What if the NFL used Harbin Point System instead of only W-L?

[TNGTony]

Whenever I listen to people talking about NFL football, the conversation invariably turns to how tough one team's schedule is versus other teams in different conferences and divisions. I haven't been motivated enough to play with the numbers, to compute the standings, but I wonder if it wouldn't be more interesting to the fans if playoff berths weren't determined strictly by W-L records, but by you W-L AND how difficult your schedule really was, the Harbin Point System. Of course there would be no L3 points or divisors since everyone is at the same level.

For those not familiar with the Harbin system, it was devised by Coach Harbin here in Ohio to figure out why some undefeated High School teams would not make it to the state championships. It is an ingenious calculation that nearly always predicts the better team in a match, and is great to use for seeding a play-off bracket.

Basically you get a certain amount of points for a win. You get a certain amount of points for a win by a team you beat in the same season. Divide the points by the number of games and you have your team average score. You get no points for a loss and no points if a team you beat loses to other opponents. After a few weeks you get to see a separation among teams that may have identical W-L records.

What would this mean for the league? Teams that really have a rough schedule still have a shot at the playoffs even with crappier records. It would keep more people interested in the playoff hunt all the way to the very last games. It is highly unlikely a team with a great record would get screwed because of the limited league, meaning that L2 points would be guaranteed at some point. But if your whole division sucks, your 6-0 division record won't be worth as much as a win against a division that is otherwise undefeated!

Just a fun exercise. I know it will never happen, but it's fun to think about who would edge who out if difficulty of schedule were a factor.

 

[Jimbo]

Whenever I listen to people talking about NFL football, the conversation invariably turns to how tough one team's schedule is versus other teams in different conferences and divisions. I haven't been motivated enough to play with the numbers, to compute the standings, but I wonder if it wouldn't be more interesting to the fans if playoff berths weren't determined strictly by W-L records, but by you W-L AND how difficult your schedule really was, the Harbin Point System. Of course there would be no L3 points or divisors since everyone is at the same level.

For those not familiar with the Harbin system, it was devised by Coach Harbin here in Ohio to figure out why some undefeated High School teams would not make it to the state championships. It is an ingenious calculation that nearly always predicts the better team in a match, and is great to use for seeding a play-off bracket.

Basically you get a certain amount of points for a win. You get a certain amount of points for a win by a team you beat in the same season. Divide the points by the number of games and you have your team average score. You get no points for a loss and no points if a team you beat loses to other opponents. After a few weeks you get to see a separation among teams that may have identical W-L records.

What would this mean for the league? Teams that really have a rough schedule still have a shot at the playoffs even with crappier records. It would keep more people interested in the playoff hunt all the way to the very last games. It is highly unlikely a team with a great record would get screwed because of the limited league, meaning that L2 points would be guaranteed at some point. But if your whole division sucks, your 6-0 division record won't be worth as much as a win against a division that is otherwise undefeated!

Just a fun exercise. I know it will never happen, but it's fun to think about who would edge who out if difficulty of schedule were a factor.

Is this the format we currently use for our High School playoffs ?
It would be interesting.

Unfortunately you couldn't use it for College as not enough teams play one another and have like opponents, even in thier own Conference, a lot of teams don't play another for several years.

 

[TNGTony]

Yes, it's the OHSAA system for the Ohio tourney seeding. Actually there is no reason why the Harbin can't be used for College. The system works great for the Ohio tourney where you have over 100 schools per division and there is absolutely no way all can play everyone else. I have been for using a Harbin point system to select 32 teams for a 5 game single elimination tourney per division for college since I finally understood how the Harbin system actually works! It ingenious! It would be great to get rid of the horrid and subjective BCS. Divide the country into 4 regions. 8 teams per region, or two regions, 16 teams per region, or a free-for-all, one tourney with the top 32 Harbin scores in a tourney seeded 1/32 - 2/31 - 3/30, etc.

