Monday, April 17, 2017

April 15th, 2017

Last week's top five teams were Dallas, Toronto, Seattle, Portland, and Atlanta. This week, all five of those teams dropped points. Kansas City cracked the top five - jumping from 6th to 2nd - after picking up 20 rating points from Portland on the road. Other big results this week include Chicago over New England (23 points), New York City over Philadelphia (22 points), and Salt Lake over Colorado (20 points).

Toronto's loss to Columbus means Dallas and Kansas City are the last two unbeaten teams in the league. Fittingly, Dallas hosts Kansas City this weekend. On the other end of the table, Philadelphia are now at an all-league all-season low of 1396 and haven't had a positive result since opening weekend (a draw on the road to Vancouver). At this point, they are predicted to finish the season on a dismal 27.97 points.

In Monte Carlo simulations, Kansas City's Supporters Shield chances dramatically increase, going from 5% to 18.4% in the span of a week. Dallas remains largely unchanged at around a 40% chance for the title.

Monday, April 10, 2017

April 9th, 2017

The biggest win of the week goes to Real Salt Lake for their 3-0 victory over Vancouver at home. The win nets RSL a 22 point boost in the Elo Ratings, moving them from dead last to 19th in one performance. The basement now belongs to Philadelphia, who suffered a 1-3 defeat at home to Portland which saw them slide 21 points.

With 10 points from 4 games, Dallas sits atop the ratings and are currently favored by the Monte Carlo simulation to win the Supporters Shield. Despite good seasonal performances by Portland and Orlando, shades of last season prevent them from topping the Elo Ratings as of yet. At 13 points from 6 games and +56 rating points on the season, Portland currently sit at 4th overall in the rankings. At 9 points from 4 games, Orlando has only netted a total of 17 rating points this season, putting them in a middling 10th place overall. This is because all three of their wins have been by a one goal margin, and their loss to Columbus negated about one and a half of those wins. Speaking of seasonal movement, the expansion clubs have both moved quite a bit so far. Atlanta is hot on Portland's heel at +53 for the season, while Minnesota has unquestionably had the worst season so far with -68 rating points total. Here is a summary of the biggest movers on the season so far:

(+56) Portland Timbers
(+53) Atlanta United FC
(+36) FC Dallas
(+35) Sporting Kansas City

(-68) Minnesota United FC
(-39) Vancouver Whitecaps FC
(-38) Philadelphia Union
(-34) New York Red Bulls

Monday, April 3, 2017

April 2, 2017

Last week was a short week (3 games) so I've included the results in Week 5.

Minnesota had down and up results, with a 2-5 loss at New England and 4-2 win to Real Salt Lake - their first ever win. The two results nearly canceled each other out, which means they didn't end up moving much this time around. Columbus ended up being the biggest positive mover, after beating both Portland and Orlando at home. The biggest negative mover was the Red Bulls, who drew Real Salt Lake at home then lost to Houston on the road. Portland and Orlando losing means no team is still perfect, though Kansas City, Dallas, and Toronto remain undefeated. Through only five games, Minnesota has let up a mind-blowing 20 goals (on pace for 136 through the season).

Dallas and Toronto are still the top teams in the league, while Seattle has slipped a bit and is closing in on getting swallowed by the rest of the pack. Here are the full ratings:

Monday, March 20, 2017

March 19, 2017

Biggest mover this week was the Columbus Crew with their 2-0 road win over DC United. Toronto and Atlanta also had good weeks. Toronto finishes on top of the Elo ratings, but Toronto, Dallas, and Seattle are close enough in points that I expect the leader to change quite a bit in the coming weeks. Likewise, the bottom five (Chicago, Vancouver, Minnesota, DC United, Salt Lake) are close enough that I'd expect them to flip roles weekly. Salt Lake gets the jeer this week.

Next week is an international weekend and MLS only has 3 games scheduled, so I may or may not update, depending on how interesting those results are.

