Jump to content

Ken Pomeroy's Ranking of D1 Programs


IUFLA

Recommended Posts

1 minute ago, IndyResident16 said:

See this is where you look silly. That computer geeks formula is more accurate than anything your eyes have ever seen. There’s no bias in any of his formula’s, it’s strictly hard numbers.   

I guarantee you I watch enough college basketball throughout the year that I can pick most of the field without any computer rankings.  I don't care about hard numbers and analytics is taking away peoples ability  to actually watch and make their own decisions.

Link to comment
Share on other sites

  • Replies 158
  • Created
  • Last Reply
3 minutes ago, IndyResident16 said:

Your untrained eye has many bias’ that a computer does not. That’s why computer rankings are entirely more accurate, there’s no bias in numbers.

I would rather have human beings pick the teams because the game is played by and watch by humans.  You probably are the person who thinks it will be great to have a robot calling ball and strikes in baseball.  Sports are suppose to be fun and entertaining but making it so scientific just takes the whole fun away from it.  The joy I get about talking about sports comes from actually talking to people who have their own thoughts and not take it from what a computer tells you what to think.

Link to comment
Share on other sites

4 minutes ago, milehiiu said:

Does it take into consideration, injuries ?   Like what IU went through last year ?

No. Again it’s a measure of predictor. KenPom doesn’t rank teams based on how  they perform rather how they are expected to perform based on efficiency. If you don’t understand how it works it’s useless to have this discussion.

Link to comment
Share on other sites

8 minutes ago, milehiiu said:

How is it that you and I are seeing more and more together on things related to College Basketball ? I agree. And as I posted previously.  I will rely on my untrained eye over a computer.

Because we grew up in a time where we actually watch the game and learned the game by watching and not by what stats told us.

Link to comment
Share on other sites

6 minutes ago, IU Scott said:

I would rather have human beings pick the teams because the game is played by and watch by humans.  You probably are the person who thinks it will be great to have a robot calling ball and strikes in baseball.  Sports are suppose to be fun and entertaining but making it so scientific just takes the whole fun away from it.  The joy I get about talking about sports comes from actually talking to people who have their own thoughts and not take it from what a computer tells you what to think.

Not at all. Why do you think Vegas always wins? Because a lot of people who think like you, think they know more “than a bunch of computer geeks” who set lines strictly based on numbers, trends, and prior results. 

Link to comment
Share on other sites

2 minutes ago, IU Scott said:

Because we grew up in a time where we actually watch the game and learned the game by watching and not by what stats told us.

I was born in the 70s and didn’t know or start to care about analytics until 10-15 years ago. So don’t act like I’m some 20 year old who doesn’t know what I’m talking about. Analytics are used widespread across all levels of sports. To ignore this is burying your head in the sand.

Link to comment
Share on other sites

17 minutes ago, IU Scott said:

I guarantee you I watch enough college basketball throughout the year that I can pick most of the field without any computer rankings.  I don't care about hard numbers and analytics is taking away peoples ability  to actually watch and make their own decisions.

Bet you wouldn’t come as close to accurate as KenPom. AND FOR THE LAST TIME, KenPom does NOT measure success. It’s a predictor of success. You don’t even know what you’re arguing at this point, just dismissing analytics because you don’t understand it. 

Link to comment
Share on other sites

47 minutes ago, IndyResident16 said:

No. Again it’s a measure of predictor. KenPom doesn’t rank teams based on how  they perform rather how they are expected to perform based on efficiency. If you don’t understand how it works it’s useless to have this discussion.

I've been following this thread for that last couple of hours or so. Most of us don't try to make a living from Vegas. We grew up watching and learning the game so we see it through a human set of eyes. Analytics is great for those that want to win a few bucks or have a different spin on discussions about the game. The use in Fantasy Leagues as well. But for most, we just enjoy looking at it from an old-fashioned POV.

Link to comment
Share on other sites

Fascinating conversation. 

Sorry @milehiiu and @IU Scott, but you guys are off base on this one.  If anything, as others have said, IU is ranked too high.  The article clearly states that the last 25 years are the basis for the ranking.  And, you can't selectively play the 'history' card.  You can't say we're a better program than so and so because of our championships, without also saying the Holy Cross is an equivalent program with Virginia.  After all, those two programs have the same number of championships in their history.  Do you agree they are equals?  I doubt you do, which means you are letting bias influence your opinion when it comes to IU, which is something computer numbers do not do.

Looking at it another way.  @IU Scott , I know you, like me, are a big Reds fan.  Do you think the Reds are better than the Dodgers?  After all, Cincy has won a World Series since the Dodgers have.  But, really, is there any comparison to the quality of the two teams in recent MLB history?

