Ranking fantasy teams from different seasons (1 Viewer)

[Rounders]

I have started to wonder if it's possible to compare FF teams from different years so that a league can determine which teams are truly the best and most dominant. I figured by looking at points scored, point differential, wins, breakdown wins (overall), and some other stats that it can be done. Perhaps a formula to incorporate it all???

Has anyone ever looked into something like this? Do any of the commissioners out there do this? Am I off my rocker?

 

[murzman]

I kind of did this for a 12 man league. This is what I did:

1. ranked each team 1 through 12 according to total points scored over the history of the league (assign 12 pts for #1 rank, 11 for #2 rank...)

2. ranked each team according to their average season points rank. ie, if Team A finished 3rd, 6th and 12th in total points over a 3 year span, his average points rank would be (3+6+12)/3 = 7 ; assign 12 pts to highest avg rank, 11 pts to second highest and so on

3. combined the above two points assigned to create an overall ranking

The reason #2 is important is so that numbers don't get scewed if one guys does amazing in 1 season and blows everyone away and then does poor other years. This year would be a good example. Anyone who drafted Brady and say Moss might easily outscore everyone by 100-300 points+ depending on your scoring system. Add that to 2 bad season for points and he will still likely be the top ranked fantasy player if you only use #1 above. This happened in my league - the guy who was #1 in total points scored finished #1, 8 and 10 in points during the 3 years being used. There is no way he is the top fantasy performer and that is why I added in #2

Other things you could use:

- season finish (ie, planing for playoffs and success in playoffs is important). I didn't take this into account because of lack of info and laziness)

- season record - I like points scored as a much better indicator though to take out the luck, bad luck in match-ups factor

- total money won - might be good for leagues with weekly payouts etc as with money leagues, the true measure of success is money won... right?

 

[immadimwitty]

 

[Rounders]

Okay here's my attempt to do this, not sure if other's will find it worthwhile, but the results in my league were interesting.

1) I ranked all the teams by winning percentage using their "all-play" records

2) I ranked all the teams by average point differential (ie I calculated the avg pf and avg pa and took the difference)

3) I gave each team a point based on their season's finish (1 pt - 1st place, 2 pts - 2nd place, etc)

I then created a weighted formula based on these assumptions:

Assumptions

1) Points scored is a true measure of good (dominant team)

2) All-play record is a good balancing statistic that minimizes the luck of a schedule

3) Winning $$ is another important measurement and those teams that finish higher get more prize money.

"Domination" Factor Formula

(All Play Rank + (Average Pt Diff *2) + Season finish Rank) / 4

For example:

My championship team from 2001 finished with a regular season record of 9-4. My team scored 804 (53.6 per game) and "allowed" 729 (48.6 per game). My all play record that year was 85-47-1 for a winning percentage of .643.

My rankings:

All Play - 5th out of 75 teams

Point Differential - 23rd out of 75 teams

Finish - 1

Formula: (5+(23*2)+1) / 4 = 52/4 = 13.0

Once this formula is applied to all 75 teams, it turns out that my 2001 championship team is ranked 15th overall with a domination factor of 13.0

Thoughts?

 

[murzman]

I think I misinterpreted your post. I now see what you mean. You are trying to compare specific teams from year to year, not an owner's performance over many years.

For #2, I would use an average points against figure for the league, not for individual teams as points against has nothing to do with team strength. I guess one could argue having least points against makes a team more dominant as they would've won their games by a wider margin...

 

[Bobcat10]

 

[Dope]

If you used MFL, you could go by the Power Rank numbers.

If not, I think the best way to do it would be to use an "All-Played" record.

Then you would see how each team, in each year, dominated that year.

For instance:

C-2005 had an All-Play record of 129-47

S-2004 had an All-Play record of 122-54

K-2006 had an All-Play record of 122-54

and so-on.

See what I mean?

 

[MOAT]

2) I ranked all the teams by average point differential (ie I calculated the avg pf and avg pa and took the difference)Thoughts?

I'm confused on why you included this. To me it would seem this would overrate a team that got lucky and caught people on their down games or played a soft schedule. To me, PF would be a better measure since your team can't do anything about points against. If my PF for the year is 900 and PA is 850 does that mean a team that is PF750/PA600 clearly better than me? A point diff from the league average might make sense though. That way all teams are compared evenly.If someone took my teams and compared PF vs possible PF I don't think I'd fair too well. My coach stinks.

 

[Rounders]

2) I ranked all the teams by average point differential (ie I calculated the avg pf and avg pa and took the difference)Thoughts?

I'm confused on why you included this. To me it would seem this would overrate a team that got lucky and caught people on their down games or played a soft schedule. To me, PF would be a better measure since your team can't do anything about points against. If my PF for the year is 900 and PA is 850 does that mean a team that is PF750/PA600 clearly better than me? A point diff from the league average might make sense though. That way all teams are compared evenly.If someone took my teams and compared PF vs possible PF I don't think I'd fair too well. My coach stinks.

The main reason I used the differential is because our league has gone through many scoring system changes throughout the years. I thought that this would represent the dominating teams by showing how they beat their opponents. I do like the idea of the point differential against the leagues PA, that may be a better way to quantify this.Thanks for poking some holes in it, I do appreciate the "other side thinking".