Check out Joeeitel.com for the current (unofficial) HS rankings in Ohio.

 

[Jimbo]

Yes, it's the OHSAA system for the Ohio tourney seeding. Actually there is no reason why the Harbin can't be used for College. The system works great for the Ohio tourney where you have over 100 schools per division and there is absolutely no way all can play everyone else. I have been for using a Harbin point system to select 32 teams for a 5 game single elimination tourney per division since I finally understood how the Harbin system actually works! It ingenious! It would be great to get rid of the horrid and subjective BCS.

Check out Joeeitel.com for the current (unofficial) HS rankings in Ohio.

Unfortunately my HS is having a down year after making the playoffs the last 3 or 4 years in a row. (Toledo Whitmer).
They have a very young team this year and a pretty new coach, his 2nd year I think, I don't like him nearly as well as the previous coaches they've had.

Thanks for the link.

Problem with that is website is, if you don't know what division your team is in, your kinda lost ...
I don't see any Search function to find a particular team.

Also I have used jjhuddle as well.

I found them, they are in Division 1 Region 1 and currently ranked 18 with a sad 4-4 record this year so far.

 

[TNGTony]

Your team is still in the hunt! This year D1 was changed to include only the largest 10% of Ohio schools and there are only two regions (basically north of I 70 and south of I 70). The rest of the schools are spread out over the existing 6 divisions in 4 regions each. This means that, unlike Div 2-7, the top 16 out of 30-some-odd in your region make it. I really dislike this because I think it dilutes the tourney, but it gives teams like yours (and one of my favorites that is having a rough year) a chance at the diluted Div 1 trophy. On the bright side, it made the D2 hunt in my region (Region 6) a LOT more interesting because two of the schools that were D1 are now at the top of D2 and one of the smaller powerhouses in D2 got bumped out to D3 where they are dominating.

Anyway, the reason I kind of brought this up was because of a little friendly argument I had with a Patriot fan. The Bengals and the Patriots have identical records. Using a Harbin system you could see objectively how hard or easy each team's schedule was.

 

[TNGTony]

So okay, just for fun, I figured the scores for the Bengals and Patriots.

	Win	Loss	L1	L1 Average	L2	L2 Average	Average
Patriots	5	2	25	3.571428571	70	1	4.571428571
Bengals	5	2	25	3.571428571	90	1.285714286	4.857142857

Explanation.
L1 = Wins x 5 points.
L1 Average = L1 points divided by number of games played.
L2 = wins by defeated opponents X 5 points (Had the Pats beat the Jets twice, the Jets victories would have counted twice in the L2 bracket for the Pats)
L2 Average = L2 points divided by 10 then divided by number of games played by the team.
Average = L1+ L2 points. Actual team ranking.

So at this point in the season, the Bengals have a slight edge in difficulty of schedule.

What killed the Patriot's L2 score was the Bucks!

 

[TNGTony]

So... I got bored:

