Separate names with a comma.
Discussion in 'Women's College' started by Soccerhunter, Nov 5, 2018.
Whoops. Error... forgot to credit Wendys with a score of 26 and to add insult to injury misspelled his/her screen name, which is now un-editable but which will be corrected next week!.
Thanks for doing this SH. You always make it fun!!
Thanks for all your hard work and efforts doing this. Makes the tourney much fun to follow along.
Round 2 contest results are in.
The new leaders are Millenium and WWC_Movement with 62 points each. But this leading duo is followed closely by four competitors only one point back.
My notes include only BOSCO calling Wake Forest as a second round winner, and only four contrarians picked NC State to advance over Santa Clara (RtD!, Kurt Kline, Soccerlol, and Stubifer.) And speaking of contrarians, a tip of the hat to NilamCA. You've got to admire someone who takes risks and sticks with his favorite teams, it's just that it sometimes doesn't pay-off... maybe next year it will!
For the Humans, this contest is wide open yet with two-thirds of the field within 6 points of each other. However, the third round tomorrow will thin the ranks considerably.
Over on the Machine/Statistical side, the Bennet Ranks has edged ahead, and like the human side, tomorrow will be the water shed. The Non-Conference RPI got a boost on Friday because NC State's non conference results were so good this season, but this wont last... Top Drawer took a hit yesterday, but could look like geniuses tomorrow if TAMU, PSU, and Virginia all win.
We shall see.... Full results below.
58 Kurt Kline
55 Tom 81
51 Glove Stinks
62 Massey Ratings
59 CPT's 5 Iteration RPI
59 United Soccer Coaches
58 NCAA RPI
56 NCAA Non-Conf RPI
53 Top Drawer Soccer
Yes, the Non-Conference RPI is heading for the trash can.
@Soccerhunter which round usually results in the most points?
If I recall correctly, we usually know the winners (or at least the top 2-3 capable of winning) before the final four, no?
I'm surprised I'm doing so well, considering I've put a lot less time into my picks this year compared to previous years. Although, I have a feeling, I'm going to get killed in the next 2 rounds.
TO answer your question, the second and third rounds (yesterday and tomorrow) are weighted the most. Each round will potentially produce 48 points in this system. (Round 2 yields 3 points for each correct pick, and Round 3 yields 6 per correct pick. The first round yields 32 points at one point per correct pick. The last three rounds yield 8 points per pick for a potential pick up of 32, 16 and 8 per round.)
The reason that the second rounds are weighted the most is because these are the rounds that typically produce the most "upsets", and if a contestant can navigate the first three rounds with so many possible picks (56 total picks for 128 possible points) then if a contestant gets a good percentage of them correct (and especially hits on the eventual elite 8) he or she will be in a good position to win the contest along with perhaps a half dozen others. Once you get to the elite 8 then most of the remaining matches would not be considered "upsets" in a real on-the-field way as it is those teams that are proving to be the best in the field in November (regardless of the expectations at the beginning of the month.) In this system one can still separate from the other contestants in the last three rounds especially if there is a variety of picks for the final four an the eventual champion, but typically, there is agreement by most contestants regarding the final four and an eventual winner.
So this system offers offers 56 points for the last three rounds but for many contestants it is a conundrum as to how to yet elevate yourself above the pack. Since there is typically a 40%-70% consensus at that stage of who the winners will be, if you want to gain on the other top contestants you have to pick an outlier that most of the competition will not and it's a risk-reward situation -not picking the team you think will win. Consequently, the real work is the first three rounds to gain a substantial lead and that is why these rounds are weighted as there are more consequential choices to get points.
Is this helpful?
Yikes! Unfortunately no miracles for Duke this year.
I'm sorry to report that Duke is getting crushed by the Hoyas.
OK... The third round is finished... and a new leader is atop the Human Division in this contest. But with 56 points yet to be awarded, nothing is set yet.
Soccerlol is laughing indeed and currently sits at #1 with 100 points. However, there is plenty of close competition with Mellennium with 98, and Dmthomas49 and Playsimple are next at 97, followed by Headspun, lovethegame, and WWC_Movement.
And while it is true that at least the top half of the contestants might have a shot numerically, it is likely that the top 6 of these 7 contestants will be the winner. A quick analysis of the remaining picks shows that only two of the final four picked by WWC_Movement are in the elite 8 as is his pick for the winner. Soccerlol and Millennium have all four of their 4th round picks still in the running after today's action, while DM, Playsimple, Headspun, and lovethegame have 3 teams in the present mix. But that's just statistical probabilities.
From a specific pick point of view, only soccerlol, Millennium, and playsimple have Georgetown picked to win. If the Hoyas polish off Baylor, these three contestants may be a leg up. And all of this list have Stanford taking it all with the exception that Soccerlol has UNC. Here again, if UNC wins, the so does Soccerlol. On the other hand if Stanford sails through, then parsing among the remaining five will depend on if they picked the right "other three" in the final four and Stanford's victim in the championship match.... and there are some key diferences there!
