So according to KenPom, we are the "luckiest" P6 team in basketball

[SpiderRico]

...and last year, we were the unluckiest.

I'm not sure what to make of that at this point. Based on what I understand about "luck" in KenPom's rankings, it says you are "luckier" the more the actual result deviates from the expected result.

But if we were super unlucky last year, does his model adjust for that this year given the fact that we returned pretty much everyone....when in reality, it's just things evening out towards the mean?

Not sure it means much, but it is interesting to root for a team that has been on the extreme outliers like that in the last 2 seasons....

 

[tksirius]

Is the expected result based on Vegas or something else?

 

[SteveGarvey1]

...and last year, we were the unluckiest.

I'm not sure what to make of that at this point. Based on what I understand about "luck" in KenPom's rankings, it says you are "luckier" the more the actual result deviates from the expected result.

But if we were super unlucky last year, does his model adjust for that this year given the fact that we returned pretty much everyone....when in reality, it's just things evening out towards the mean?

Not sure it means much, but it is interesting to root for a team that has been on the extreme outliers like that in the last 2 seasons....

We were lucky to beat Pitt and Northwestern. Otherwise, I wouldn't consider any of our other wins "lucky".

 

[CP87]

Is the expected result based on Vegas or something else?

No, I think it is based upon point-differential.

In that sense, "lucky" really comes down to winning close games. This is where the arguments from sport-folk vs. analytics-folk arise. The coaches, former-players, etc. swear that there is a "clutch gene" that some players/teams have that let them win those close ones over and over.

The analytics people think "clutch" is a mirage that our brains create, and that players/teams are what thousands of minutes of playing time says they are, and they cannot somehow become different just because a game is on the line.

The answer is probably somewhere in the middle. Anxiety increases in "clutch" situations, and some players/teams handle that better than others. But people also take this to ridiculous extremes, claiming things like LeBron James is a choke artist (that has mostly died away, but it was really prominent for awhile) even though his statistical production in close games was otherworldly. But he missed an important shot here or there, thus he is a choker (and all of the shots missed by true winners like MJ or Kobe are never brought up).

 

[SayHawkKid]

...and last year, we were the unluckiest.

I'm not sure what to make of that at this point. Based on what I understand about "luck" in KenPom's rankings, it says you are "luckier" the more the actual result deviates from the expected result.

But if we were super unlucky last year, does his model adjust for that this year given the fact that we returned pretty much everyone....when in reality, it's just things evening out towards the mean?

Not sure it means much, but it is interesting to root for a team that has been on the extreme outliers like that in the last 2 seasons....

Buffalo (KenPom # 24) plays in the MAC and has lost 3 games this year to teams with an average KenPom rating of 85. Iowa (KenPom # 27) plays in the B1G and has lost 5 games to teams with an average KenPom rating of 16.

So you can see it is all very logical and easy to understand.

 

[PCHawk]

No, I think it is based upon point-differential.

In that sense, "lucky" really comes down to winning close games. This is where the arguments from sport-folk vs. analytics-folk arise. The coaches, former-players, etc. swear that there is a "clutch gene" that some players/teams have that let them win those close ones over and over.

The analytics people think "clutch" is a mirage that our brains create, and that players/teams are what thousands of minutes of playing time says they are, and they cannot somehow become different just because a game is on the line.

The answer is probably somewhere in the middle. Anxiety increases in "clutch" situations, and some players/teams handle that better than others. But people also take this to ridiculous extremes, claiming things like LeBron James is a choke artist (that has mostly died away, but it was really prominent for awhile) even though his statistical production in close games was otherworldly. But he missed an important shot here or there, thus he is a choker (and all of the shots missed by true winners like MJ or Kobe are never brought up).

There are players that get anxious in close games and have the "anti clutch gene". Then there are players that get an overwhelming sense of confidence when the game is on the line. Those are the 9nes who have the clutch gene. Most players fall somewhere in the middle. I'm not sure how anyone who knows anything about sports or humans can argue that. What exactly are they saying? That ever single person handles pressure the exact same way? Or that the odds of making a shot that you throw up in a panic is the same as a shot you shoot with extreme confidence?

Bohannon said "I knew it was going to go in before it left my hand". That doesn't mean he had 100% chance of making it. It just means a really good shooter shot it with extreme confidence. The odds were pretty good.

 

[BryceC]

I'm not sure what to make of that at this point. Based on what I understand about "luck" in KenPom's rankings, it says you are "luckier" the more the actual result deviates from the expected result.

Correct.

But if we were super unlucky last year, does his model adjust for that this year given the fact that we returned pretty much everyone....when in reality, it's just things evening out towards the mean?

Nope, last year has no bearing on this year.

"What do all the columns mean?

The new ones are Cons (Consistency) and Luck. The easiest one to understand is Luck, which is the deviation in winning percentage between a team’s actual record and their expected record using the correlated gaussian method. The luck factor has nothing to do with the rating calculation, but a team that is very lucky (positive numbers) will tend to be rated lower by my system than their record would suggest."