"--------	Win	Loss	L1	L1 Average	L2	L2 Average	Average	Weighted
Chiefs--	7	0	35	5	70	1	6
Seahawks	6	1	30	4.285714286	80	1.142857143	5.428571429
Saints	5	1	25	4.166666667	70	1.166666667	5.333333333	4.571428571
Broncos	6	1	30	4.285714286	60	0.857142857	5.142857143
Colts---	5	2	25	3.571428571	95	1.357142857	4.928571429
Bengals	5	2	25	3.571428571	90	1.285714286	4.857142857
49ers---	5	2	25	3.571428571	75	1.071428571	4.642857143
Patriots	5	2	25	3.571428571	70	1	4.571428571
Packers	4	2	20	3.333333333	60	1	4.333333333	3.714285714
Chargers	4	3	20	2.857142857	60	0.857142857	3.714285714
Jets----	4	3	20	2.857142857	50	0.714285714	3.571428571
Lions---	4	3	20	2.857142857	50	0.714285714	3.571428571
Cowboys	4	3	20	2.857142857	45	0.642857143	3.5
Bears---	4	3	20	2.857142857	40	0.571428571	3.428571429
Dolphins	3	3	15	2.5	50	0.833333333	3.333333333	2.857142857
Titans	3	4	15	2.142857143	50	0.714285714	2.857142857
Panthers	3	3	15	2.5	20	0.333333333	2.833333333	2.428571429
Ravens	3	4	15	2.142857143	45	0.642857143	2.785714286
Browns	3	4	15	2.142857143	45	0.642857143	2.785714286
Bills	3	4	15	2.142857143	45	0.642857143	2.785714286
Cardinals	3	4	15	2.142857143	35	0.5	2.642857143
Rams-----	3	4	15	2.142857143	25	0.357142857	2.5
Eagles	3	4	15	2.142857143	10	0.142857143	2.285714286
Steelers	2	4	10	1.666666667	35	0.583333333	2.25	1.928571429
Raiders	2	4	10	1.666666667	20	0.333333333	2	1.714285714
Texans	2	5	10	1.428571429	35	0.5	1.928571429
Falcons	2	4	10	1.666666667	15	0.25	1.916666667	1.642857143
Redskins	2	5	10	1.428571429	30	0.428571429	1.857142857
Vikings	1	4	5	1	10	0.2	1.2	1.028571429
Jaguars	0	7	0	0	0	0	0
Giants	0	6	0	0	0	0	0
Buccs	0	6	0	0	0	0	0

 

[Ramy]

Fun stuff to look at for sure.

 

[TNGTony]

And after last night's Giants win a few minor changes to "my" rankings:

"--------	Win	Loss	L1	L1 Average	L2	L2 Average	Average	Weighted
Chiefs--	7	0	35	5	75	1.071428571	6.071428571
Seahawks	6	1	30	4.285714286	80	1.142857143	5.428571429
Saints	5	1	25	4.166666667	70	1.166666667	5.333333333	4.571428571
Broncos	6	1	30	4.285714286	65	0.928571429	5.214285714
Colts---	5	2	25	3.571428571	95	1.357142857	4.928571429
Bengals	5	2	25	3.571428571	90	1.285714286	4.857142857
49ers---	5	2	25	3.571428571	75	1.071428571	4.642857143
Patriots	5	2	25	3.571428571	70	1	4.571428571
Packers	4	2	20	3.333333333	60	1	4.333333333	3.714285714
Chargers	4	3	20	2.857142857	60	0.857142857	3.714285714
Jets----	4	3	20	2.857142857	50	0.714285714	3.571428571
Lions---	4	3	20	2.857142857	50	0.714285714	3.571428571
Cowboys	4	3	20	2.857142857	50	0.714285714	3.571428571
Bears---	4	3	20	2.857142857	45	0.642857143	3.5
Dolphins	3	3	15	2.5	50	0.833333333	3.333333333	2.857142857
Panthers	3	3	15	2.5	25	0.416666667	2.916666667	2.5
Titans	3	4	15	2.142857143	50	0.714285714	2.857142857
Ravens	3	4	15	2.142857143	45	0.642857143	2.785714286
Browns	3	4	15	2.142857143	45	0.642857143	2.785714286
Bills	3	4	15	2.142857143	45	0.642857143	2.785714286
Cardinals	3	4	15	2.142857143	35	0.5	2.642857143
Rams-----	3	4	15	2.142857143	25	0.357142857	2.5
Eagles	3	4	15	2.142857143	15	0.214285714	2.357142857
Steelers	2	4	10	1.666666667	35	0.583333333	2.25	1.928571429
Raiders	2	4	10	1.666666667	20	0.333333333	2	1.714285714
Texans	2	5	10	1.428571429	35	0.5	1.928571429
Falcons	2	4	10	1.666666667	15	0.25	1.916666667	1.642857143
Redskins	2	5	10	1.428571429	30	0.428571429	1.857142857
Vikings	1	5	5	0.833333333	10	0.166666667	1	0.857142857
Giants	1	6	5	0.714285714	5	0.071428571	0.785714286
Jaguars	0	7	0	0	0	0	0
Buccs	0	6	0	0	0	0	0

 

[Derwin0]

Since all teams in a division play pretty much the same schedule, and each division champion gets a playoff berth. I don't see this system improving anything. Especially with there being only 2 wild card slots in each conference.

