Just for fun and to pass time, I'll try to mine my data base for interesting D1 women's soccer trivia to see if anyone can come up with correct answers. My data base covers the years since 2007 (roughly 40,000 games), so all the answers must be from then to the present. Also, the trivia are limited to regular season (incuding conference tournaments) games, so they don't include NCAA Tournament games. Today's questions: What is the greatest upset that resulted in a tie? What is the greatest upset that resulted in a loss? Note: These will be games in which the team that got upset had roughly a 99.1% statistical win likelihood!
Ooo...I like it and I’ll give it a go. Loss: UNC 0 - 1 William and Mary (2007) ? Tie: Couldn’t find one just searching on my phone. Will look harder on my laptop later.
The first game that comes to my mind is Princeton beating UNC, 2-1 in OT, in the 2017 NCAA tournament at WakeMed. UNC was ranked #2 at the time and Princeton was #13 so I'm not sure that is the correct answer due to the Tigers' relatively high ranking. The William & Mary win over UNC that OGsoccercoach mentioned is a good one but it was really early in the season when both teams were 4-2. That Tarheels team was #2 at the time and was loaded with talent. Tobin Heath, Casey Nogueira, Allie Long, Yael Averbuch, Meghan Klingenberg, and Ashlyn Harris were all on that team. Overall it was probably a more talented UNC team than the one that Princeton beat in 2017. The tie is a hard a question and I really have no idea. I considered UCLA's draw with Duke in the 2017 NCAA semifinals to be an upset but there are probably other ties that are far bigger upsets. UCLA ended up advancing to the final vs Stanford on PKs. cpthomas' qualification of "these will be games in which the team that got upset had roughly a 99.1% statistical win likelihood" almost assures that my answers are not correct.
2010 NCAA tournament Washington vs #2 ranked Portland. 1-1 tie UW win in PK’s. Portland was loaded and one of the favorites to win tournament
In 2012 Ohio State and Oakland tied 1-1 in the first round of NCAA Tournament, with Oakland moving on in PKs. Officially goes down as a tie with Oakland advancing. In 2008 South Dakota State over Colorado won 1-0 in the first round of NCAA Tournament.
All those games are great ones and memorable upsets. I may have used the wrong term with "great" upsets, as they all were great upsets. I remember all of them, and I was at the Portland/Washington game. In terms of the greatest upset, in terms of the widest gap between the teams, however, none of them makes the grade. Using the North Carolina loss to William and Mary, as an example, UNC ended the regular season with an RPI of 0.6883 and a #1 rank. W&M eded with an RPI of 0.6100 and a #26 rank. With that rating difference, and no adjustment for home field advantage, UNC's win likelihood was 82.0%, which means you could expect it not to win 1 out of 5 games. I think the other games guessed might be in the same ball park. According to the current version of the RPI, the greatest upset ending in a tie was in 2009, Samford 0 v Wofford 0 @ Samford. Samford had a rating/rank of 0.5708/55 and Wofford was 0.3792/301. Next comes 2015 with LSU 1 v Northwestern State 1 @ LSU. LSU was 0.5958/40 and Northwestern State 0.4218/280. Then comes 2017 with South Carolina 1 v Georgia 1. South Carolina was 0.6834/3 and Georgia 0.5171/129. The greatest upset ending in a loss was in 2017, Vanderbilt 0 v Rhode Island 1 @ Vanderbilt. Vanderbilt was 0.5918/39 and Rhode Island 0.4358/269. For these games based on the location-adjusted RPI difference between teams, the favored but upset team had a 99.1% win likelihood, which means if the teams played 100 games, you would expect the favored team to win 99 of them. I checked Massey, and using his ratings (the best, I think), the greatest upset ending in a tie was 2013 Baylor 0 v McNeese State 0 @ Baylor. He had Baylor at 50 and McNeese State at 278. He had the LSU v Northwestern State tie coming in next. For the greatest upset ending in a loss he likewise had the Vanderbilt v Rhode Island game. In these games, the favored but upset team had a 99.7% win likelihood.
