Kia Nurse: early commitment reports; rankings debate

[ctchamps]

It's a great game to observe the coaching styles and the way coaches handle the players with two of the three schools Kia is choosing between. She also gets to observe the atmosphere at Penn State vs. at UConn.

 

[UcMiami]

One outlier and one not ranked which may simply mean we don't know who you talking about or we only cover the US athletes in which case it should bear no weighting at all.

Ice - the problem is at the depth of coverage that Vowel goes to, he ends up dealing with lots of players that don't make it on to lists because they didn't get seen by that ranker or didn't rank highly enough for whatever number they rank to and so that NR is 'valid' in that case. We all see rankings that look way out of line, but trying to interpret why is a fool's errand and adds subjectivity into a process he is trying to keep objective - his raw data is subjective to start and he doesn't want to add to it.
I am in awe of the dedication, enjoy the end product, and take it all with a grain of salt. If Geno wants the player and she commits, that is enough for me - Gardler and Swanier have given me as much pleasure and sweated as much blood for Uconn as players who were ranked much higher.

I would make another suggestion, and one similar to something I once proposed for team rankings: choose a standard cut-off number for all rankings (say 100) and assign that value to anyone who is either unranked or ranked above that number. The accuracy of comparative ranks gets worse the lower down you go to the point of absurdity - after 50, the difference between a rank of 55 and one of 85 becomes almost arbitrary in predicting performance or value and is more likely based on an evaluators days of viewing and not a players skill - the situation of the school they attend and the coaching they receive will have more bearing than the talent they enter with or rather the evaluation they received.

(My suggestion for RPI or similar was to pool the values of wins based on blocks within ranking significantly above or below a schools ranking - for a Uconn in a typical year a win against anyone ranked outside of the top 20 is really the same and to assign a higher value to #35 vs # 55 is sort of silly and the same is true of a loss. The same would be true for a midlevel team ranked in the thirties - wins against top tens are huge and unexpected, wins against teams ranked #65 or #85 should be expected and should really carry a less disparate value in determining their position.)

 

[meyers7]

No doubt. But it has to make sense to you that he adjust her ranking so his overall rankings are accurate, not just based on a formula.

No, it should be based on a formula. It's just a compilation. If he adjusts her ranking, he's ranking her. Why not then adjust other girls he doesn't think are ranked correctly?

If you guys want to complain (and well this is the BY), complain about the services that don't rank Foreigners. Or complain about the ones that do. There's where the problem lies.

 

[Zorro]

Would love to see her playing here. She certainly has the skills and size to play either the 1, 2 or the 3. I looked at her video, and was particularly impressed with her set shot. Or, her jump shot. Or, I guess you could call it maybe a jumping set shot with a hop? Or...oh, never mind.

 

[RockyMTblue2]

I don't think a recruit would go to a game between Pennsylvania and U Conn at Pennsylvania to announce for U Conn( or v-a-v for that matter). We will hear early next week.

 

[Adesmar123]

It's a great game to observe the coaching styles and the way coaches handle the players with two of the three schools Kia is choosing between. She also gets to observe the atmosphere at Penn State vs. at UConn.

Are we collectively trying to nurse this thread along until she commits?

 

[meyers7]

Ice - the problem is at the depth of coverage that Vowel goes to, he ends up dealing with lots of players that don't make it on to lists because they didn't get seen by that ranker or didn't rank highly enough for whatever number they rank to and so that NR is 'valid' in that case. We all see rankings that look way out of line, but trying to interpret why is a fool's errand and adds subjectivity into a process he is trying to keep objective - his raw data is subjective to start and he doesn't want to add to it.
I am in awe of the dedication, enjoy the end product, and take it all with a grain of salt. If Geno wants the player and she commits, that is enough for me - Gardler and Swanier have given me as much pleasure and sweated as much blood for Uconn as players who were ranked much higher.

Well said.

I would make another suggestion, and one similar to something I once proposed for team rankings: choose a standard cut-off number for all rankings (say 100) and assign that value to anyone who is either unranked or ranked above that number.

That's what I would do to. But more probably because I am lazy and it would make it easier.

 

[reno tony]

If she comes to UCONN there is one stat the Boneyarders will need accurately defined. How tall is she?

 

[EricLA]

If she comes to UCONN there is one stat the Boneyarders will need accurately defined. How tall is she?

Canada National Team website lists her at 6'0". The US recruiting websites list her at 5'11". UCONN would list her at 5'9".

 

[easttexastrash]

I know you are hoping that she verbals to Penn State instead of UCONN because it would make a great UCONN 2014 class even better.

I have no dog in the fight. UCONN will be the best team with or without her.

 

[easttexastrash]

Are we collectively trying to nurse this thread along until she commits?

I stopped nursing at 5.

 

[EricLA]

I have no dog in the fight. UCONN will be the best team with or without her.

Well ok then. But how about a bear?!!

 

[easttexastrash]

I don't think a recruit would go to a game between Pennsylvania and U Conn at Pennsylvania to announce for U Conn( or v-a-v for that matter). We will hear early next week.

That would be my point, also. Let's check her itinerary and see when she is scheduled to fly home. But making a point to fly to Penn to make a verbal to UCONN the next day would seem illogical.

 

[easttexastrash]

Well ok then. But how about a bear?!!

Baylor is now a cub, not quite bear...yet

 

[Zorro]

I imagine that Kia is getting an eyeful of what happens to a star player on a so-so team when they play a team that plays defense. Not much fun for Maggie so far.