Much more to see!
Over on the Machine/Statistical Division, the pattern is becoming clear. NC State's and Wake Forest's magical runs have ended and so has the fortunes of the "Non Conference RPI" entry. And it looks like the Top Drawer bracket picks will be next to last. The remaining four are all within easy striking distance of each other (104 to 100) and so one hiccup could change the pattern, and since there is only one small difference in their picks from here on in, we can predict the winner if only one thing changes. If Georgetown beats UNC or UCLA in the semifinals, then the entry based on the United Soccer Coaches poll will win, if UNC or UCLA beats Georgetown, then the present leader, Bennett Ranks will take the prize... but barely.
One thing is for sure: Once again the Statistical Division on average will be way ahead of the humans.
Full Scores follow:
82 Kurt Kline
81 Glove Stinks
79 Tom 81
104 Massey Ratings
101 CPT's 5 Iteration RPI
101 United Soccer Coaches
100 NCAA RPI
83 Top Drawer Soccer
80 NCAA Non-Conf RPI
Great, and very quick, work Soccerhunter. What is BennettRank's score?
As abysmally as the NCAA's Non-Conference RPI has done, Top Drawer Soccer is only slightly ahead of it.
Something related to which I'm tracking final results, from WWC_Movement: "RPI sucks royal."
Great! Another Sleep Deprived mistake... Sorry. I meant "Massey" but somehow typed "Bennett". Not sure how to explain that!! Nest week I'll try to do better!
Ah, that makes me happy. In my opinion, Massey is the best and this year's numbers are bearing it out so far.
The mistake probably has something to do with the two consonants in the middle.
Well, the FSU-PSU match was a hard one for me.
I picked FSU to win in my bracket, so I guess it's a small win for me, even though PSU lost.
PSU showed out in the 1st half.
FSU regained footing in 2nd.
Nothing to be ashamed of there my friend!
Those 2 free kicks from about 23 yards out were your best chances. Gotta put it on goal!
FSU home crowd was unreal. Sold out and loud.
I'm certain that made the difference.
Good luck in the final four!
The 4th Round is finished and we have a typical set of possibilities based on the current top scorers in our prediction contest.
Soccerlol holds the lead with 132 points followed closely by Millennium with 130 points. Then there is a 9 point gap and three more closely grouped contestants: Dmthomas49 with 121, and lovethegame and Wendy's with 119 each. The first and second place winners will come form this group.
Here are the scenarios. If FSU and Georgetown win their next games and since no one selected Georgetown to make the finals, then nothing will change in the present scoring for the 25 human contestants except for the two contestants who chose FSU to beat Stanford and to then win the finals. Both of these contestants are at the 103 and 94 point level and have no chance of overhauling the top three but Tom81 could could possibly tie for 4th if FSU beats UNC in the finals.
To put it simply if Stanford and UNC win their next games and UNC takes the championship, then Soccerlol will be the undisputed winner. If Stanford wins the championship over UNC then Millennium will take the bow as the contest winner. We won't know until the last whistle blows next Sunday.
Over on the Machine/Statistical division, the non-conference RPI and the Top Drawer bracket have dropped out of contention which leaves the other four entries in a virtual tie (all within 4 points.) The scenarios are that if Stanford beats UNC for the championship, the United Soccer Coaches will fall off by 8 points and the Massey ratings will take the prize (But the Iteration 5 RPI and the NCAA RPI tucked in close behind.) If Georgetown beats UNC, then the United Soccer Coaches poll will come out on top with the other three close behind.
Here are the results as of 11/24/18
106 Kurt Kline
103 Tom 81
97 Glove Stinks
136 Massey Ratings
133 CPT's 5 Iteration RPI
133 United Soccer Coaches
132 NCAA RPI
107 Top Drawer Soccer
80 NCAA Non-Conf RPI
Thanks for the update and hard work you put into this contest @Soccerhunter !!
I agree with MiLLeNNiuM, thanks a ton, Soccerhunter. And, very fast work after each round!
Hopefully, this year's contest will put to rest anyone's contention that Top Drawer Soccer has credibility at ranking teams. (I consider the NCAA's NonConference RPI to have been put to rest a long time ago.)
On the other hand, I'm surprised by the United Soccer Coaches' success. They've done well. The individual conferences' coaches also did quite well, from a computer/machine perspective, in their pre-season predictions of where teams would end up in their conference standings. I guess I should pay more attention to how the coaches rank teams.
Thanks again soccerhunter. It's a lot of extra work for you but hopefully someone comes and takes over next time. I wouldn't mind doing it next year if I only knew how the calculations worked and all that. I'd probably just keep it the same basic format as you have to make it easier.
This further confirms my opinion that TDS is oft ill-informed about all things soccer. Sometimes I think that the primary purpose of TDS is to bolster the egos of parents of kids that play at the club ages.
You only think that sometimes?
It is primarily a "for profit" business. The worrisome part is what it means when one of those ego-driven parents can't even score higher than TDS. So embarrassing...