the road. Yards for and against, and Yards Per Attempt, are much better future predictors, but they also aren't without shortcomings. This method is meant to take away those shortcomings. All you need to know for the raw numbers is how far a team could drive the ball, and how far they did drive the ball. If a team started at it's own 20 they could drive 80 yards. If they drive 75 yards and it results in no points they still get credit for a 75 out of 80 possible yards.
CircleTheWagons Posted September 21, 2006 Posted September 21, 2006 Shamrock- This doesn't factor in points in anyway. It might be a little over the top for some but I don't feel points scored and allowed is the best statistic to use for future predictions. 781572[/snapback] Count me as one who thinks it's over the top What would you think about giving half of the points for the remaining yards in a successful field goal (if the drive stalled at the 20 but the team came away with a field goal they get 10 additional points). Personally, I would try to add additional points for different parts of the field - driving from your own 20 to your 40 seems different than driving through the red zone for a TD. Anyway thanks for taking the time to share your stuff - it's the best part of the Wall for me.
Mikie2times Posted September 21, 2006 Author Posted September 21, 2006 Count me as one who thinks it's over the top What would you think about giving half of the points for the remaining yards in a successful field goal (if the drive stalled at the 20 but the team came away with a field goal they get 10 additional points). Personally, I would try to add additional points for different parts of the field - driving from your own 20 to your 40 seems different than driving through the red zone for a TD. Anyway thanks for taking the time to share your stuff - it's the best part of the Wall for me. 781668[/snapback] The half yardage credit for a successful FG isn't a bad idea. As for your other comments at the core of this system is the fact that it credits offenses for having to drive farther with the ball, and weakens defenses for having a smaller field to defend. The cumulative totals on the year are the numbers I use, not the average % of each individual drive. So if the Bills averaged starting on there own 35 this year, and had 19 total drives the yard potential on offense would be 100-35= 65, multiplied by the amount of drives (19) for a total of 1235 potential yards. Whatever they actually gained is divided by that number for the raw offensive ranking. I then apply the SOS methods I've explained for the final numbers.
CircleTheWagons Posted September 21, 2006 Posted September 21, 2006 it credits offenses for having to drive farther with the ball, and weakens defenses for having a smaller field to defend.781685[/snapback] Are you sure that's what your system does? It seems to me that it punishes offenses for having to drive further (3 and out at own 20 is way worse for your raw offensive ranking than 3 and out at the opponents 15).
.thumb.png.1b7fb4086290c3afc001a7900b31fab0.png)
Recommended Posts