Link to comment
Share on other sites

8 minutes ago, 5fouls said:

Fascinating conversation. 

Sorry @milehiiu and @IU Scott, but you guys are off base on this one.  If anything, as others have said, IU is ranked too high.  The article clearly states that the last 25 years are the basis for the ranking.  And, you can't selectively play the 'history' card.  You can't say we're a better program than so and so because of our championships, without also saying the Holy Cross is an equivalent program with Virginia.  After all, those two programs have the same number of championships in their history.  Do you agree they are equals?  I doubt you do, which means you are letting bias influence your opinion when it comes to IU, which is something computer numbers do not do.

Looking at it another way.  @IU Scott , I know you, like me, are a big Reds fan.  Do you think the Reds are better than the Dodgers?  After all, Cincy has won a World Series since the Dodgers have.  But, really, is there any comparison to the quality of the two teams in recent MLB history?

There are several dimensions to determining how good a program is. KenPom uses actual basketball game stats to derive his rankings. That is fine if that is all that interests you. But I can rest assure you that if you use other parameters such as attendance,  profit, tv ratings etc. IU ranks in the top 5 of programs over the last 25 yrs. Add in history, tradition...well, who needs KenPom.

Link to comment
Share on other sites

1 hour ago, IndyResident16 said:

I was born in the 70s and didn’t know or start to care about analytics until 10-15 years ago. So don’t act like I’m some 20 year old who doesn’t know what I’m talking about. Analytics are used widespread across all levels of sports. To ignore this is burying your head in the sand.

For you they might be great but I have no use for them.  I don't have to have about 25 stats to tell me why my team won or lost a game when I can watch it with my own two eyes.

Link to comment
Share on other sites

1 hour ago, IndyResident16 said:

Bet you wouldn’t come as close to accurate as KenPom. AND FOR THE LAST TIME, KenPom does NOT measure success. It’s a predictor of success. You don’t even know what you’re arguing at this point, just dismissing analytics because you don’t understand it. 

I don't have a need to understand them because they are useless to me.  What I hate about people who like analytics is that those people think they are so much smarter about the game but I don't think you are.  You might know the stats and what they tell you but do you really no more about the actual game of basketball because you can look up stats.  It is like that Brian Kenny on the MLB network who thinks he invented the game because he uses analytics and will argue with guys who actually managed or played the game and thinks he knows more than they do.

Link to comment
Share on other sites

37 minutes ago, Moyeneeded said:

I've been following this thread for that last couple of hours or so. Most of us don't try to make a living from Vegas. We grew up watching and learning the game so we see it through a human set of eyes. Analytics is great for those that want to win a few bucks or have a different spin on discussions about the game. The use in Fantasy Leagues as well. But for most, we just enjoy looking at it from an old-fashioned POV.

thank you

Link to comment
Share on other sites

24 minutes ago, 5fouls said:

Fascinating conversation. 

Sorry @milehiiu and @IU Scott, but you guys are off base on this one.  If anything, as others have said, IU is ranked too high.  The article clearly states that the last 25 years are the basis for the ranking.  And, you can't selectively play the 'history' card.  You can't say we're a better program than so and so because of our championships, without also saying the Holy Cross is an equivalent program with Virginia.  After all, those two programs have the same number of championships in their history.  Do you agree they are equals?  I doubt you do, which means you are letting bias influence your opinion when it comes to IU, which is something computer numbers do not do.

Looking at it another way.  @IU Scott , I know you, like me, are a big Reds fan.  Do you think the Reds are better than the Dodgers?  After all, Cincy has won a World Series since the Dodgers have.  But, really, is there any comparison to the quality of the two teams in recent MLB history?

I am not saying that we should be higher but I am just arguing about analytics as a whole and not just this article

Link to comment
Share on other sites

24 minutes ago, IU Scott said:

I don't have a need to understand them because they are useless to me.  What I hate about people who like analytics is that those people think they are so much smarter about the game but I don't think you are.  You might know the stats and what they tell you but do you really no more about the actual game of basketball because you can look up stats.  It is like that Brian Kenny on the MLB network who thinks he invented the game because he uses analytics and will argue with guys who actually managed or played the game and thinks he knows more than they do.

They’re useless to you because you don’t understand them. There isn’t a program or franchise at either the collegiate or professional level who doesn’t incorporate analytics into scouting, game planning, drafting, etc. Coaches/managers at all levels use analytics in their daily coaching. From lineups, to spray charts, to shifts, to matchups, analytics are incorporated into the game and you’re lying to yourself if you think otherwise. Saying “I don’t need to understand analytics because they are useless to me” is the same as saying “I’m dense and don’t understand them”. It’s no different than when my father used to tell me texting was useless, until I actually showed him how to text and he’s never stopped since.

