The opening week of the season was disappointment via surprise.
Toronto FC, a massive home favourite which I bet at 1.70 before the price plummeted below 1.50, laid an egg. They had no shortage of chances and finished with an xG of 1.40 to 1.28 of Columbus despite losing 0-2 at home.
The most substantial difference between model and market pricing of the opening week was Philadelphia, who took care of business. The 2-0 final scoreline was even still a bit misleading as the xG final scoreline read 3.50-0.60 in favour of the home side.
Portland was the long shot of the week, and after their price shortened from where I bet it at 3.19 to below 2.60, the market swung back. LA released a solid lineup and drove the price back above 3.50. The Timbers lost away 2-1 which was indicative of the xG scoreline which finished 3-1.75.
New England Revolution vs Colorado Rapids
This week, the combination of the three models points out a couple of big advantages. Here are the probabilities I have generated for each match in Week Two:
Chicago vs Kansas City
Columbus Crew vs Montreal Impact
Houston Dynamo vs Vancouver Whitecaps
Orlando City vs Minnesota
New York Red Bulls vs Portland Timbers
Atlanta United vs DC United
New York City vs LA Galaxy
The first bet of note is New England with a model price near 1.30. This price undoubtedly accounts for a lot of noise. Anytime a pricing difference is this vast, I stop and question why. The answer this week is the weighting of Poisson accounting for Colorado and their poor history playing away from home.
The market is also overreacting to what happened last week where New England was pinned back and pummeled with shots for 90 minutes.
Poisson is weighting New England higher because of their strong home numbers carrying over. I do not think 1.30 is accurate, but, I can make a case for 1.50 and see significant value in the current price of 1.91.
Chicago makes their season debut at home this week and is the second pricing standout. With odds of 2.22, the model identifies value with the projected price at 1.66.
Kansas City underachieved last week coming up empty despite generating chances that equated to 1.55 expected goals. I feel there is a little bit of noise in the number, but do not think a price of 1.80 to 1.85 is unjust in this spot.
Atlanta is my big wager of the week. The market price currently is 1.57 and dropping quickly towards the model price of 1.28.
This match is a case of extreme market overreaction. Atlanta got blasted on the road in Houston losing 4-0 – which does not look pretty by any standard. However, there is a reason for optimism considering they generated 1.74 expected goals, the biggest difference between expected goals difference of any team all week.
DC United is a different story. Despite being a man up for 55 minutes, they lost the possession battle 60/40, had more passes in their defensive end in the second half than Orlando and gave up the late equaliser. Taking Atlanta at the hefty price is well warranted this week.
New York City and Los Angeles should see plenty of opportunities and the model projects a high percentage chance of both teams scoring multiple goals.
The distribution shows both teams with more than 55% chances to score two goals or more. Both sides scored a pair in the opening week, and no team generated more expected goals than LA (3.05). Over 2.5 goals is priced currently at 1.74, well short of the model projected price of 1.50.