For the next question, let's try a Hall of Shame one: Since 2007, which team was the stronger team in the most one-sided matchup?
Stanford's shameless and classless 15-0 drubbing of Prairie View A&M in the postseason this past season comes to mind. 41 vs 0 shots.
I'll check that one out -- it's not a regular season game, so there's not full shame since Stanford didn't voluntarily schedule the game. But, maybe it will be up there with the best -- worst?
I missed/forgot the regular season game requirement. This is a hard question. I would assume that the answer will be a team that is a P5 school vs a lower level mid-major. I tried to cheat and google an answer. I don't want to side-track the thread but I did find some interesting soccer blowouts. A game in the playoffs for the Madagascar national championship in 2002 AS Adema 149 SO l'Emyrne 0 In a a 2019 women's game in Brazil in the Rio state league Flamengo 56 Greminho 0 Sorry for the thread drift but I found these so remarkable that I was wondering if I was reading correctly....
No problem, PlaySimple, with going to the NCAA Tournament or even non-college games. This is for fun and to pass the sequestration time.
While I'm at it, I'll try to post some stats information related to the trivia questions. For the first question, about the greatest upsets, here are some tidbits. Home Field Advantage. For the version of the RPI the NCAA currently uses for women's soccer, home field advantage is worth 0.0150 -- that's on average. If you want to determine the result likelihood of a game, the first step is to determine (or estimate) each team's RPI rating. Second, you calculate the difference between the two teams' ratings. Third, if the game is at a neutral site, you leave rating difference as is. If the higher rated team is the home team, you increase the rating difference by 0.0150. If the higher rated team is the away team, you decrease the difference by 0.0150. It's possible, in subtracting 0.0150, that you will change which team is higher rated. Result Likelihood. The fourth step is to compare the game location-adjusted rating difference to a probability table. For the version of the RPI the NCAA currently uses for women's soccer, this is the probability table: On this table, with your location-adjusted rating difference in hand, you look at the right-hand column to see which number your difference is closest to. When you've found that number, then you are in the row that tells you the likelihood of your higher rated team winning, tieing, and losing. Using the 9/3/17 Vanderbilt 0 v Rhode Island 1 game as an example, and based on the teams' end of season RPI ratings: 1. Vanderbilt's RPI is 0.5918 and Rhode Island's 0.4358. 2. The rating difference is 0.1560. 3. Vanderbilt is the home team, so add 0.0150 to the rating difference to get the location-adjusted rating difference of 0.1710. 4. Look on the right-hand column of the table for the number that is closest to 0.1710. That turns out to be 0.1801 at the top of the table. This row therefore gives you the game's result probability: Vanderbilt has a 99.1% chance of winning, a 0.7% chance of tieing, and a 0.2% chance of losing. The 0.2% chance of losing means that if the teams were to play 500 games, you would expect Vanderbilt to lose 1 time. They did it! The table shows something interesting: For D1 women's soccer, there is virtually no rating difference at which the better rated teams is absolutely certain of winning. In other words, there can be an upset in virtually any game. Looking at all games together, rather than broken into the table's groupings, the likelihood of the better rated team (after the game location adjustment) winning is 72.7%, tieing 10.7%, and losing 16.6%. In other words, from a rating perspective you are likely to see an upset in roughly 1 out of 4 games. I don't know what this table would look like for other sports, but I suspect that the upset likelihoods for soccer are high compared to other sports. Also, the likelihood of a tie is 10.7% looking at all games, and even for the most closely rated games is only 16.3%. So, if you predict a game will be tied, your chance of being right is less than 1 in 5. If you are betting on a tie, you'd better get really good odds. In the table, each row's numbers are based on that row representing 2,000 games. With that size data set, the numbers are reliable.