 

[ArvadaHawk]

There are players that get anxious in close games and have the "anti clutch gene". Then there are players that get an overwhelming sense of confidence when the game is on the line. Those are the 9nes who have the clutch gene. Most players fall somewhere in the middle. I'm not sure how anyone who knows anything about sports or humans can argue that. What exactly are they saying? That ever single person handles pressure the exact same way? Or that the odds of making a shot that you throw up in a panic is the same as a shot you shoot with extreme confidence?

Bohannon said "I knew it was going to go in before it left my hand". That doesn't mean he had 100% chance of making it. It just means a really good shooter shot it with extreme confidence. The odds were pretty good.

ANYONE who has played sports, at least through the high school level, understands there is a "clutch" gene. like PC said, it's a combination of confidence, calm, competitiveness, talent, desire, physical attributes and (sport) IQ. Anyone think Jordan wasn't clutch? Ask the Cleveland Cavaliers and Craig Ehlo. Anyone think Reggie Jackson wasn't clutch in October? Extra credit to those that can identify what Reggie Jackson's nickname was.

 

[sioux34]

Im very anti baseball, and even I know the extra credit answer.

 

[CP87]

There are players that get anxious in close games and have the "anti clutch gene". Then there are players that get an overwhelming sense of confidence when the game is on the line. Those are the 9nes who have the clutch gene. Most players fall somewhere in the middle. I'm not sure how anyone who knows anything about sports or humans can argue that. What exactly are they saying? That ever single person handles pressure the exact same way? Or that the odds of making a shot that you throw up in a panic is the same as a shot you shoot with extreme confidence?

Bohannon said "I knew it was going to go in before it left my hand". That doesn't mean he had 100% chance of making it. It just means a really good shooter shot it with extreme confidence. The odds were pretty good.

I think it is mostly based upon evidence that a team's records in close games seems to be pretty random from year to year. If team A wins 80% of their close games one year while team B wins 20% of their close games, and they both bring back the same players and coaches, team A should also win more close games the next year if they were truly "clutch." I don't think the evidence really bares that out. Over time, almost all teams are 50/50 over a large enough sample of close games.

There are outliers, however. One that comes to mind is the Memphis Grizzlies for about 5+ years during their grit-and-grind glory days, they were way better than expected in close games. I think the T-Wolves had a similar stretch around the same time where they were leaning on young players and they were way worse than normal in close games. Who knows, "Lucky" might really mean you have a bunch of experienced guys who have been there before and are not overwhelmed by the moment.

If you want to dive into some statistical evidence refuting the existence of "clutch hitting" in baseball, here you go:

http://research.sabr.org/journals/the-statistical-mirage-of-clutch-hitting

 

[Fryowa]

KenPom's overall ranking is probably the most accurate and objective rating out there. If the exact formulas were made publicly available (which I'm well aware wouldn't happen), I'm fine with using it or something like it to determine the field.

Teams 1-64 go to the dance. 1-4 get 1 seeds, 5-8 get 2 seeds, and so on. Pretty tough to argue about not making it if you're arguing about being the 65th best team in the country.

People would bitch about Conference champ auto bids...fine...

Give the conference champs the auto bids still, and use a KenPom-esque rating system for the next 32 teams, using the ranking to seed all 64 teams.

 

[trj]

No, I think it is based upon point-differential.

In that sense, "lucky" really comes down to winning close games. This is where the arguments from sport-folk vs. analytics-folk arise. The coaches, former-players, etc. swear that there is a "clutch gene" that some players/teams have that let them win those close ones over and over.

The analytics people think "clutch" is a mirage that our brains create, and that players/teams are what thousands of minutes of playing time says they are, and they cannot somehow become different just because a game is on the line.

The answer is probably somewhere in the middle. Anxiety increases in "clutch" situations, and some players/teams handle that better than others. But people also take this to ridiculous extremes, claiming things like LeBron James is a choke artist (that has mostly died away, but it was really prominent for awhile) even though his statistical production in close games was otherworldly. But he missed an important shot here or there, thus he is a choker (and all of the shots missed by true winners like MJ or Kobe are never brought up).

LeBron being a choke artist basically stemmed from the finals against Dallas. A series where he didn't play well. More times than not, in "clutch" situations, LJ is making the right basketball play.

 

[JG10Hawk]

Buffalo (KenPom # 24) plays in the MAC and has lost 3 games this year to teams with an average KenPom rating of 85. Iowa (KenPom # 27) plays in the B1G and has lost 5 games to teams with an average KenPom rating of 16.

So you can see it is all very logical and easy to understand.

The luck part may not be all that logical but his ranking system is.