 

[TNGTony]

Even if no one else is interested, this is fun to follow. Some errors corrected from last week that gave too many L2 points:
One meaningless stat, the Bengals have by far the most L2 points so far, meaning they have beat the teams with the better records.
This assumes the Vikings don't have a hell of a come-back to win tonight.

"--------	Win	Loss	L1	L1 Average	L2	L2 Average	Average	Weighted
Chiefs--	8	0	40	5	95	1.1875	6.1875
Seahawks	6	1	30	4.285714286	90	1.285714286	5.571428571	4.875
Broncos	7	1	35	4.375	90	1.125	5.5
Bengals	6	2	30	3.75	135	1.6875	5.4375
Saints	6	1	30	4.285714286	70	1	5.285714286	4.625
Colts---	5	2	25	3.571428571	105	1.5	5.071428571	4.4375
Patriots	6	2	30	3.75	105	1.3125	5.0625
49ers---	6	2	30	3.75	80	1	4.75
Packers	5	2	25	3.571428571	65	0.928571429	4.5	3.9375
Lions---	5	3	25	3.125	75	0.9375	4.0625
Bears---	4	3	20	2.857142857	70	1	3.857142857	3.375
Chargers	4	3	20	2.857142857	55	0.785714286	3.642857143	3.1875
Cardinals	4	4	20	2.5	95	1.1875	3.6875
Panthers	4	3	20	2.857142857	40	0.571428571	3.428571429	3
Titans	3	4	15	2.142857143	70	1	3.142857143	2.75
Jets----	4	4	20	2.5	30	0.375	2.875
Dolphins	3	4	15	2.142857143	55	0.785714286	2.928571429	2.5625
Ravens	3	4	15	2.142857143	45	0.642857143	2.785714286	2.4375
Cowboys	4	4	20	2.5	20	0.25	2.75
Rams-----	3	4	15	2.142857143	45	0.642857143	2.785714286	2.4375
Raiders	3	4	15	2.142857143	15	0.214285714	2.357142857	2.0625
Bills	3	5	15	1.875	45	0.5625	2.4375
Browns	3	5	15	1.875	40	0.5	2.375
Eagles	3	5	15	1.875	10	0.125	2
Texans	2	5	10	1.428571429	45	0.642857143	2.071428571	1.8125
Steelers	2	5	10	1.428571429	35	0.5	1.928571429	1.6875
Redskins	2	5	10	1.428571429	35	0.5	1.928571429	1.6875
Falcons	2	5	10	1.428571429	15	0.214285714	1.642857143	1.4375
Giants	2	6	10	1.25	25	0.3125	1.5625
Vikings	1	6	5	0.714285714	10	0.142857143	0.857142857	0.75
Jaguars	0	8	0	0	0	0	0
Buccs	0	7	0	0	0	0	0

 

[Teehar]

That is pretty interesting Tony.It appears to still be pretty close to actual NFL standings though.I see a couple of differences,but for the most part it's very close.I wonder how it compares to some of the power rankings?

 

[TNGTony]

Really, the whole idea for me was to figure out the "better" team with identical W/L record even if they don't play each other.