Back to the Hall of Shame, for the teams that scheduled the most one-sided games in their favor. This didn't include NCAA Tournament games, but one guess was Stanford in this year's NCAA Tournament opener against Prairie View, which Stanford won 15-0. Stanford's regular season RPI was 0.7070/#1 and Prairie View's was 0.4554/246. The RPI difference was 0.2516, to which one would add 0.0150 for Stanford's home field advantage. As a check, Stanford's Massey had Stanford at 2.4245/1 and Prairie View at 0.5549/304, for a difference of 1.8696. No, that wasn't the most one-sided game. According to the RPI, the correct quiz answer is 9/19/14 Arkansas Pine Bluff 0 v Southern California 9 @ So Cal. UAPB's RPI was 0.2838/331 and So Cal's was 0.5858/19, for a rating difference of 0.3020, before the home field adjustment. (UAPB's rank made it next to last in Division 1.) Next comes 9/14/12 Jackson State 0 v Florida State 5 @ FSU. Jackson State was at 0.3784/298; Florida State at 0.6708/5; rating difference of 0.2924, before the home field adjustment. It looks like FSU took its foot off the gas in that one. According to Massey, the most one-sided game was 9/19/07 Alabama State 0 v Alabama 9 @ Alabama. Alabama State was at -2.2980/314; Alabama was at 1.1490/130; the rating difference was 3.4470, before the home field advantage adjustment. If you noted the negative Massey rating for Alabama State, that says they were really, really bad. There were only 314 teams in Division 1 at that time, so they were at the bottom of the rankings. For the next most one-sided game, Massey has the above Arkansas Pine Bluff (-1.3900/331) v Southern California (1.9200/20) game. Rating difference of 3.3100. If you're wondering why the RPI doesn't have the Alabama State v Alabama game as the most one-sided, it's because the RPI, due to its structure, has trouble rating teams from the Southwestern Athletic Conference. It greatly overrates them. So, it overrated Alatama State to the extent this game doesn't appear to have been the most one-sided.
Since the Stanford v Prairie View game came up, for today's question lets go with: What is the most goals scored in a game by a team since 2007? Who scored the goals, and against whom? (Go earlier in time, if you want.)
I tried to cheat on this one but couldn't get an answer for most goals by a team in a game since 2007. I did find a game in which Mississippi Valley scored 25 goals vs Alcorn in October of 2006. Most goals in a game since 2007 is Laadi Issaka with 7 and oddly enough that was also in a Mississippi Val. vs. Alcorn game on Oct. 28, 2016. The NCAA DI recorn book is here. http://fs.ncaa.org/Docs/stats/w_soccer_RB/2019/D1.pdf This is unrelated but when I was looking through the record book I saw that one of my all-time favorite players, UNC's Yael Averbuch, is tied for the record for fastest goal in a game at 4 seconds vs Yale in 2006. Many of you have probably already seen the video of the goal but if not, I found it on YouTube:
PlaySimple's 25 for Mississippi Valley v Alcorn State in 2006 is great. I figured someone would find the record book. Since 2007, here's what we have: 8/23/15: @ Howard 17 v Hampton 0 9/18/09: South Carolina State 17 @ Alabama State 1 8/25/15: @ VMI 16 v Hampton 0 8/21/15: Canisius 16 @Hampton 1 8/25/19: Louisiana Tech 15 @ Alcorn State 0 9/6/09: @ Troy 15 v Alabama State 0 8/31/14: @ Jacksonville State 15 v Mississippi Valley 0 9/6/09: @ McNeese State 15 v Alcorn State 0 8/21/10: @ Miami OH 15 v Alabama A&M 0 2015 was Hampton's first year with soccer. Other than that, all of the losing teams are Southwestern Athletic Conference teams. Perhaps this puts Stanford's 15-0 win over Prairie View in context. Which would be more humiliating? Losing 15-0 to Stanford, ranked #1, or losing 15-0 or worse to any of the above winners? It seems that at least some weak teams have no sportsmanship compunction about beating the crap out of a very weak opponent.
