343rd in Luck per Ken Pom

[Elonsmusk]

At this stage of the season MU is rated 343 in terms of luck.  Meaning we are the 21st unluckiest team out of 364.  Of course some of the Shaka skeptics will point toward toward his coaching, yet I'd consider Chuckie Hepburn going Markus Howard mode against us, and us being on the short end of a historically tilted FT disparity 49 to 19 both are indicative of bad "luck."

Shaka's a top 85-90 percentile coach.  There is room for improvement in some end of game situationals, and last night he should have used a timeout to stem the tide and calm the team in the last 5 minutes of the game.  Yet, being in a 30 free throw attempt deficit, you'd have to be a magician to win that type of game.

[CTWarrior]

We know why the rating says we are unlucky.  Our wins have been by large margins and our losses have been by close margins.

I don't think we were actually "unlucky".  We deserved each of those losses with poor play late.

[TAMU, Knower of Ball]

We know why the rating says we are unlucky.  Our wins have been by large margins and our losses have been by close margins.

Correct. The idea behind this stat is that things outside a team's control (said another way, luck) result in swinging the outcome by a few points in one direction in every game. Reality is that if we got one more bounce to go our way in each of our losses, we may be undefeated or at least have two less losses. Conversely, if we got one less bounce to go our way in each of our wins, we'd still have 9 wins. Shows have razor thin the margin can be.

I don't think we were actually "unlucky".  We deserved each of those losses with poor play late.

And if any of the victors lost those games, they would have deserved the loss for poor play at other parts of the game. Losing a lead late and failing to come all the way back from an early deficit both count the same. Also, the poor play late definitely applies to Purdue and Providence, but in Mississippi State and Wisconsin we were the ones who came back but couldn't finish at the end.

[The Equalizer]

At this stage of the season MU is rated 343 in terms of luck.  Meaning we are the 21st unluckiest team out of 364.  Of course some of the Shaka skeptics will point toward toward his coaching, yet I'd consider Chuckie Hepburn going Markus Howard mode against us, and us being on the short end of a historically tilted FT disparity 49 to 19 both are indicative of bad "luck."

Shaka's a top 85-90 percentile coach.  There is room for improvement in some end of game situationals, and last night he should have used a timeout to stem the tide and calm the team in the last 5 minutes of the game.  Yet, being in a 30 free throw attempt deficit, you'd have to be a magician to win that type of game.

You're reading too much into it.  "Luck" in the KenPom ratings is simply the term he uses to describe the deviation from his expectations. 

The unexpected closeness of our win against Chicago State probably factored more into this than Chucky Hepburn.





   

[Dr. Blackheart]

You're reading too much into it.  "Luck" in the KenPom ratings is simply the term he uses to describe the deviation from his expectations. 

The unexpected closeness of our win against Chicago State probably factored more into this than Chucky Hepburn.

And may be more of a statement of Pomeroy's model accuracy for MU than our luck.

[wisblue]

Pomeroy even says that he uses the word “luck” simply to represent the difference between actual W-L results and what might be expected from his model. It isn’t really “luck” in the same sense as the luck of something truly random like tossing a coin or rolling dice.

IMHO you could substitute the word “clutch” to identify teams that are good at pulling out tight games, not lucky to win them, and teams  that lose close games because they don’t execute well in tight situations, not because they are “unlucky”.

I think MU is more a case of being poor at executing down the stretch in close games than being “unlucky”.

[GoldenEagles03]

I'm a believer of "you create your own luck".

Lucky plays happen all the time, but only get noticed or analyzed when they happen late.  If you are constantly failing to pull away late then games will be close allowing either you or your opponent "lucky" opportunities near the end of games.

Up 8 with 5 minutes to go is a game a good team wins. Ideally you learn from a close loss and win the next one but Marquette is now 0-4 in close games 2 of which combined for 3 OTs.  They just aren't a good basketball team (right now) in close situations. If it doesn't get fixed they absolutely will not be a tournament team. If they can learn and grow and execute, then they might give themselves a shot.

