The off-season has continued to throw the combined Canadian-American soccer world into confusion. With many issues, including NASL’s lawsuit, still pending, that means there may be some changes to the calculations for which I have not accounted (yet). There’s a lot of small tweaks, and some big consequences.
Currently, the magic 8-ball points to the NASL’s dissolution. Even if they do continue operation, their move to the FIFA calendar means I won’t need to worry about them until July. I can revisit how they fit in this at that time. In the meantime, I’ve left NASL, as well as NISA, USL3, and CanPL off the ratings. If/When the landscape changes, I will integrate the leagues using the formula announced in the 2017 primer.
New to the background of these ratings, the NSPL will be tracked. The league contains no Canadian teams, but as the defacto 4th division in the US Pyramid alongside the PDL, they are solidy integrated in to the US Open Cup and their play affects the ratings of PDL, USL and MLS teams on the US-side directly, and the Canadian teams indirectly as a result of league play and the Voyageur Cup.
The exact numbers are in the tabs below, however, I have chosen to mimic the weighting of the PDL for the NPSL. American lower-tier enthusiasts will have a better idea of the two leagues equivalencies, so I invite them to provide feedback on this decision in advance of Open Cup & League play.
The first big change, and visible, change will be the inclusion of the CONCACAF Champions League (CCL). With its new format and my movement to continuous rank, see below, it only makes sense. With CCL matches working as home and away matches, the same aggregate adjusting will occur. See the tabs below for the exact weighting. The starting point for non-Can/US teams will be the average of qualifying Can/US teams, which I know over-rates some, and under-rates others, but I can’t track all of CONCACAF. For 2018, that starting point is 1442.980.
This also leaves room to include the CONCACAF League, if an entry point is created there for a Canadian Premier League team at some point in the future.
The other major change is the “continuous” aspect of the ratings. Teams begin 2018 with the same ratings they ended last year. This allows for the team’s ratings to come from a larger base of data, which will improve the overall accuracy of the ratings in the long term.
That being said, one of my reasons for restarting ratings was the high turnover rate of players that many teams experience. This is especially true of lower leagues where it may literally be a new team from one year to the next. So I will also be doing a 1-year ratings, with all scores reset every league’s year. So if NASL does manage to run from August 2018 into 2019, 1-year ratings would reset again next August.
This provides a team with 2 elo-ratings: continuous and this year. I will display as a single table, sorted by this year’s position, with the continuous ranking included, as seen to the left. With Toronto FC beginning CCL play February 20th, expect elo-ranking to start moving at the end of the month!
2018 Starting Points
Since I have both 1-year and cumulative ratings, this may be a little more confusing for some people.
For the 1-year rankings, everyone is reset. Additionally, teams that are new to the leagues are set at these values for both 1-year and cumulative ratings.
If a team moved leagues, but remained in the same pyramid division (ie from NASL to USL), their cumulative rating travels with them.
I wanted to increase the L1O & PLSQ ratings, but as they are “holding” players for the CanPL, I opted to remain with the same starting points as 2017, and this will likely remain the same for 2019 (unless they destroy Ottawa Fury in the Voyageur’s Cup).
As a result, these are the starting league averages for our elo rating calculations:
You’ll also notice the cumulative league averages are no longer the starting default. There are several reasons. First, some elo-rating points were lost last year to sub-PDL teams during the US Open Cup (including that amazing run by what was basically just a pub team). Secondly, some teams, both above and below the averages, folded or transferred to a different league. And thirdly, new teams started at the original, 1-year, starting point. This has created some drop in the overall rating totals.
With the inclusion of CCL this year, I suspect we’ll see some points filter back into these totals, which should buoy the averages.
2018 Match Weights
You may notice there’s a bit of a jump from MLS to CCL. This is to leave some room for the CONCACAF League, which may need to be tracked in future years. Cup, Playoff, Conference Finals, Regional Finals, and Final Match weights may change if the format of those competitions change, but I think the listed weights will probably work well.