For today's trivia: Which conference, on average, has the highest percentage of home games? Edit: Should have been trivium. My days of Latin are coming back to haunt me.
My first thought was that it had to be the Ivy League. I am basing this on the league philosophy that student is emphasized in the student-athlete equation. The very first rule in the Ivy manual regarding travel: "All travel should be arranged so as to minimize missed class time." Ivies also have more constrictive rules governing practice times and travel. Not being allowed to get on the road for an away contest until later in the day limits where teams can go. And with the Ivies starting practices later in the year before other school, there is a limitation on early season travel. The first permissible practice date in women’s soccer is only 10 days prior to the first permissible contest date. The first permissible contest date in women’s soccer is 13 Fridays before Thanksgiving. This past year that wasn't until the last Friday in August. Many teams travel in the early part of the season and that is pretty much unheard of in the Ivy League. I didn't have the ambition to check beyond last year but based on last year, my hunch was correct. If I did my math correctly, this is what I came up with: Ivy 56% of contests are at home ACC 53.1 Big 10 52.5 WAC 50.8 Big South 50.5 PAC 12 50.2 Mountain West 49.8 Big East 49.7 Big West 49.7 SEC 49.4 At the bottom was Big Sky at 40.3 I ran these numbers hastily so they very well could be wrong.
Conferences with the highest home field percentages: Michael M's SEC is correct, over time. And PlaySimple's numbers for 2019 are excellent, with the Ivy League on top. Below are two tables and a chart: The first table shows game location data and percentages since 2013. I use 2013 since that was that was the first year after completion of a period of major conference membership changes. There have been some minor ones since then, but the conferences essentially have been pretty stable since 2013. The table shows home field as a percentage of all games that were not at neutral sites and as a percentage of non-conference games not at a neutral site. The second table is for 2019 and shows detailed breakdowns by conference. The chart shows the relationship between conference strength as measured by average ARPIs and conference home field imbalances. The relationship between the two is clear as a bell.
Given the stringent travel rules in the Ivy, I surprised that those schools don't have the highest percentage of home games every year.
Any mod......^ why does the software on this site only allow a 30 minute window to edit posts? "I surprised"
I forgot to post a trivia question yesterday. So, here's one for today: What is the average tenure of today's D1 women's soccer head coaches? What is it for female head coaches? What is it for males? I have no idea what the answer is, but I think I have the data to come up with the correct answer. Make your best guesses!
I would have no idea of where to look for the data on this except to select random samples of schools and look at the tenures and then average it out. (actually it wouldn't be so random because I would select a certain number of schools that I know are coached by women and the same number coached by men). When I think of prolific soccer programs with a lot of wins, from any NCAA division, most have coaches that have been in their positions for a long time. Men - Anson Dorrance - 41 years coach of UNC women and 9 of those years also coaching the men's team. - Brian McManus - 30 years at UCSD (now retired) - Pete Felske - 32 years at D3 Wheaton - Jerry Smith - 33 years at Santa Clara - Mark Krikorian - 15 years at Florida State (Hartford and Franklin pierce before that) Women - Becky Burleigh - 25 years at Florida - Aliceann Wilber - 40 years at D3 William Smith - Amy Reifert - 29 years at D3 Chicago - Nancy Feldman - was at Boston U for 23 years - Paula Wilkins at Wisconsin for 14 years As I said, I really have no idea about the average tenure of coaches. If I had to guess I would say that it is between 5 & 6 years with women having a slightly longer tenure than men. Without trying to wade into a controversial area, I think that universities keep women as coaches longer because there is a lack of them. I've always thought that if there are multiple qualified coaches for an open position that the best coach should get the job without consideration to the sex of the coach. Also, if there are two coaches that are equally qualified for head coaching position of a women's soccer team and one is a woman and the other is a man that the woman should get the job. I'll catch criticism for this but my opinion is that, in general, women do a better job of coaching women than men do if all else is equal.