2018 Season- Game Prediction Thread

Through the July 7 games:

 

Predictions for mid-week games of week 16:

North Carolina 2 v Washington 1 (could be 2-0)
Houston 1 v Orlando 2
Seattle 1 v Utah 1

For Seattle v Utah there's an interesting coincidence. Over the "full season" from last year to this, for home games Seattle's goals for average has been 1.30 and its goals against has been 1.00. For Utah for away games, its goals for average has been 1.00 and its goals against has been 1.30.

 

Prediction for mid-week games (wow, the schedule looks increasingly busy, these days: is it because ToN is approaching? ):

NC 1 WAS 0
HOU 1 ORL 1
SEA 3 UTA 1

 

NC 3 WAS 0
HOU 1 ORL 2
SEA 2 UTA 1

 

NC 3 WAS 0
HOU 1 ORL 2
SEA 2 UTA 1

I'll second lunatica's prediction.

 

It's a strategy!

 

NC 2 WAS 0
HOU 1 ORL 1
SEA 1 UTA 1

 

Here are some tables from the data mine, derived from the last 98 games (roughly 2/3 from 2018 and the last 1/3 of 2017):

 

One more table showing our predictors' scoring trends:

 

Wednesday's matches -

NC - DC, 4 - 0
Hou - Orl, 0 - 1
Sea, - Utah, 1 - 1

 

NC 3-0 WAS
HOU 2-1 ORL
SEA 1-2 UTA

 

Fascinating to me that there is no home field advantage in the NWSL? Am I reading that correctly?

 

NC 2-0 WAS
HOU 1-2 ORL
SEA 1-2 UTA

 

At least this season, yup! I pointed it out in one of the gameweek threads a month or two ago, and cpt and I have been tracking it on and off since.

 

I agree with STT, that's how it looks based on this season (plus the last third of 2017). For the 2017 season as a whole, it appeared there was a significant, but not huge, home field advantage. I'm inclined to think that our data sample may be too small. It will be interesting to see, come the end of the 2018 season, what the numbers show for the season as a whole.

Also, although it may not make a whole lot of difference, the top of the "points per game" table has played 4 more away games than home, while the bottom of the table (including Chicago) has played 4 more games home than away. That will balance out over the last third of the season.

 

Since I have a little time, one more table for this week, showing how predictors have done under the scoring system per game predicted. This takes into account that some predictors haven't made predictions for all games. (Notwithstanding the second column says there have been 54 games, there actually have been 70). This also includes the shadow predictor "1 to 1," who actually is in first place for exact score predictions.

 

NC 2-1 WAS
HOU 2-1 ORL
SEA 2-2 UTA

 

North Carolina 3, Washington 0
Houston 2, Orlando 2
Seattle 2, Utah 1

 

Wednesday games:

North Carolina 2 v Washington 0
------------ Ordega and Pugh Out & Banini Questionable; NC can even rest Questionable Zerboni and Mewis pending call ups.
Houston 1 v Orlando 2
------------- bounceback continues for ORL > improving HOU plus homefield.
------------ { and I did not check HOU homefield succes %, which may be terrible}
Seattle 1 v Utah 2 -
--------------[ at Rio Tinto 0-0; last game POR 4 UTA 0, SEA 3 HOU 1, also SEA 1 POR 0]:
--------------Press plus POR loss > Rapinoe minus Fishlock plus homefield {I didn't check that one either}
---------------maybe I should wait to see if Fishlock is playing?

I looked at @cpthomas and @BlueCrimson -- the Robot can't predict upsets and futile humans who predict upsets (UTAH) are doomed to long term failure. BlueCrimson must think the Force is strong with Pinoe.

 

REVISED CoachJon Prediction Wednesday, @cpthomas

> Broon Out and Fishlock In<

Seattle 2 v Utah 1

 

Got it, you beat the clock. I wondered if "Broon out" would result in prediction changes.

 

Whoopie, everyone who predicted beat the robot today and BlueCrimson led the charge with 6 out of a possible 9 points:

 

Great move!!!

Down with the robot overlord! This is a riot!

 

I gotta say, even though I predicted against them and I didn't earn any points, I loved the Dash win!