06-05-2019, 11:32 AM
(This post was last modified: 06-05-2019, 12:01 PM by ColoradoTree.)
In terms of Pac-12 teams, we have 3 hosts--#1 Stanford, #7 UCLA, and #10 Washington. #16 USC is the #2 at UW's regional, #18 Cal is the #2 at Kansas's regional, #34 ASU is the #3 at Vanderbilt's regional, #36 Washington State is the #2 at our regional, #44 Utah is the #3 at Michigan's regional, and #49 Arizona is the #3 at Pepperdine's regional.
Colorado looks to have opted out entirely, and it's unlikely that they'll get to #90 Oregon. Some teams pass at this point in the draft, since they don't see the value in paying to fly across the country to get crushed in the first round, but not that many teams pass.
#53 South Florida has opted to be the #4 in our regional, meaning it chose to play us rather than be the #3 at Georgia, the #4 at UNC, the #3 at Duke, the #4 at Vanderbilt, the #4 at Texas, or the #4 at in-state Florida State. Maybe it's a "let's challenge ourselves by taking on the best" kind of thing, but there are a lot closer and theoretically less dangerous teams that they passed on playing to play us.
That's an excellent question. I struggle with that myself. All I really know is two things: (1) UTR doesn't even count matches where the UTR difference between players coming in was 2.50 or more and the favored player wins as expected, since those should be blowouts; and (2) UTR reps say that matches where players are within a point of each other should be competitive. I saw one post somewhere that said where there is a difference of more than 1.0 points, the lower-rated player wins just 3% of the time. But we rarely see such lopsided matches in NCAA tournament play, so it would be far more meaningful to see data about how often players win when their UTR is 0.1 higher, or 0.25/0.5/0.75 higher, etc. I don't think UTR has released that, though. I wish they would, so I'd know how linear the difference was between a coin flip with a tied UTR and a near-certain outcome when the difference is 1.0 or more. Are 0.25 differences also basically coin flips, or is it more like the favored player wins 2/3 of the time? If someone has done that analysis, I haven't seen it.
From my unscientific tracking of how we've been doing the last couple of years at NCAAs, it sure seems like there's a meaningful difference in being favored by 0.8 or so than there is in being favored by 0.25 or 0.05. If someone like Gordon (11.72 at season's end) were to play someone ranked at 11.00 or 10.9, I would expect Michaela to win that match handily.
FYI, both Gordon and Lord had Top 10 UTRs at the end of the NCAA season. Gordon was #6, with her 11.72 just 0.18 behind the top score of 11.9, shared by Michigan's Kate Fahey and Miami's Estela Perez-Somarriba (Georgia's Jokic was in third with 11.85). Lord tied for 9th with UNC's Makenna Jones, as they both had 11.67.
Colorado looks to have opted out entirely, and it's unlikely that they'll get to #90 Oregon. Some teams pass at this point in the draft, since they don't see the value in paying to fly across the country to get crushed in the first round, but not that many teams pass.
#53 South Florida has opted to be the #4 in our regional, meaning it chose to play us rather than be the #3 at Georgia, the #4 at UNC, the #3 at Duke, the #4 at Vanderbilt, the #4 at Texas, or the #4 at in-state Florida State. Maybe it's a "let's challenge ourselves by taking on the best" kind of thing, but there are a lot closer and theoretically less dangerous teams that they passed on playing to play us.
(06-05-2019, 11:21 AM)oldalum Wrote: What is considered a meaningful difference in UTR?
That's an excellent question. I struggle with that myself. All I really know is two things: (1) UTR doesn't even count matches where the UTR difference between players coming in was 2.50 or more and the favored player wins as expected, since those should be blowouts; and (2) UTR reps say that matches where players are within a point of each other should be competitive. I saw one post somewhere that said where there is a difference of more than 1.0 points, the lower-rated player wins just 3% of the time. But we rarely see such lopsided matches in NCAA tournament play, so it would be far more meaningful to see data about how often players win when their UTR is 0.1 higher, or 0.25/0.5/0.75 higher, etc. I don't think UTR has released that, though. I wish they would, so I'd know how linear the difference was between a coin flip with a tied UTR and a near-certain outcome when the difference is 1.0 or more. Are 0.25 differences also basically coin flips, or is it more like the favored player wins 2/3 of the time? If someone has done that analysis, I haven't seen it.
From my unscientific tracking of how we've been doing the last couple of years at NCAAs, it sure seems like there's a meaningful difference in being favored by 0.8 or so than there is in being favored by 0.25 or 0.05. If someone like Gordon (11.72 at season's end) were to play someone ranked at 11.00 or 10.9, I would expect Michaela to win that match handily.
FYI, both Gordon and Lord had Top 10 UTRs at the end of the NCAA season. Gordon was #6, with her 11.72 just 0.18 behind the top score of 11.9, shared by Michigan's Kate Fahey and Miami's Estela Perez-Somarriba (Georgia's Jokic was in third with 11.85). Lord tied for 9th with UNC's Makenna Jones, as they both had 11.67.