I limited the calculations to 3 decimal places to make it easier to read
Here are the standings after the Seahawks shellacking of the Rams:

"--------	Win	Loss	L1	L1 Average	L2	L2 Average	Average	Weighted
Chiefs--	8	0	40	5.000	100	1.250	6.250
Seahawks	7	1	35	4.375	105	1.313	5.688
Broncos	7	1	35	4.375	90	1.125	5.500
Bengals	6	2	30	3.750	135	1.688	5.438
Saints	6	1	30	4.286	70	1.000	5.286	4.625
Patriots	6	2	30	3.750	120	1.500	5.250
Colts---	5	2	25	3.571	95	1.357	4.929	4.313
49ers---	6	2	30	3.750	80	1.000	4.750
Packers	5	2	25	3.571	60	0.857	4.429	3.875
Lions---	5	3	25	3.125	80	1.000	4.125
Bears---	4	3	20	2.857	70	1.000	3.857	3.375
Cardinals	4	4	20	2.500	90	1.125	3.625
Chargers	4	3	20	2.857	55	0.786	3.643	3.188
Panthers	4	3	20	2.857	45	0.643	3.500	3.063
Titans	3	4	15	2.143	70	1.000	3.143	2.750
Dolphins	3	4	15	2.143	60	0.857	3.000	2.625
Jets----	4	4	20	2.500	30	0.375	2.875
Ravens	3	4	15	2.143	40	0.571	2.714	2.375
Rams-----	3	4	15	2.143	40	0.571	2.714	2.375
Cowboys	4	4	20	2.500	15	0.188	2.688
Bills	3	5	15	1.875	50	0.625	2.500
Browns	3	5	15	1.875	40	0.500	2.375
Raiders	3	4	15	2.143	20	0.286	2.429	2.125
Texans	2	5	10	1.429	45	0.643	2.071	1.813
Eagles	3	5	15	1.875	5	0.063	1.938
Steelers	2	5	10	1.429	40	0.571	2.000	1.750
Redskins	2	5	10	1.429	30	0.429	1.857	1.625
Falcons	2	5	10	1.429	15	0.214	1.643	1.438
Giants	2	6	10	1.250	25	0.313	1.563
Vikings	1	6	5	0.714	15	0.214	0.929	0.813
Jaguars	0	8	0	0.000	0	0.000	0.000
Buccs	0	7	0	0.000	0	0.000	0.000

 

[Juan]

What if we taught kids effort was more about important than results....wait we do...that's why we have a bunch of kids who think 2+2=5

Sent from my DROID RAZR HD using SatelliteGuys mobile app

 

[Ramy]

Really, the whole idea for me was to figure out the "better" team with identical W/L record even if they don't play each other.

I limited the calculations to 3 decimal places to make it easier to read
Here are the standings after the Seahawks shellacking of the Rams:

Shellacking? They barely won.

 

[Derwin0]

What if we taught kids effort was more about important than results....wait we do...that's why we have a bunch of kids who think 2+2=5

Didn't we teach that back in 1984?

 

[TNGTony]

More worthless numbers that are fun to look at:
Sorted by "Weighted" since most teams have only played 8 games and some have played 9 games.