 

[AndyCT]

No, it should be based on a formula. It's just a compilation. If he adjusts her ranking, he's ranking her. Why not then adjust other girls he doesn't think are ranked correctly?

If you guys want to complain (and well this is the BY), complain about the services that don't rank Foreigners. Or complain about the ones that do. There's where the problem lies.

No, correcting her compilation is not “ranking her”, anymore than making the compilation in the first place is, right? By its very nature compiling disparate rankings into a single composite number is “ranking” her.

That said, if a number is an obvious outlier it should be ignored – when we’re looking at numbers like 11 and 14, and then there is a number like 156, that 156 needs to be evaluated, and probably ignored. I understand and agree that the root problem is really with who ranked her 156 (and her “not ranked”) and that probably comes from her not being seen enough up in Canada – but that should actually be reason enough ignore that 156 and NR – with other rankings all in the top 15, “not ranked” is not the same thing as “unranked”. If she was from somewhere in the States where she might be seen by more ranking services, those two “numbers” (156 and NR) might be legitimate and should carry weight; in this case they shouldn’t. Or as they used to sing when we were kids, “One of these things is not like the others, one of these things just doesn’t belong…”

I understand the problems involved – I have the same problem in “fusing” different data into a single master list for 3000-5000 people were I work (I’ll be spending the rest of the afternoon after the PSU game doing that); sometimes I have to find a tiebreaker and other times decide to just ignore numbers that don’t make sense, and making those decisions is actually the hardest part of the job. But if I don’t, it just goes back to that old computer axiom, “Garbage in, garbage out…”

 

[EricLA]

I imagine that Kia is getting an eyeful of what happens to a star player on a so-so team when they play a team that plays defense. Not much fun for Maggie so far.

And then BAM. Huge 2nd half for her.

Lucas is a player. Love her hustle and intensity. But while PSU made it a good battle, it would have been far different if UCONN had KML and Morgan. I hope that's one thing Kia takes away from the game. Also that on any given night, a different player may step up. Geno will play the kids who are impacting the game and playing really well.

 

[RockyMTblue2]

I imagine that Kia is getting an eyeful of what happens to a star player on a so-so team when they play a team that plays defense. Not much fun for Maggie so far.

At the end of coverage as Penn players lined up for some Alma Mater ritual the camera focused on her for a while - she was totally gassed, with a very dejected look, and wiping away tears.

 

[UConnCat]

That would be my point, also. Let's check her itinerary and see when she is scheduled to fly home. But making a point to fly to Penn to make a verbal to UCONN the next day would seem illogical.

It would be illogical for her to fly to State College; one doesn't just fly in and out of there unless you own a private jet. She probably drove (5hrs to Hamilton, Ontario). She already took an official visit to Penn State so this trip was on her dime.

 

[jdubct]

Canada National Team website lists her at 6'0". The US recruiting websites list her at 5'11". UCONN would list her at 5'9".

Uconn would list her as 5'7", Jefferson's "height." As if height really matters, within reason, it's the heart that counts.

 

[meyers7]

Well ok then. But how about a bear?!!

There's always a bear in a fight. At least around here. Or a dog. Or a dog and a bear.

 

[UConnChapette]

I don't think a recruit would go to a game between Pennsylvania and U Conn at Pennsylvania to announce for U Conn( or v-a-v for that matter). We will hear early next week.

Can't help the fact that the game between two of her three finalists is at Penn State. I tend to think she is about certain where she wants to commit, but wanted to see the two play heads up just to be absolutely sure. If the game happened to be in Connecticut, would you be thinking she is about to commit to UConn? Simply basing your thoughts based on the game location is illogical. Both schools are on her list of finalists. Maybe her parents don't get the Big 10 Network like I don't and this was the only way to see the game? I just don't think you can read anything into her attending the game as meaning she wouldn't go if she was not going to commit to PSU. She may indeed do so. Or she may not. But the game was in PA, not CT. It is what it is.

Sent from my Nexus 7 using Tapatalk

 

[UConnCat]

I can see Nurse picking Penn State. That was a great atmosphere today. Both starting guards for PSU graduate, so there will be plenty of playing time for her. She'll start right away at PSU.

 

[meyers7]

No, correcting her compilation is not “ranking her”, anymore than making the compilation in the first place is, right?

No.

By its very nature compiling disparate rankings into a single composite number is “ranking” her.

No. "Ranking" her is watching, observing her, evaluating her, and then comparing her to other players who have been watched, observed, evaluated.

That said, if a number is an obvious outlier it should be ignored – when we’re looking at numbers like 11 and 14, and then there is a number like 156, that 156 needs to be evaluated, and probably ignored.

Or maybe the 11 and 14 need to be ignored. How about say 7, 8, 13, 21. Is 21 now an outlier? Then instead of just compiling the numbers that people who evaluate them, you start making decisions on what you think the players are like.

If you change the formula, or formulate what you consider and "outlier", then, then you can adjust the raw data. But just changing the raw data.....just because you don't like it, or you don't think it fits, then no you can't do that without prejudice.

 

[Sonny44]

No.

No. "Ranking" her is watching, observing her, evaluating her, and then comparing her to other players who have been watched, observed, evaluated.


Or maybe the 11 and 14 need to be ignored. How about say 7, 8, 13, 21. Is 21 now an outlier? Then instead of just compiling the numbers that people who evaluate them, you start making decisions on what you think the players are like.

If you change the formula, or formulate what you consider and "outlier", then, then you can adjust the raw data. But just changing the raw data.....just because you don't like it, or you don't think it fits, then no you can't do that without prejudice.

Run the Wilcoxon test for statistical outliers.