liesse00 yes, you were great on the pre-season ladder. And, great to see you in this phase of the contest. It must be going to be a close game today between the Thorns and the Reign. We have 6 picking the Thorns and 5 the Reign.
The moment of the season is so important that I still wish @SiberianThunderT, the only missing predictor at the moment, will join too before the start of the game. I know he's not really a regular this season: just a wish! He was the one implementing these predictions in previous seasons, so we'd sure want to have him here as our special guest for the night!
Standings after the first playoff game. Balance UT and CoachJon scored! And the majority (by 6 to 5) was (were/was?) correct.
The preceding post showed the table for our playoff predictions part of the contest based on the actual playoff brackets, after the first semifinal game. For the pre-season ladder part of the contest, in addition to the ladder predictions there also were playoff predictions. Here's the table for that part of the contest. Predictors got 2 points for correctly predicting an advancing team and an additional point for correctly predicting the teams method of advancing, i.e., Regulation time, Overtime, or Kicks from the Mark. As you can see, Balance UT and the robot correctly predicted the Thorns would advance in Regulation time; and BlueCrimson and liesse00 correctly predicted the Thorns would advance but had it in Overtime. None of the others who predicted had the Thorns advancing. From here on, the tables are going to be including every one who participated in any part of the contest. If you see a 0 score for a participant for a part of the contest, that may mean the participant made a prediction but ended with a 0 score or it may mean the participant didn't make a prediction for that part of the contest. This is the easiest way for me to do a total contest table, which you'll see in the next post.
And, here's a table for the total contest standings to date. You'll see that for the pre-season predictions part of the contest, I've broken out the ladder predictions and the playoffs predictions and then included the total of the two. The Regular Season Games Score and Pre Season Ladder Score columns, and associated standings, are final. I'll update the other columns as we through the remaining playoff games.
So right now we're at 6 say Courage win and 4 say Red Stars with Coach Jon -- and the Siberian -- yet to say. It looks like another close game, according to those "in the know."
Still almost 2 hrs to kickoff...... NC 2 CHI 1 regular time. Actually I wouldn't mind Sam Kerr winning this game and the next one
Still almost 2 hours untill kickoff... NC 2 CHI 1 - actually I wouldn't mind if Sam Kerr won this one and the next one.
OK, I don't like this, but this is why the robot does better than humans: Portland 1 v North Carolina 2 Don't forget, we all need to predict the championship game MVP. I need to figure out a stats way to do that, to be true to the robot.
And, updated tables coming tomorrow. For Blissett regarding actual playoff predictions: . Condolences. But, not so bad on pre-season playoff predictions. I was at the game tonight, and it was very weird being there and not really having a team to root for.
I know, I know... Makes me doubt that I could get the same score if I predict of if i don't predict the final match! I'll probably throw a prediction anyway, but not right now: I'll try to think it off better this time around.
I'll make a wild prediction. NC 3, Port 0. A blowout. MVP Jess McDonald Portland opened last year's championship game rough. Heath and Raso should have been yellow-carded early in the game. The ref lost control. The ref this year is going to be stricter. North Carolina has the psychological edge. Speed kills. NC has it, and Portland's speediest player, Raso, is injured. Franch might save Portland.
I'm puzzling over a stats way to predict the MVP. I have a set of stats, courtesy of my friend Chris Henderson, so it's not a question of having a computer-driven prediction of who the actual MVP will be. From the stats, it's Horan for the Thorns or Dunn for the Courage. And, between those two, it's Horan. The problem is, the stats don't distinguish based on which team wins the game, yet it's almost certain that the pick will come from the winning team. And, the prediction contest doesn't ask who actually will be the most valuable player in the game but who those that pick will pick as the MVP. Those two things are not necessarily the same. The robot has picked the Courage to win the game. The stats system says that Horan has been the best performing player over the season. So do I pick Dunn because she's a Courage player and I think she'll get the pick if the robot is right and the Courage wins? But that would be having the robot's pick take into account a human prejudice that the MVP must be a player on the winning team. A little research while I'm writing this gives me the answer. Stats say that the MVP will come from the winning team. In fact, in every NWSL championship game so far, the MVP has come from the winning team. So, from the robot's statistics perspective, the Courage will be the winning team and the MVP therefore will be from the Courage. MVP: Dunn There, I've done my duty to the robot, even if I don't like what the robot says. Which perhaps says something about why the robot is doing well in the contest -- strict objectivity vs a mix of objectivity and subjectivity.
Latest tables, through the semi-finals. One game to go: Pre-Season Playoff Predictions: Actual Playoff Predictions: Entire Contest:
does the robot pump out any stats about the liklihood of the same team winning every time when there are 4 contests between the same 2 combatants in the same season? i don't see how the courage can beat a good team like the thorns for a fourth time this season. i especially don't see how the courage can beat the thorns for a second time this year in that house that is their fortified home.... ...but when I watch the courage play, I think to myself, "they're going to win!"
Nope, plus the data base would be way to small to use it from a stats perspective. But in the meantime, just for fun, I have some stats tables for all of this season's games to date: Yes, in the end, home field advantage did assert itself. It's not big, but is right about where I would expect it to be. Predicting a team getting more than 2 goals in a game is pretty risky.
Speaking of too small data-bases, current NWSL stats say that it's statistically impossible for a team that won the Shield to also win the Championship. The incidence of this kind of event is currently 0%. Of course "0%" is a figure of speech, since we're way far from having even just the one hundred games that could somehow justify the use of the "percent" word in its actual meaning! So, we can say that from the point of view of this particular piece of stats, NC Courage should surely lose, no matter the opponent, because they won the Shield. But, of course, the sample is irrelevant and, even from just a statistical point of view, the "curse" is fighting with other more meaningful pieces of statistics from this season, so I'd argue that Courage aren't necessarily doomed.