"--------	Win	Loss	L1	L1 Average	L2	L2 Average	Average	Weighted
Chiefs--	9	0	45	5.000	135	1.500	6.500	6.500
Seahawks	8	1	40	4.444	120	1.333	5.778	5.778
Patriots	7	2	35	3.889	110	1.222	5.111	5.111
Broncos	7	1	35	4.375	100	1.250	5.625	5.000
Bengals	6	3	30	3.333	135	1.500	4.833	4.833
Colts---	6	2	30	3.750	130	1.625	5.375	4.778
49ers---	6	2	30	3.750	90	1.125	4.875	4.333
Saints	6	2	30	3.750	85	1.063	4.813	4.278
Jets----	5	4	25	2.778	90	1.000	3.778	3.778
Packers	5	2	25	3.571	80	1.143	4.714	3.667
Cowboys	5	4	25	2.778	65	0.722	3.500	3.500
Lions---	5	3	25	3.125	85	1.063	4.188	3.257
Panthers	5	3	25	3.125	40	0.500	3.625	3.222
Dolphins	4	4	20	2.500	90	1.125	3.625	3.222
Chargers	4	4	20	2.500	75	0.938	3.438	3.056
Titans	4	4	20	2.500	70	0.875	3.375	3.000
Browns	4	5	20	2.222	65	0.722	2.944	2.944
Cardinals	4	4	20	2.500	60	0.750	3.250	2.889
Bears---	4	3	20	2.857	55	0.786	3.643	2.833
Eagles	4	5	20	2.222	40	0.444	2.667	2.667
Bills	3	6	15	1.667	60	0.667	2.333	2.333
Redskins	3	5	15	1.875	55	0.688	2.563	2.278
Ravens	3	5	15	1.875	50	0.625	2.500	2.222
Rams-----	3	5	15	1.875	30	0.375	2.250	2.000
Raiders	3	5	15	1.875	30	0.375	2.250	2.000
Texans	2	6	10	1.250	40	0.500	1.750	1.556
Steelers	2	6	10	1.250	40	0.500	1.750	1.556
Giants	2	6	10	1.250	25	0.313	1.563	1.389
Falcons	2	6	10	1.250	15	0.188	1.438	1.278
Vikings	1	7	5	0.625	10	0.125	0.750	0.667
Jaguars	0	8	0	0.000	0	0.000	0.000	0.000
Buccs	0	8	0	0.000	0	0.000	0.000	0.000

 

[Teehar]

Gotta disagree with the numbers this week.No way the Bengals should be ahead of the 49ers,Saints,and Colts.And the Jets?Seriously

 

[AntAltMike]

My home state, New Hampshire, has been using a weighted ranking system over 50 years. When I was in high school in the late 1960s, football programs were ranked into categories based on male enrollment, with the six largest schools being called Division 1 and the next, including my school, were in Division 2. There may have been more than 6 Division 3 schools. In the so-called "standings", a team got credit for 1 out of 1 for beating a Division 1 school, .8 out of 1 for beating a Division 2 school, .6 out of one for beating a Division 3 school. Our football team went 10-0 in 1968 and beat the best and second best Division 1 schools along the way, but if it had lost one game to anyone, then it could have finished no higher than second, in spite of having beaten 9-1 Manchester Central by a score of 50-6.

Nashua had one high school for over 4,000 students. We had 1,400. Do you realize what that means for a football team? Football teams need nine large lineman. That means that, on the average, their ninth largest lineman would tend to be as big as our third largest lineman, and our ninth largest lineman, a starter, would have been about as big as their 25th largest lineman, but of course, they didn't even have a 25th lineman.

No way will that ranking system fly in the NFL. Wins and losses are easy to conceptualize. The fans don't even want to contemplate ties. Most hockey fans can't figure out in their heads what a tie-loss is when they try to estimate their team's prospects of moving up or down in the standings. Billions of dollars are made off the gambling-supported/gambling-free NFL, and no way in hell is anyone going to tinker with that money making cow to satisfy some statistical geeks.

 

[TNGTony]

Gotta disagree with the numbers this week.No way the Bengals should be ahead of the 49ers,Saints,and Colts.And the Jets?Seriously

Yeah... I triple checked my formula for the weighting system for an unbalanced schedule (8 games v 9). It all came down to L3. Statistically the Colts and Bengals are a dead heat. 9ers and Saints have played teams that have very few victories themselves. To put things in perspective, the Chiefs have won every one of the their games. The Bengals only 6 games. However the teams these guys have beaten have the same total of victories among them! IOW, the Bengals have beaten "better" teams than the Chiefs.
Again, this is all for fun. Unlike the HS or College teams this system is supposed to rate, all NFL teams are "equal" in that they have the same pool to choose from "enrollment" is the same, divisions are equal.
One of these days I may add a "points vs victories" differential that would weight the standings to offensive output and defensive capability.

oh... and the Bengals did beat the Jets.