I’ve never pretended like I know more about the game than others simply because I understand how analytics work. Have I suggested otherwise? I was a 3 sport athlete in high school and didn’t have a single clue about analytics or advanced stats, didn’t know they existed. But incorporating analytics from what I previously knew has changed my perspective. If you actually follow KenPom’s model, he’s a lot more accurate than the guys who continually say “I know the game better after watching with my own eyes for X and X years”.

Link to comment
Share on other sites

32 minutes ago, Moyeneeded said:

There are several dimensions to determining how good a program is. KenPom uses actual basketball game stats to derive his rankings. That is fine if that is all that interests you. But I can rest assure you that if you use other parameters such as attendance,  profit, tv ratings etc. IU ranks in the top 5 of programs over the last 25 yrs. Add in history, tradition...well, who needs KenPom.

Profit, TV ratings, attendance has absolutely ZERO to do with on-court success or predictive measures of success. Again, you guys are wanting this KenPom article to be something it isn’t. Read the article. The article isn’t designed to rank teams based on historical success, TV rankings, etc, it’s simply a predictive measure of success on how teams should fare based on the data he’s been compiling since 1997. Had he been collecting data since 1967 then IU is likely ranked much higher.

Link to comment
Share on other sites

49 minutes ago, IU Scott said:

For you they might be great but I have no use for them.  I don't have to have about 25 stats to tell me why my team won or lost a game when I can watch it with my own two eyes.

Again, you don’t understand KenPom. KenPom doesn’t tell you why you won or lost the game. KenPom tells you who should win and by how much based on different levels of efficiency. I don’t think you understand what predictive means. 

Link to comment
Share on other sites

Apologies in advance if this comes off harsh or as a manifesto.

1. Why start at 1997? Why does traditional Hawaiian food feature a lot of pork, fish and pineapple? The answer is the same: because that's what's available. Arbitrarily dismissing this because of the 1997 start date is like criticizing Hawaiian food for not featuring more caribou. Pomeroy has been compiling data since that point. Given enough time and resources, he could probably go back further, but he hasn't yet.

2. Why does Pomeroy base his ratings through January of a given year on a blend of the prior season's data and the current season's data? Sample size. If I send out a survey to 10 people, and 7 answer a certain way, is it fair to extrapolate that data and say that 70% of the residents of that state would answer that way? Not really. Let's say I was conducting such a survey, and in the state of Indiana, I could only send out those 10 surveys. I could use the data I have, which I know isn't likely to be accurate, or I could integrate data that is similar. In my example, that means I could conduct surveys in Illinois or Ohio. Is that still going to be an exact predictor of Indiana? No, but it's the next best thing, and it's going to produce results that are more sound than a mere 10 examples. Oh, by the way, the survey is about rock salt preference in the winter, and 5 out of the 10 residents are recent immigrants from Qatar and only speak and write in Urdu. That's how accurate of a predictor a lot of the cupcake non conference games are.

If Pomeroy were to only base results on the current season, you'd get anomalies like the example above. Is using past data blended with current perfect? No, but it's slightly better than not.

3. A rose by any other name does smell sweeter. There's a huge difference in the AP/Coaches "rankings" and a statistical ranking like the KenPom metrics. The subjective polls are intended to represent a moment in time: that one particular week. These are the top teams for this particular week. Kenpom's metrics aren't meant to capture a moment in time. They're meant to say that IF current and historic trends hold, at the end of the season, this will be the likely outcome. Predictive vs moment in time.

Now, to answer the question about injuries, yes, they are accounted for. A team's statistics are the sum of its players performances. If a team loses a key player, their aggregate offensive and defensive stats will begin to decline. This will start to decrease their rank accordingly. The same would hold true in a subjective poll. If a team has an injury, but keeps winning, won't their subjective rating improve or hold steady?

4. Indiana isnt the group that should have a beef with these team rankings. Sure, we have 5 titles, but all were before the tabulated period. UConn has FOUR titles within the tabulated period and they're BEHIND US. Anybody consider that?