[TAMU, Knower of Ball]

Pomeroy even says that he uses the word “luck” simply to represent the difference between actual W-L results and what might be expected from his model. It isn’t really “luck” in the same sense as the luck of something truly random like tossing a coin or rolling dice.

IMHO you could substitute the word “clutch” to identify teams that are good at pulling out tight games, not lucky to win them, and teams  that lose close games because they don’t execute well in tight situations, not because they are “unlucky”.

I think MU is more a case of being poor at executing down the stretch in close games than being “unlucky”.

How "clutch" a team is certainly matters. But there's also a reality that there are factors that play into an outcome of a basketball game that teams have no control over. The opponent's career 40% FT shooter goes 9/9 from the line? Nothing you can do about that. Guy makes a well defended 40 fade away 40 footer at the Buzzer? Nothing you can do about that. Ref misses more fouls/violations committed by opponent/makes more erroneous calls on you? Nothing you can do about that. There are even dozens more micro examples of things that impact the outcome of games.

All of these things that you can't really do anything about can impact the outcome of a game and you might call that "luck". Luck tends to even out over time but in individual games one team could be more "lucky" than the other. Good teams take luck out of the question by playing well enough that even a few more bad bounces on their side of the ledger doesn't lead to a loss.

People don't like talking about luck because they view it as an excuse. It doesn't excuse anything but it is a reality in sports. We are unlucky in the sense that in each of our losses, we only need one or two more breaks on our side to potential turn the Ls into Ws whereas in our wins, our opponents would have needed a ton of extra things to break their way in order to change the outcome. In the end, it doesn't matter. A "lucky" win is still a win, an "unlucky" loss is still a loss. Don't put yourself in a position where luck can be the difference in the outcome.

[Uncle Rico]

How "clutch" a team is certainly matters. But there's also a reality that there are factors that play into an outcome of a basketball game that teams have no control over. The opponent's career 40% FT shooter goes 9/9 from the line? Nothing you can do about that. Guy makes a well defended 40 fade away 40 footer at the Buzzer? Nothing you can do about that. Ref misses more fouls/violations committed by opponent/makes more erroneous calls on you? Nothing you can do about that. There are even dozens more micro examples of things that impact the outcome of games.

All of these things that you can't really do anything about can impact the outcome of a game and you might call that "luck". Luck tends to even out over time but in individual games one team could be more "lucky" than the other. Good teams take luck out of the question by playing well enough that even a few more bad bounces on their side of the ledger doesn't lead to a loss.

People don't like talking about luck because they view it as an excuse. It doesn't excuse anything but it is a reality in sports. We are unlucky in the sense that in each of our losses, we only need one or two more breaks on our side to potential turn the Ls into Ws whereas in our wins, our opponents would have needed a ton of extra things to break their way in order to change the outcome. In the end, it doesn't matter. A "lucky" win is still a win, an "unlucky" loss is still a loss. Don't put yourself in a position where luck can be the difference in the outcome.

Anyone that dismisses luck as a factor in any sport is an idiot

[wisblue]

How "clutch" a team is certainly matters. But there's also a reality that there are factors that play into an outcome of a basketball game that teams have no control over. The opponent's career 40% FT shooter goes 9/9 from the line? Nothing you can do about that. Guy makes a well defended 40 fade away 40 footer at the Buzzer? Nothing you can do about that. Ref misses more fouls/violations committed by opponent/makes more erroneous calls on you? Nothing you can do about that. There are even dozens more micro examples of things that impact the outcome of games.

All of these things that you can't really do anything about can impact the outcome of a game and you might call that "luck". Luck tends to even out over time but in individual games one team could be more "lucky" than the other. Good teams take luck out of the question by playing well enough that even a few more bad bounces on their side of the ledger doesn't lead to a loss.

