Pump the brakes guys, don't forget...

[VikingCat21]

ESPN's BPI has us ranked 22nd going into next season. The team may have looked good in the Bahamas and they may be oozing national title potential, but don't forget that according to ESPN's BPI metric we will have to hold off teams like TCU, Northwestern, St. John's and Butler while trying to leap frog the powerhouses in front of us like Illinois, Alabama, Iowa, Mississippi St, Virginia Tech, Kansas St, Syracuse, West Virginia, and Marquette. Hopefully by January we can break into the top 15.

 

[Bkocats]

OP you're so right. We really need to take a step back and listen to the experts. They definitely know what they're talking about

 

[shaudyshaud]

**** ESPN

 

[JerseyCat84]

All those teams aside from W Virginia wouldn’t stand a chance in hell.

 

[ManitouDan]

Espn .. is that the network that used to do sports ?

 

[Big_Blue79]

BPI is based on a formula. The godfather of college basketball analytics, Ken Pomeroy (of KenPom.com fame), has flat-out stated that even his projections are inherently flawed in the early going due to lack of good data. Same here. In UK's case, it's a young team with little returning production and a lot of projection. Of course UK will look bad in BPI in August.

Crapping on BPI for it's early season projections of a hard to project team is pretty dumb. You're either ignorant as to how it works, or harping on a known weakness of it (and similar) systems.

 

[HICATFAN]

**** ESPN

loveESPN ???

Sorry shaudy... I may have misinterpreted that...



AlohaCat

 

[Bungalow Bill]

You lost me at ESPN.

 

[kyjeff1]

BPI is based on a formula. The godfather of college basketball analytics, Ken Pomeroy (of KenPom.com fame), has flat-out stated that even his projections are inherently flawed in the early going due to lack of good data. Same here. In UK's case, it's a young team with little returning production and a lot of projection. Of course UK will look bad in BPI in August.

Crapping on BPI for it's early season projections of a hard to project team is pretty dumb. You're either ignorant as to how it works, or harping on a known weakness of it (and similar) systems.

The OP was being sarcastic.

 

[Levibooty]

So until the season has played out a few weeks ESPN could stand for Extra Sensory Projection Network

 

[VikingCat21]

BPI is based on a formula. The godfather of college basketball analytics, Ken Pomeroy (of KenPom.com fame), has flat-out stated that even his projections are inherently flawed in the early going due to lack of good data. Same here. In UK's case, it's a young team with little returning production and a lot of projection. Of course UK will look bad in BPI in August.

Crapping on BPI for it's early season projections of a hard to project team is pretty dumb. You're either ignorant as to how it works, or harping on a known weakness of it (and similar) systems.

I understand how the BPI and similar metrics work. I also realize it is August and once the games start there will be a lot of movement up and down the rankings as more data becomes available. But listing one of the two or three best teams in the country at 22nd, and listing a Marquette team that finished 7th in the Big East last year and lost their leading scorer at 4th is just stupid. There's no point in putting out a preseason ranking for a metric that needs relevant data to be accurate until said data is available.

 

[Cardsstink]

Haters have to hold onto hope as long as possible, too.

 

[BlueBlood66]

I understand how the BPI and similar metrics work. I also realize it is August and once the games start there will be a lot of movement up and down the rankings as more data becomes available. But listing one of the two or three best teams in the country at 22nd, and listing a Marquette team that finished 7th in the Big East last year and lost their leading scorer at 4th is just stupid. There's no point in putting out a preseason ranking for a metric that needs relevant data to be accurate until said data is available.

There’s no point in putting ranking metrics out at all other than just something for fans to click on and discuss. Doesn’t matter if it’s pre-season or halfway through the season.

Us discussing it in a thread alone makes it justifiable to ESPN.

It’s an entertainment company like every other “sports” company.

 

[Newcattitudes]

OP you're so right. We really need to take a step back and listen to the experts. They definitely know what they're talking about

Same media that picked Hillary?

 

[Blueisbest]

If that was the 22nd best team in the country I watched last week then, Good God, how good are the other 21?

 

[Aike]

BPI is based on a formula. The godfather of college basketball analytics, Ken Pomeroy (of KenPom.com fame), has flat-out stated that even his projections are inherently flawed in the early going due to lack of good data. Same here. In UK's case, it's a young team with little returning production and a lot of projection. Of course UK will look bad in BPI in August.

Crapping on BPI for it's early season projections of a hard to project team is pretty dumb. You're either ignorant as to how it works, or harping on a known weakness of it (and similar) systems.

If a poll of human voters can figure out that UK is top 5 or better, then someone can build an algorithm that comes to a similar conclusion.

Crapping on BPI for getting its projections so obviously wrong is the opposite of dumb. Why would anyone give credit to a computer projection that is less accurate than a monkey throwing darts?

 

[Big_Blue79]

If a poll of human voters can figure out that UK is top 5 or better, then someone can build an algorithm that comes to a similar conclusion.

Crapping on BPI for getting its projections so obviously wrong is the opposite of dumb. Why would anyone give credit to a computer projection that is less accurate than a monkey throwing darts?