 

Link to comment
Share on other sites

I find that my brain does the same thing the statistics are telling me. I know that our team doesn’t shoot the three well and that perhaps the other team doesn’t give up a lot of points in the paint or they get a lot of points in transition and we turn the ball over at a high rate all those things may add up to say give me a good indication we may struggle against them. Does it mean we won’t win? Heck no. We may get streaky or we may take care of the ball and get back on D. Heck the other team may get food poisoning or their best player may get in foul trouble or break up with his girlfriend and the game may go differently then the analytics would suggest. It’s why college basketball is so fascinating...a lot more volitility with the variables. In the NBA where there are professionals they eliminate a lot of the outside factors and the players generally perform a lot more consistently so I think it’s more precise but in the end I think I can look at teams and I can tell you what the numbers are probably going to say about them. Sometimes I’m surprised...sometimes the schedule early on can hide flaws etc but usually with about 10 games to go in a season I think I can tell you the numbers and I think usually the numbers will back up what I think about a team. Not always but ones I’ve watched often enough I feel pretty confident about. I’d say overall the analytics usually line up pretty close to the eye test. I’d say even the people that don’t buy into the analytics don’t realize their eyes take in the info and their brain calculates it and they formulate their opinion based on the statistics they see. Problem comes in when they don’t retain accurately what they are seeing and are misrepresenting the numbers in their opinions. The paper doesn’t lie so time to time i do like to match up what I think I saw with the numbers.

Link to comment
Share on other sites

1 hour ago, IU Scott said:

For you they might be great but I have no use for them.  I don't have to have about 25 stats to tell me why my team won or lost a game when I can watch it with my own two eyes.

Picture perfect illustration.

Using statistics isn't about explaining why a particular game was won or lost. It's aggregating each of those games and using those outcomes to predict future outcomes.

And I'm sure you'll deny this up one side and down the other, but "eyeball" factors (aka anomalies, much of the time) do factor in to statistical models.

Case in point: I build models that use past data to predict a given community's future energy demand. Demand in Dayton Ohio in July is going to be lower than normal. Why? Because tornados knocked out a big industrial section, which are heavy power consumers. I can't use statistics to explain this. In fact, I'll remove it from future models, or at least try to normalize it, as it represents an anomaly, not the trend.

 

Link to comment
Share on other sites

5 hours ago, IndyResident16 said:

You wouldn’t have the slightest clue as how to configure his ratings. Hate analytics all you want, they’re a part of the game and how it’s played and coached and more importantly factors into NCAA seeding and inclusion.

Ok your posts have been getting personal and offensive. Mile  doesn’t appreciate the article. Leave it there. 

Link to comment
Share on other sites

11 hours ago, IndyResident16 said:

They’re useless to you because you don’t understand them. There isn’t a program or franchise at either the collegiate or professional level who doesn’t incorporate analytics into scouting, game planning, drafting, etc. Coaches/managers at all levels use analytics in their daily coaching. From lineups, to spray charts, to shifts, to matchups, analytics are incorporated into the game and you’re lying to yourself if you think otherwise. Saying “I don’t need to understand analytics because they are useless to me” is the same as saying “I’m dense and don’t understand them”. It’s no different than when my father used to tell me texting was useless, until I actually showed him how to text and he’s never stopped since.

I’ve never pretended like I know more about the game than others simply because I understand how analytics work. Have I suggested otherwise? I was a 3 sport athlete in high school and didn’t have a single clue about analytics or advanced stats, didn’t know they existed. But incorporating analytics from what I previously knew has changed my perspective. If you actually follow KenPom’s model, he’s a lot more accurate than the guys who continually say “I know the game better after watching with my own eyes for X and X years”.

That is fine if team use them but I am talking about a fan perspective and I can enjoy the game without having to know every little detail of the stats.  I don't need 25 different stats to tell me why we won a game or why we lost the game because I cant know that by just watching the game.

 

Without even reading the article about the top programs I bet I can come close just off the top of my head who were the programs that were ahead of us.  If I got it straight that they had 17 programs ahead of us and 4 in the big ten so here is my stab at this.

MSU, UW, OSU, UM

UCONN, Villanova, Gonzaga, UVA

UNC, Duke, UL, Syracuse

Florida, UK

Kansas

Arizona, UCLA

Link to comment
Share on other sites

9 hours ago, Zlinedavid said:

Picture perfect illustration.

Using statistics isn't about explaining why a particular game was won or lost. It's aggregating each of those games and using those outcomes to predict future outcomes.

And I'm sure you'll deny this up one side and down the other, but "eyeball" factors (aka anomalies, much of the time) do factor in to statistical models.

Case in point: I build models that use past data to predict a given community's future energy demand. Demand in Dayton Ohio in July is going to be lower than normal. Why? Because tornados knocked out a big industrial section, which are heavy power consumers. I can't use statistics to explain this. In fact, I'll remove it from future models, or at least try to normalize it, as it represents an anomaly, not the trend.

 

Again as a basketball fan I couldn't care less about who or what the computer thinks will happen.  I just watch basketball for entertainment and my love for the game of basketball and not stats.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.




×
×
  • Create New...