Wins Tree -- something to play with

granger

Six-man expert
Founder
Here's a simple link to some data I have available...

http://sixmanfootball.com/wins_tree.php

you can try to list a school and click and see their wins, as well as indirect wins. Level 1 wins are direct. Level 2 wins are when you have defeated someone and they defeat someone.... on and on... obviously could go on and on, but I take the best win. The JV teams included have played a varsity competition.

I hope to make this more advanced in the future
 
Actually it has nothing to do with my ranking system. It is just something interesting... but you can see which teams have some solid influence
 
I think it is interesting to compare a team like Throckmorton to a team like Ira to a team like Richland Springs... the sphere of influence is amazing. Throckmorton's win over Dallas The Covenant and RS win over Austin Hill Country really help because those teams are so successful and outside the norm of who everybody is playing.

I am also not completely honest when I say it has nothing to do with the rankings. It does in a way. The computer is always re-evaluating each win/loss throughout the season, so if you are beating teams that are winning... it helps.... and vice versa
 
I am not following exactly how this works.

I thought I would try it out so I just entered a school starting with an "A" that I have read a lot about on this site, Aquilla.
There were 4 levels under them and our team, Tyler HEAT, was in the 4th level. I have no gripe about being on the 4th level, we may be even lower than that under some teams. My question is that there are no teams on level 1, 2, or 3 that we have even played. In level 4, the same level we are in, there are 4 of the 5 teams we have played. One beat us by 45, one we beat by 45, and 2 we beat by lesser margins. The other team that beat us by 45 is not shown anywhere in the list.

I am not really that concerned about our particular showing, it just seemed to me that if I am understanding this right, there should have been a team on level 3 that beat us and one on level 2 that beat them, and level 1 that beat them, and that team was beat by Aquilla. What am I not getting?
 
Is this wins tree the reason that the power rating for a given week has rarely been the prior week's power rating, plus or minus the change?

I've noticed over the course of this season that The current week's change may be 3 or 4 points, but the power rating will have only increased 1 or 2 points from the previous week's power rating.
 
Heb 12:29":3784fios said:
I am not following exactly how this works.

I thought I would try it out so I just entered a school starting with an "A" that I have read a lot about on this site, Aquilla.
There were 4 levels under them and our team, Tyler HEAT, was in the 4th level. I have no gripe about being on the 4th level, we may be even lower than that under some teams. My question is that there are no teams on level 1, 2, or 3 that we have even played. In level 4, the same level we are in, there are 4 of the 5 teams we have played. One beat us by 45, one we beat by 45, and 2 we beat by lesser margins. The other team that beat us by 45 is not shown anywhere in the list.

I am not really that concerned about our particular showing, it just seemed to me that if I am understanding this right, there should have been a team on level 3 that beat us and one on level 2 that beat them, and level 1 that beat them, and that team was beat by Aquilla. What am I not getting?
I'm getting a headache............think I'll just wait for December.
 
I posted an explanation on the page now that is a little better but here's a good in-depth way to look at this.

Let's start with Tyler HEAT.
They are 3-2 and as we can see, they have def. Bloomburg, Corsicana Agape and Fruitvale JV. Those are their Level 1 wins (aka Direct Wins). None of those teams has won a game, so Tyler HEAT's 'branches' stop there.

OK, so back to Aquilla and how we get to Tyler HEAT on the fourth level.

Aquilla has defeated Milford, who def. Walnut Springs, who def. Rising Star, who def. Brookesmith, who def. Weatherford Home School, who def. Longview Trinity, who def. Tyler HEAT. That would make it a level 7 win.... I think somewhere along the lines the numbers are wrong. I am fairly confident the teams are correct, just their appropriate levels....

I'll look into this and say more when I find out.

As for RainJack, I am not sure what you mean. There are ratings and rankings. A rating is a number a ranking is ordinal (the order). I know you know that, but I am confused by the way the questions was posed.
 
Interesting, pablum for the proponents that if their team beat a team that is ranked higher then their team, then their team HAS to be ranked higher.

Interesting, but I'm more interested in the method you use to walk the tree, recursive call or a aggregation of down stream victories? How hard would it be to do the inverse, ie, losses :)
 
fajitapete":z0hopfdz said:
Interesting, pablum for the proponents that if their team beat a team that is ranked higher then their team, then their team HAS to be ranked higher.

Interesting, but I'm more interested in the method you use to walk the tree, recursive call or a aggregation of down stream victories? How hard would it be to do the inverse, ie, losses :)

No one is making that argument, Pete. No one.
 
granger":rq1g63tu said:
As for RainJack, I am not sure what you mean. There are ratings and rankings. A rating is a number a ranking is ordinal (the order). I know you know that, but I am confused by the way the questions was posed.

I'm not that concerned with rankings. But, you rank according to the ratings, right? A team with a rating of 200 will be ranked higher than a team with a rating of 175, no? But on to my question.

Let's use Valley as an example. In the week 5 rankings, Valley had a rating of 286.32. In week 6 rankings, Valley now has a power rating of 294.80 with a change of 3.39.

How does Valley get to a power rating of 294.80 and only have a change of 3.39? By my math, 294.80 minus 3.39 is 291.41. The change from week 5 to week 6 (if looking at the actual change in ratings week-to-week) is 8.48.

Do you see what I'm getting at? I'm wondering if the win tree affects prior week ratings.
 
Granger on a date:
dating_statisticians.png
 
Can you guys not hijack this thread? You have 10,000 other threads you can go and hijack. Pick one of them and show Granger a modicum of respect.
 
rainjacktx":3p5o91no said:
Can you guys not hijack this thread? You have 10,000 other threads you can go and hijack. Pick one of them and show Granger a modicum of respect.
We aren't trying to hijack this thread. It's actually interesting.
 
Back
Top