Because it's more accurate later in the year with more and better data, which is inherent in its design and obvious to anyone who is able to follow a hyperlink or two (or use common sense). Don't be dumb, Aike. You're better than that.

 

[Big_Blue79]

There's no point in putting out a preseason ranking for a metric that needs relevant data to be accurate until said data is available.

People click on the link, which generates ad revenue. Then outraged people like OP post about it to places like this, which induces other to click on the link, which generates ad revenue. So there is absolutely a point in putting it out at this time of year. It's just not the one you want.

 

[IL Wildcat]

Espn .. is that the network that used to do sports ?

Yes, the same.

Now they're attempting to wow us with personality-driven talk-shows. And they wonder why they've lost their base of support.

 

[Aike]

Because it's more accurate later in the year with more and better data, which is inherent in its design and obvious to anyone who is able to follow a hyperlink or two (or use common sense). Don't be dumb, Aike. You're better than that.

What’s the purpose of using an inaccurate model for preseason rankings? Surely they could have changed the weighting of certain factors to produce something more precise. Maybe weighted five star recruits on some sort of sliding scale related to expected playing time?

Again, any half moron with 2 eyeballs can tell this team is much better than number 22 in the country. It isn’t as simple as the model getting better with more data. This model is specifically a preseason one. They intentionally chose to underweight incoming talent, fully knowing what that would do to Kentucky’s ranking.

So pardon me for calling a spade a spade. As a predictive model, which this is clearly intended to be, it is lacking.

Of course it was very likely mis-weighted for clickbait reasons. The E does stand for Entertainment, after all.

 

[ORCAT]

ESPN's BPI has us ranked 22nd going into next season. The team may have looked good in the Bahamas and they may be oozing national title potential, but don't forget that according to ESPN's BPI metric we will have to hold off teams like TCU, Northwestern, St. John's and Butler while trying to leap frog the powerhouses in front of us like Illinois, Alabama, Iowa, Mississippi St, Virginia Tech, Kansas St, Syracuse, West Virginia, and Marquette. Hopefully by January we can break into the top 15.

Great post, love this one, filled with sarcasm. Great job. VikingCat21.

 

[KYFOSSIL]

ESPN's BPI has us ranked 22nd going into next season. The team may have looked good in the Bahamas and they may be oozing national title potential, but don't forget that according to ESPN's BPI metric we will have to hold off teams like TCU, Northwestern, St. John's and Butler while trying to leap frog the powerhouses in front of us like Illinois, Alabama, Iowa, Mississippi St, Virginia Tech, Kansas St, Syracuse, West Virginia, and Marquette. Hopefully by January we can break into the top 15.

I Love the Irony when some says “pump” the breaks” and has an avatar from Smoky and the Bandit

 

[Big_Blue79]

What’s the purpose of using an inaccurate model for preseason rankings? Surely they could have changed the weighting of certain factors to produce something more precise. Maybe weighted five star recruits on some sort of sliding scale related to expected playing time?

Again, any half moron with 2 eyeballs can tell this team is much better than number 22 in the country. It isn’t as simple as the model getting better with more data. This model is specifically a preseason one. They intentionally chose to underweight incoming talent, fully knowing what that would do to Kentucky’s ranking.

So pardon me for calling a spade a spade. As a predictive model, which this is clearly intended to be, it is lacking.

Of course it was very likely mis-weighted for clickbait reasons. The E does stand for Entertainment, after all.

The model is not specifically a preseason one (it's a power ranking system like human polls, RPI, Sangarin, and KenPom), but the article is (it also it titled "Why does BPI rate Marquette higher than Duke or Kentucky?", which means it actually answers your gripes). If you can't figure out the difference, then I'll have to retract my statement that you're better than to fall for bad arguments. The model updates throughout the season based on more and better data, and is used throughout the season, primarily around the NCAA Tournament (who should be in, seeding, predictions, toughest bracket, etc...). And just because a model has flaws does not mean the whole thing should be thrown out, but I suspect now that is an argument above your head so why even bother?

Also are you seriously trying to say that ESPN created BPI with the specific intent of damaging Kentucky's preseason ranking? Or purposefully "mis-weighted" it to harm Kentucky (or Duke, I guess)? That's one of the dumber things I've read on RR in a long time.

 

[Aike]

The model is not specifically a preseason one (it's a power ranking system like human polls, RPI, Sangarin, and KenPom), but the article is (it also it titled "Why does BPI rate Marquette higher than Duke or Kentucky?", which means it actually answers your gripes). If you can't figure out the difference, then I'll have to retract my statement that you're better than to fall for bad arguments. The model updates throughout the season based on more and better data, and is used throughout the season, primarily around the NCAA Tournament (who should be in, seeding, predictions, toughest bracket, etc...). And just because a model has flaws does not mean the whole thing should be thrown out, but I suspect now that is an argument above your head so why even bother?

Also are you seriously trying to say that ESPN created BPI with the specific intent of damaging Kentucky's preseason ranking? Or purposefully "mis-weighted" it to harm Kentucky (or Duke, I guess)? That's one of the dumber things I've read on RR in a long time.