Kenpom's rankings are based off net adjusted efficiency or what he calls "adjusted efficiency margin". This is not directly based on whether a team wins or loses. The adjusted part is him factoring in the strength of opponent when calculating offensive and defensive efficiency (estimated points scored / points given up per 100 possessions) which is then netted to give him the margin used to rank teams.

Edit: The reason Buffalo is ranked ahead of Iowa is simply their average efficiency margin is better, even when adjusted for opponent strength like Kenpom does.

 

[revkev73]

Iowa = Lucky (Instant Classic Oxymoron)

 

[revkev73]

Ken Harrelson "It isn't how many, but when, that separates the greats from average"

 

[CP87]

ANYONE who has played sports, at least through the high school level, understands there is a "clutch" gene. like PC said, it's a combination of confidence, calm, competitiveness, talent, desire, physical attributes and (sport) IQ. Anyone think Jordan wasn't clutch? Ask the Cleveland Cavaliers and Craig Ehlo. Anyone think Reggie Jackson wasn't clutch in October? Extra credit to those that can identify what Reggie Jackson's nickname was.

It is those same people that argue there is no such thing as "the hot hand." And evidence supports that, just because you have made 3 shots in a row, you don't have an increased likelihood of making that 4th shot (compared to missing 3 shots in a row). But all of these things are so hard to tease out; that "hot" shooter is actually more likely to attempt a more difficult 4th shot, so making a more difficult shot at the same rate might really indicate that he is "hot."

But then you get back to the people who have actually played BB, and they will all recall a time that they just couldn't miss for an extended period. Things like this are often too complex to be able to definitively prove, but that doesn't mean they don't exist. There is a middle ground where we can be informed by the stats but not let them entirely run the decision-making process. The human brain is fallible and susceptible to biases and all sorts of reasoning flaws, but it is still the most complex and nuanced computer available.

 

[CP87]

LeBron being a choke artist basically stemmed from the finals against Dallas. A series where he didn't play well. More times than not, in "clutch" situations, LJ is making the right basketball play.

And he did pretty much choke in that series (underperformed under pressure). But for several years thereafter people bent over backwards to find evidence to support their choker narrative (see Skip Bayless) even though that series was the outlier, not the norm.

 

[Robowe]

ANYONE who has played sports, at least through the high school level, understands there is a "clutch" gene. like PC said, it's a combination of confidence, calm, competitiveness, talent, desire, physical attributes and (sport) IQ. Anyone think Jordan wasn't clutch? Ask the Cleveland Cavaliers and Craig Ehlo. Anyone think Reggie Jackson wasn't clutch in October? Extra credit to those that can identify what Reggie Jackson's nickname was.

Reggie was, and always will be, Mr October. 1978 young Bob Welch gets him to end Game 2 (1st clip). However, Game 6, Reggie comes back to take him deep (2nd clip).

 

[Robowe]

No, I think it is based upon point-differential.

In that sense, "lucky" really comes down to winning close games. This is where the arguments from sport-folk vs. analytics-folk arise. The coaches, former-players, etc. swear that there is a "clutch gene" that some players/teams have that let them win those close ones over and over.

The analytics people think "clutch" is a mirage that our brains create, and that players/teams are what thousands of minutes of playing time says they are, and they cannot somehow become different just because a game is on the line.

The answer is probably somewhere in the middle. Anxiety increases in "clutch" situations, and some players/teams handle that better than others. But people also take this to ridiculous extremes, claiming things like LeBron James is a choke artist (that has mostly died away, but it was really prominent for awhile) even though his statistical production in close games was otherworldly. But he missed an important shot here or there, thus he is a choker (and all of the shots missed by true winners like MJ or Kobe are never brought up).

Here are some misses by MJ in the Finals. Now, the last one is really desperation, but he didn't make it every time. However, a couple rattled in ... and out.

 

[PCHawk]

It is those same people that argue there is no such thing as "the hot hand." And evidence supports that, just because you have made 3 shots in a row, you don't have an increased likelihood of making that 4th shot (compared to missing 3 shots in a row). But all of these things are so hard to tease out; that "hot" shooter is actually more likely to attempt a more difficult 4th shot, so making a more difficult shot at the same rate might really indicate that he is "hot."

But then you get back to the people who have actually played BB, and they will all recall a time that they just couldn't miss for an extended period. Things like this are often too complex to be able to definitively prove, but that doesn't mean they don't exist. There is a middle ground where we can be informed by the stats but not let them entirely run the decision-making process. The human brain is fallible and susceptible to biases and all sorts of reasoning flaws, but it is still the most complex and nuanced computer available.

Anyone who has played any sport at a decent level knows you can get hot and cold. When you're hot, you absolutely have a better chance of making the shot. Take pool for example. Some nights I can live up over a ball and have no idea if I will make the shot or not. Then sometimes I line up and it's like I can see a line drawn from the ball to the pocket. I know without a doubt I'm going to make it. My odds are maybe 50% on my bad nights. When I'm on, it's like 95%. Its not the odds playing themselves out. It's me being on fire.