USL League One Ratings: Week 17

So the action this past week was fairly exciting, even if some of the better matches this past week were just really, really interesting ways to get to a draw. That doesn’t do much for us in the ratings department, as you’ll see below:

Rank	Team	Rating
1	Union Omaha (7-5-1)	2.104
2	Chattanooga Red Wolves (6-6-1)	1.533
3	Greenville Triumph (6-5-3)	1.082
4	Forward Madison (4-6-3)	0.824
5	FC Tucson (4-4-5)	0.335
6	Richmond Kickers (4-5-5)	0.254
7	Toronto FC II (5-4-6)	0.253
8	North Texas SC (4-3-5)	0.079
9	Fort Lauderdale CF (5-4-7)	0.058
10	New England Revolution II (6-1-7)	-0.253
11	South Georgia Tormenta (6-2-9)	-0.503
12	North Carolina FC (2-3-7)	-0.777

So I’m not including the weekly ratings change table this week because, amazingly, we saw zero teams change positions as far as ranking. Fort Lauderdale also crossed the zero threshold, giving us nine out of twelve clubs with a positive rating. It’s becoming obvious that so long as a team has more combined wins and draws than losses, they’re more likely to have a positive rating.

This makes some sense, considering that all draws result in a positive match rating. This is something I might take into account when I’m evaluating this rating for next year, but we are way, way too far away from lighting that candle. The reason why I may want to to adjust this in the future is because, of course, not all draws should be considered equal. In my opinion, I don’t think any team had a draw that should be considered a “negative” result this past week (maybe Madison), but I think it’s entirely logical to think that Greenville’s draw at home to North Carolina earlier this season was considered a negative result for them.

So, another way to interpret these ratings are with my before mentioned proportional analysis. By rating everybody on a scale that constantly fluctuates depending on what the highest and lowest ratings are, we can reorganize these ratings so that they’re displayed as if they’re calculated with some sort of “zero-sum” method.

In the predictability analysis I’m privately tracking (which has it’s positives and negatives at the moment), I only care about what a team’s rating is in proportion to the best team. In order to calculate this, I add the absolute value of the lowest rating to all twelve teams. This makes the lowest rating zero, and the highest rating can now be used as the denominator to calculate all percentages. Think of it as considering the highest rated team as the absolute best team (100%), the lowest rated team as the absolute worst team (0%), and every other team is graded somewhere in the middle. If you’d rather have a numeric scale similar to the ratings themselves, where 0 is the midpoint, you’d simply take the extra step of subtracting each adjusted rating by half of the highest rating. This can be read as:

(R + ab(L)) – ((H + ab(L)) / 2)

In the above formula, R is the team’s rating, ab(L) is the absolute value of the lowest rated team, and H is the highest rated team. Here is Greenville’s adjusted rating using this method (current rating: 1.082):

(1.082 + ab(-.777)) – ((2.104 + ab(-.777)) / 2) = 0.4185

Here’s the entire table using this calculation:

Team	Adjusted Rating
Omaha	1.4405
Chattanooga	0.8695
Greenville	0.4185
Madison	0.1605
Tucson	-0.3285
Richmond	-0.4095
Toronto	-0.4105
North Texas	-0.5845
Fort Lauderdale	-0.6055
New England	-0.9165
South Georgia	-1.1665
North Carolina	-1.4405

This might be unnecessary, but this observation makes it a little easier to spot the noticeable gaps at the top and bottom of the table. It also might support rethinking at what point do we start seeing teams as being mediocre. Madison’s rating is the closest rating to zero here, and being a team that has only one more win than they have losses, a whole bunch of draws, and a goal difference of -1, this view might actually be a more accurate representation of how they stand.

We have a few matchups this week that might give us some hints as to which teams are ready for the midseason grind. We may have the answer to that for a couple of teams, but the jury is still out for many. We’ll see how these ratings adjust from that next week.