You really are a piece of...work.

This was a model built for preseason rankings. That’s why it includes details such as number of 5 star recruits. So it wasn’t “inherent in the design” of this model to get better throughout the season. This model was built for predictive purposes. For the preseason.

Do you really think they will include info about 5 star recruits by March? Of course not. The model in March will be a different model built on different principles. And obviously different data.

I’m not privy to the discussions at ESPN closed door meetings. I’m sure the analytics team pounding out this model had hundreds, if not thousands of tweaks to choose from. Maybe they really think they got it right with the one they trotted out. I would imagine that the marketing guys at their website were perfectly comfortable with a model that nearly left Kentucky out of the top 25. Almost sounds like a clickbait headline.

 

[jackcarlson]

the only thing I find to worry about with this team is: how is Cal going to find minutes for everyone who deserves them? He may start off by telling some of these guys that he fully expects them to be here 2 or 3 years. Obviously doesn't apply to Travis, but could apply to all the guards and definitely Richards.

 

[Bkocats]

the only thing I find to worry about with this team is: how is Cal going to find minutes for everyone who deserves them? He may start off by telling some of these guys that he fully expects them to be here 2 or 3 years. Obviously doesn't apply to Travis, but could apply to all the guards and definitely Richards.

wasn't a problem in '14-15
And I don't think it's a problem this year; these guys are hungry and Cal probably won't have to tighten the rotation by the end of the season
Just my opinion

 

[kev69]

Espn .. is that the network that used to do sports ?

ESPN has been a complete joke for years.

 

[morgousky]

What’s the purpose of using an inaccurate model for preseason rankings? Surely they could have changed the weighting of certain factors to produce something more precise. Maybe weighted five star recruits on some sort of sliding scale related to expected playing time?

Again, any half moron with 2 eyeballs can tell this team is much better than number 22 in the country. It isn’t as simple as the model getting better with more data. This model is specifically a preseason one. They intentionally chose to underweight incoming talent, fully knowing what that would do to Kentucky’s ranking.

So pardon me for calling a spade a spade. As a predictive model, which this is clearly intended to be, it is lacking.

Of course it was very likely mis-weighted for clickbait reasons. The E does stand for Entertainment, after all.

This word data......uugghhhh....

In my no no list of sports words right along with "anecdotal"

Wish sometimes people would just say Hey, team looks top 5 to me.

 

[Aike]

This word data......uugghhhh....

In my no no list of sports words right along with "anecdotal"

Wish sometimes people would just say Hey, team looks top 5 to me.

Data is here to stay. But common sense should still apply.

 

[Bluegrassking]

the only thing I find to worry about with this team is: how is Cal going to find minutes for everyone who deserves them? He may start off by telling some of these guys that he fully expects them to be here 2 or 3 years. Obviously doesn't apply to Travis, but could apply to all the guards and definitely Richards.

I find it difficult at the moment to imagine any if the bigs back. If he had eligibility, I'd say Travis would be the most likely.

Montgomery despite a super limited showing looks like the highest ceiling guy to me.

PJ seemingly walked the edge of the knife to come back and if he is consistent and builds at a normal progression then it will be damn difficult to keep him out of the first round.

Richards has always been not being a tomato can away from top 20 and probable lottery status with his size, consistency from at least 15 feet, and agility to allow him to switch on smaller players.

Of course Reid is a Graduate transfer.

The options seem to me are Richards being no help or Montgomery being gimpy and falling behind too far to get the PT, which means his experience would be minimal.

I expect coach to keep them all fresh and looking good while riding the hot hand/dominating performance which hides the cold/bad performances. These guys are all four going to look very appealing if they bust their asses consistently.

To get more than one back would mean trouble like 98.999975% of the time.

A couple or three guards may be possible though.

 

[HICATFAN]

I love all of these discussions.... good stuff!!!

But one thing that still is so salient ITS AUGUST!!!

I can’t wait to see what Coach Cal, KP and the other coaching (strength) staff is going to do with this team... it could get “unreal”....

AlohaCat

 

[BourbonBalz]

The model is not specifically a preseason one (it's a power ranking system like human polls, RPI, Sangarin, and KenPom), but the article is (it also it titled "Why does BPI rate Marquette higher than Duke or Kentucky?", which means it actually answers your gripes). If you can't figure out the difference, then I'll have to retract my statement that you're better than to fall for bad arguments. The model updates throughout the season based on more and better data, and is used throughout the season, primarily around the NCAA Tournament (who should be in, seeding, predictions, toughest bracket, etc...). And just because a model has flaws does not mean the whole thing should be thrown out, but I suspect now that is an argument above your head so why even bother?

Also are you seriously trying to say that ESPN created BPI with the specific intent of damaging Kentucky's preseason ranking? Or purposefully "mis-weighted" it to harm Kentucky (or Duke, I guess)? That's one of the dumber things I've read on RR in a long time.

So, the bottom line is you’re just a condescending asshole?