Kia Nurse: early commitment reports; rankings debate

[JS]

I'd hate to try to decide which restaurant to go to with any of you folks.

Or at least any two of you.

If you disagreed, which you probably would, you'd argue 'til we starved.

 

[JRRRJ]

I'd hate to try to decide which restaurant to go to with any of you folks.

Or at least any two of you.

If you disagreed, which you probably would, you'd argue 'til we starved.

Not me, I eat anything.

Except salty fish.

 

[Zorro]

Geno's, mebbe?

 

[pinotbear]

I'd hate to try to decide which restaurant to go to with any of you folks.

Or at least any two of you.

If you disagreed, which you probably would, you'd argue 'til we starved.

 

[cohenzone]

Geno's, mebbe?

Yuck.

 

[baggerbob]

The only ranking that matters in the end is Geno wanted her, case is closed.

 

[ctchamps]

The only ranking that matters in the end is Geno wanted her, case is closed.

That's the way I see it. Hard to figure why so much discussion about her ranking when this is the salient point.

 

[HuskyFanBlue]

o

 

[Icebear]

Not me, I eat anything.

Except salty fish.

Especially, hairy, salty fish, except when used in salad dressing.

 

[doggydaddy]

I'd hate to try to decide which restaurant to go to with any of you folks.

Or at least any two of you.

If you disagreed, which you probably would, you'd argue 'til we starved.

I would just argue the method of deciding. I'd bring a sandwich.

 

[Icebear]

Of course you'd want us to jump at the first opportunity rather than push the decision forward with a hop, skip, and a jump.

 

[Adesmar123]

This is looking more and more like an off season thread. I thought it would have died a timely death after the season began.

 

[jsloanjr]

Who cares about all this back story today. She is now a Husky and we are very lucky to have her and a great job on the coaching staff............

 

[AndyCT]

And that's fine. Then you should know (really if you are in the mathematical world, you should know this) you just can't throw out outliers because you don't like them, or think they are outliers. You have to have a formula for determining what is an outlier and then that has to be applied to "all" the players. Not just, well we don't like this one on this player, because we like this player and want her to look better, so let's change it (ignore it).

Actually, I really am in the mathematical world. I come from a family of “technical people” – all the boys became engineers and the girls are all analysts. While I initially worked in engineering, today I do statistical work on my off days (which is why I don’t post very often – busy busy busy) on staffing for both my company (secondary) and the labor union (primary) which includes projecting pilot staffing levels in different bases and airplanes based on past trends. So I am all about data and trends on a level that would make your head hurt.

You seem to be stuck on the idea that I and most of the others are arbitrarily throwing out the low ranking because it is low - that’s simply not so – and that we “have to have a formula for determining what is an outlier and then that has to be applied to "all" the players” – that’s also not so. Judgment is an important human trait and it doesn’t have to rely on some (supposedly) infallible mathematical formula.

Most animals can recognize patterns and deviations. Studies have shown that pigeons can tell a Van Gogh from a Picasso. They don’t actually recognize the paintings (nor would I), they simply recognize patterns in the painters’ styles. So do humans, and (almost) all of us look at the pattern 11-11-14-156 and know that “one of these things is not like the others, one of these things just doesn’t belong.” A pigeon would, anyway.

Formulas can certainly be used to screen for outliers, but the final decision can (and usually should) still be made by a human (all my programming at work does just that – it flags outliers, but I decide whether to keep them or not). As another poster mentioned we could use a Wilcoxon filter (or a Kamman filter or others) but something much simpler like standard deviation, or even just a wide percentage would also work – if a number is more than 50% or a hundred numbers different it is flagged to be evaluated. Or simply, like the pigeon, if it just doesn’t look right.


JRRRJ – makes a good point, I think I may have said something kind of similar earlier about the fact that the rating services themselves are compilers. By their nature anything they put out can be iffy. Generally there is a consensus on the better players that are observed a lot so the numbers come out closely for the higher rankings; there is much more spread among the lower rankings – kind of a crapshoot the lower you get...which is why blind adherence to including anything they publish doesn't make sense. We know they can be wrong.

The bottom line is that, while Kia’s Boneyard ranking isn’t all that important, if it is going to be useful, it needs to be as accurate as it can be given the limitations that JRRRJ mentioned – subjective, sparse, and observation-biased data. Otherwise, what’s the point in using a number that we know isn't really right?

And really, all the arguing aside, does anyone really believe that Geno thinks he signed the 156th best player in the country? I bet he doesn’t…

 

[HGN]

You may be right, but it struck me this morning that she's just opening herself up to more pressure by not announcing now. Every program must still be pursuing her, and she might be getting even more pressure from the Gamecocks now than later.

Sam , I think you have a good point. But being thought of as the No.1 player has a lot of pressure with it. She's probably damned if she do , and damned if she don't. She and her family probably are playing this out the best way they know how. Whether announcing now is better , or later , one is just as bad as the other in their eyes.

I guess we have to wait until Spring to find out. In the meantime let's keep our fingers crossed.

 

[meyers7]

Not speaking for Eric, but don't you see how this looks to many people to be a unique situation and not just a formulatic change for everyone?