Monday, March 13, 2017

March 12, 2017

A couple of interesting results from this weekend:

Los Angeles lost their second straight game to start the season, falling to a rating of 1496. This is the first time Los Angeles has been under 1500 since July 8th, 2012, and the worst rating they've had since June 20th, 2012.

Minnesota's massive 42 point drop by way of their 1 to 6 home defeat to Atlanta was the 19th largest single-game rating swing of all time - out of 5030 games played in MLS's history. We don't have to go back very far in time to see the last one to surpass it, though - remember that time Toronto beat NYCFC 5-0 on the road in the playoffs last November?

Minnesota is now ranked dead last on Elo Rating, End of Season Points Predicted, and Monte Carlo simulation. In addition, the change made a huge impact on the strength of West vs. East. The West's average of 1500.90 is the lowest it has been since April 20th, 2013.

Atlanta's outlook received quite a bit of boost due that game as well. They're now predicted to finish 3rd in the East and the Monte Carlo simulator has them at over a 80% chance to make the playoffs.

Three teams remain perfect through week 2: Portland, Houston, and San Jose from the West, and New York Red Bulls from the East.

Monday, March 6, 2017

March 5, 2017

The first week of the 2017 season is in the books, and the biggest winners are the New York Red Bulls, Dallas, Portland, and Houston. All four teams saw more than a 3.5 point increase in their predicted finish to the season. Dallas launches to the top of the Elo Ratings and Seattle slides down to 4th.

In the Monte Carlo simulator, the Red Bulls and Dallas are front-runners for the Supporters Shield and Seattle's chances took a massive hit.

Monday, February 20, 2017

2017 MLS Season Prediction

Last season I stopped updating this site, though I maintained my rater and made periodic updates to Twitter (@AlanSchrader). This year I'm going to go for a compromise: less frequent updates on a more consistent basis. Maybe every other week or every month or something like that. We'll see what happens as the year plays out.

I also, admittedly, have a tendency to want to post less when Sporting Kansas City is doing poorly. If they are running the table, I'll probably be updating every week.

Anywho, before we get into the numbers I need to talk about a change to the rater I made this offseason. I've been extremely reticent to force Elo ratings to regress to the mean in the offseason. Most other sports Elo raters include this manual adjustment at the end of each season to allow for time off, personnel changes, etc. I've never included it because Elo's system is self-correcting and I've made the case that I like to watch those changes over the course of the season. For the past several seasons, I have noticed that the end-of-season ratings, when used in a Monte Carlo simulation of the following season before a single game has been played, give some extreme results. For instance, last year, Chicago ended the season with a rating of 1384, which after Monte Carlo simulation resulted in a probability of winning the Supporters Shield of 0% (and a chance of making the playoffs at just over 6%). On the other end, Seattle ended the season with a rating of 1629, which after Monte Carlo simulation resulted in a 36% chance of winning the Supporters Shield. While I do believe Seattle have a significantly higher probability of winning than Chicago do, I believe that spread is too extreme on both ends.

After playing around with several different factors in the rating formulas, I decided that the end-of-season regression to the mean was, indeed, the best way to fix this issue. I've retroactively added this to all years and now my rater acts as if I'd been doing this all along. One particular consequence to this revisionist history is every single Elo exhibit I've ever posted to this blog is now wrong. :D

So after redoing past calculations, implementing the end-of-season correction to year-end 2016 ratings, and loading in this season's schedule, we have the following exhibit with which to kick the season off:

I have to admit these results seem much more reasonable than their predecessors. One thing I wanted to point out: Kansas City, Colorado, Portland, and Minnesota all have nearly identical ratings, and their predicted finish is almost entirely based on strength of schedule. That's a fairly good case study on just how much home/away matchups against certain teams play into these predictions, even when all four of those teams are playing in the same conference and have similar schedules.

Now on to the Monte Carlo simulation:

These numbers are a lot more reasonable than the previous run (see the deltas for how much these numbers changed). While 5-1 odds for Seattle is probably too high and 667-1 odds for Chicago is probably too low, these are at least in the ballpark now.

Those are all of the words I have for now. If you made it this far, thanks for reading.