People don't like talking about luck because they view it as an excuse. It doesn't excuse anything but it is a reality in sports. We are unlucky in the sense that in each of our losses, we only need one or two more breaks on our side to potential turn the Ls into Ws whereas in our wins, our opponents would have needed a ton of extra things to break their way in order to change the outcome. In the end, it doesn't matter. A "lucky" win is still a win, an "unlucky" loss is still a loss. Don't put yourself in a position where luck can be the difference in the outcome.

I think you are misreading my point.

I am not saying there is no luck in basketball. I’m just pointing out that what Pomeroy labels as “luck” isn’t necessarily luck, and he makes that pretty clear.

I used the word “clutch” as an alternative word that would be just as inaccurately simplistic as “luck”.

I do, think, however, that MU losing all 4 of its close games this season is a reflection of a lot more than just bad luck.

I think their worst luck of the season came in the first half of the UW game when the Badgers made a much greater percentage of their 3 point shots than they would usually make given the nature of the shots and the players taking them. I know there was some angst on the board about MU playing bad defense against the threes, but some of those shots (like the ones Hepburn made from way behind the arc as the shot clock was running out) gave UW a big lead that proved to be too much for MU to overcome.

The other 3 losses all featured some pretty abysmal execution, especially on offense, in the last quarter of the game.

[TAMU, Knower of Ball]

The other 3 losses all featured some pretty abysmal execution, especially on offense, in the last quarter of the game.

I understood your point. I was explaining another definition of luck. Both are true.

[MU82]

Anyone that dismisses luck as a factor in any sport is an idiot

Yessir.

Good teams take luck out of the question by playing well enough that even a few more bad bounces on their side of the ledger doesn't lead to a loss.

And yessir to you, too.

All of that can be true ... and is true in Marquette's case IMHO.

[bilsu]

Sometimes teams do lose, because of bad luck.

A couple of years ago MU was unlucky in a lost to Xavier. Xavier missed the shot to tie the game and the miss shot bounced all the way out to Kunkel who hit a three to win the game. The missed shot ending up in Kunkel's hands was pure luck for Xavier and therefore unlucky for MU.

[TAMU, Knower of Ball]

Sometimes teams do lose, because of bad luck.

A couple of years ago MU was unlucky in a lost to Xavier. Xavier missed the shot to tie the game and the miss shot bounced all the way out to Kunkel who hit a three to win the game. The missed shot ending up in Kunkel's hands was pure luck for Xavier and therefore unlucky for MU.

IIRC, the refs also missed a moving screen on the first shot, also bad luck. Part of the game

[DoctorV]

Sometimes teams do lose, because of bad luck.

A couple of years ago MU was unlucky in a lost to Xavier. Xavier missed the shot to tie the game and the miss shot bounced all the way out to Kunkel who hit a three to win the game. The missed shot ending up in Kunkel's hands was pure luck for Xavier and therefore unlucky for MU.

Fu*k Cunckel

[Class71]

Ken Pom's model is stochastic. That is outcomes are based on probabilities, probabilities that are continually changing as more data are collected. That is why his projections of winners, for example, indicating team A will beat team B 65% of the time does not always predict the actual result.

Having said that luck has nothing to do with a team that is expected to win when it does not win. The errors are due to uncertainty that is caused by the limitations of the model. The better the model the lower the uncertainly, the better the predictions.

The worst model possible predicts all teams have a 50/50 chance of winning. Clearly a useless model.

To test the value of a model it's predictions need to be measured against actual results. The more the predictions are close to actual outcomes (100% vs 50%) the better the model. If the model's predictions are considerably less than 50% then it might be a good model that has reverse correlation.

So is Ken Pom's model any good? Start by looking how his predictions stack up against actual results over several years for many teams. Then look at his methodology if he is willing to share.

Final thought is, with enough data, this problem is likely best solved with one of the existing AI algorithms. AI algorithms learn or adjust as you feed it more data. That is why data is the key to the future and why companies and governments are so interested in collecting data on everything.