Yes, and that's the problem.

 

[meyers7]

I absolutely agree. In Neopolitan ice cream, the chocolate needs to be in the middle.

Are you freaking kidding me??????

 

[UcMiami]

This is looking more and more like an off season thread. I thought it would have died a timely death after the season began.

Silly you!
They only play 2 or 3 games a week taking up 3 hours - add in 3 hours before each game to predict the outcome and analyze, and 24 hours after the game to critique the announcers, pan the other teams 'superstar', take umbrage at some throw away line by the coach and or one of the players, and praise our superior players (except the ones who had a less than world class outing) and you have only consumed 60 to 90 hours out of a possible 168 hours! What are we true fans supposed to do with all that down time!?!?

 

[meyers7]

i havn't read the entire thread

Finally someone with some intelligence.

 

[doggydaddy]

Yes, and that's the problem.

Please see Andy's post.

It's just Sesame Street.

11,11,14,156.

You don't need a system to see that if you include the 156 you have screwed up the rating.

 

[meyers7]

Just stop. Stop trying to put words in my mouth or decide what I'm trying to say. Seriously, knock it off. I gave my reasons, you disagree, and continue to go on and on and on and on and on. You don't agree - I get it. You think we "only want ourselves to feel better." I guarantee you that's not the case and I've said as much. So stop trying to attribute something to my comments that do not exist.

Sorry, the facts is the facts. Deny all you want. We know. we know (we know).

It's just not accurate.

But I am sure, if everyone did exactly as you wished, exactly what would make you happy, then, AND only then, it would be accurate.

Yea, we get that about you.

 

[meyers7]

I'd hate to try to decide which restaurant to go to with any of you folks.

Or at least any two of you.

If you disagreed, which you probably would, you'd argue 'til we starved.

No me baby, I ain't never starved.

Now if we got into dividing up the bill........oh boy.

 

[meyers7]

I'd bring a sandwich.

You know something.....I think you would.

 

[UcMiami]

Andy - very interesting and I agree having spent most of my working life compiling databases from suspect sources and dealing with the issues of outliers in that data. The business was tracking pricing in newspaper advertising for products and the problem was to determine if the $9.99 price for a $300 camera was first a data entry error internally, but also how to handle the case where the store actually advertised that price and was it a misprint or a loss leader (teaser to get customers in the store. It was a published price, so do you report it to the client or not?
And while I agree that there is generally a consensus on a player, that may also be pernicious - Joe says player x is great, Judy says player x is great, do I want to stake my reputation on saying she isn't or do I just role with the flow and agree. Happens less at the top end, but I think it happens more frequently in reverse so a player ranked in the 30's that someone thinks should be in the top ten doesn't get that ranking.
I do think after about player 30 the standard deviation gets pretty wide and when you get out above 50 the rankings begin to be meaningless but can have a huge difference in a players overall ranking.
In an ideal world if I were to do this work (and I am not going to and I am not complaining about the work that VG does in any way) I would probably assign a value of 50 to all players ranked 50 or above (or unranked) in all services. That would still not answer the special case for a player like Kia, but it wouldn't create quite the mess that an NR or a rank of 156 does. And it would discount the input data at a point where its reliability goes from suspect to ____. And I suspect it would have no effect or minimal effect on the final rankings of schools in the top 20.

 

[meyers7]

Actually, I really am in the mathematical world. I come from a family of “technical people” – all the boys became engineers and the girls are all analysts. While I initially worked in engineering, today I do statistical work on my off days (which is why I don’t post very often – busy busy busy) on staffing for both my company (secondary) and the labor union (primary) which includes projecting pilot staffing levels in different bases and airplanes based on past trends. So I am all about data and trends on a level that would make your head hurt.

Yada, yada, yada. Sorry, that don't impress me much (as the song goes). I've met many doctors, engineers, lawyers and especially educators, who aren't very bright. Not lumping you in, just sayin'. And engineers, I've worked with many of them who don't know their head from a hole in the ground, especially when it comes to how something actually works out in the real world. But I digress.

Nothing personal, but it's the internet. I really don't take a whole lot of stock in anyone telling me how smart they are.

You seem to be stuck on the idea that I and most of the others are arbitrarily throwing out the low ranking because it is low - that’s simply not so –

Actually that's very, very so. That's exactly why some people want to throw it out.

Judgment is an important human trait and it doesn’t have to rely on some (supposedly) infallible mathematical formula.

Ding, ding, ding. Thank you. Exactly what I was saying. People want to make a biased judgement call on a number they don't like. But only that number. They don't want to look at any other outliers, only the one.

Formulas can certainly be used to screen for outliers, but the final decision can (and usually should) still be made by a human (all my programming at work does just that – it flags outliers, but I decide whether to keep them or not).

Ding, ding, ding again. See above.

And really, all the arguing aside, does anyone really believe that Geno thinks he signed the 156th best player in the country? I bet he doesn’t…

We of course know Geno doesn't give a $*&^ about the ranking numbers. However, if he did, would he think he got the #156?, #14? How about Edwards? #8?, #32? What if we'd gotten Rooks, #60, #18.