I know it is frequently said WCC is the strongest conference top to bottom, and I think I mostly agree with that sentiment. Going into this season I strongly felt that way, but seeing a few WCC teams that seem to have early season sentiment being too high, I am not sure now. So far the WCC has 3 wins and 7 losses against the Big West. 2 wins over Cal State Fullerton and 1 over UCSB. The Big West has wins over Pepperdine (3), Davis, LMU, and SM (2). I believe it is safe to assume CSF is the weakest in the BW so the BW wins seem to be of a higher caliber as well. There are several more matchups this weekend.
I think both conferences are great and wide open. @waterpolowednesdays did an episode last year suggesting how the NCAA tournament could be restructured to reward regular season play. Seeing how much parity we have in the conferences this year I think this would be a great year to have what they suggested. I do think the gap from MPSF to everyone else is as big as ever, but within the conferences it is as close as ever and very exciting.
I think there are 4 teams that could win the WCC and 4 that could win the Big West this year.
After today the Big West is now 9-3 against the WCC. Every WCC team has a loss against the BW except AF and every BW team has a win against the WCC except Fullerton.
Big West is definitely having a moment. They’ve won 67% of WCC match ups vs an aggregate win probability of just 51%. That said, eight of those games were very close - 2 or less goal margin of victory. Long Beach and CBU are both above 1,550 in WW’s Elo Ratings while most of the WCC hovers around the 1,500 breakeven point. The next few weeks will be exciting.
Elo is a zero sum ranking system where points are added or deducted based on a team’s margin of victory in each match up - think of the points like poker chips. If you split the chips evenly among all 30 D1 teams, they’d have 1500 points each. The difference in the points allows you to create a win probability.
You’re absolutely right. Probabilities aren’t fortune telling. Would you really want to live in a world where someone could tell you the 100% correct outcome of every game before it happened?
If you look to the far right of that table above you’ll see a column with the header “Brier”. The Brier score quantifies the accuracy each prediction. Lower is better. Anything under .250 usually means the model picked the winner correctly. Overall the model is running about 78% accurate in picking the winner of each match.
I noticed you cherry picked the outcome the least accurate prediction. I’m curious, did YOU have LMU taking down UCSD on your bingo card?
If you’re still reading my response, this is what the win probability actually means: based on the Elo differential, UCSD had 65% probability of winning the game. Said another way, if UCSD played LMU three games in a row, the most likely outcome would be that they would win two out of three.
The expected goals, expected goals against and expected margin of victory are aggregates of 765 D1 games played since the start of 2024. Think of it as the ‘average’ scores expected of teams with that Elo differential.
The chart below describes how LMU won. The vertical line in the bar graph represents the expected outcome from the model. Anything to the right means the team overperformed expectations, anything to the left means they underperformed. Blue is offense - goals scored, Red is defense - goals against.
In this case LMU was expected to score about 11 goals but they underperformed, only scoring 10. They won on defense. Expected goals against was around 16 but they allowed only 9. That seven goal differential is what swung the margin of victory from -5.5 to a one goal win.
Big West and WCC split games right down the middle this weekend - some close ones and a couple of surprises. Overall, Big West still outpacing WCC, winning 13 of 20 match ups.
