Ben Johnson is a Disaster

Donate To GopherHole
dpodoll68 said:
Holy balls.

First of all, the post that Bad quoted was in regard to your comment about something you made up that I said about "football vs. Wisconsin". Try to keep up - difficult with limited brainpower, I know.

Secondly, please point out where in that post I said "he['ll] never be good enough to coach as an assistant in the B1G." Here's a free hint - peruse the word "probably."

You are not bright. But it's a Friday and it's fun playing with the challenged, particularly a total dipshit who has the balls to call me a "moron" when he can barely read and certainly can't follow an argument that he himself is feebly attempting to make.
Click to expand...
Peace on earth. Goodwill to all.
 


BroncoRedux said:
Minnesota Nice? This thread now contains misogynist, racist, and homophobic references (really, butthurt?). Surely there must be others to attack? Get started, I know you'll do the haters proud.
Click to expand...
I've read a lot of brain dead shit on this board over the years. But wow this one might take the cake
 












tikited said:
I love reading Jay's hot takes. I feel kinda bad for him being so sure he's right though.
Click to expand...
Bad Gopher said:
According to Ted Knight's character in Caddyshack, the world needs ditch diggers. It also needs cesspool builders. Mission accomplished: this thread is now a cesspool.
Click to expand...
I've been keeping up in all the other threads. But I dropped out of this cesspool on page 8 LOL. I'll read pages 9-15 when I get a chance. I jumped back in and read pages 16-17.
 





formerlybis said:
False? Please explain.

Or do you not believe in analytics (especially when you disagree with the conclusions)?
Click to expand...
I don't know anything about how this is calculated, but I would've guessed our chances of going winless in conference are more like 50%, not 4%. These prognosticators probably haven't watched us in person.
 

Spaulding!No! said:
If we had known the Pitino replacement was baby Ben maybe we would a just shut up and sat down.
Click to expand...
That's actually a good idea. A first-time head coach isn't expected to bring a contending team immediately. We SHOULD shut up and sit down until his recruits are juniors & sophomores.
 


MNJay said:
That's actually a good idea. A first-time head coach isn't expected to bring a contending team immediately. We SHOULD shut up and sit down until his recruits are juniors & sophomores.
Click to expand...

Tell that to the other first time head coaches who have significantly better teams.

Plus this entire team is Ben’s recruits, and 7 of the 12 on scholarship are sophomores and older.
 

GopherWeatherGuy said:
Tell that to the other first time head coaches who have significantly better teams.

Plus this entire team is Ben’s recruits, and 7 of the 12 on scholarship are sophomores and older.
Click to expand...
Have you considered running for political office? You'd be great at it.
 

Allshade said:
I don't know anything about how this is calculated, but I would've guessed our chances of going winless in conference are more like 50%, not 4%. These prognosticators probably haven't watched us in person.
Click to expand...
I know this was in jest, but I have this overwhelming desire to make the world more literate on statistics...

The probability of rolling snake eyes on two dice is 1 in 36 - 1/6 x 1/6. That's less than 3%. The probability of Minnesota going winless over the remaining 18 (!) conference games is calculated at over 4%.

Saying it again for emphasis, the chances that Minnesota goes winless in conference - with 18 chances - is greater than rolling snake eyes.
 

formerlybis said:
I know this was in jest, but I have this overwhelming desire to make the world more literate on statistics...

The probability of rolling snake eyes on two dice is 1 in 36 - 1/6 x 1/6. That's less than 3%. The probability of Minnesota going winless over the remaining 18 (!) conference games is calculated at over 4%.

Saying it again for emphasis, the chances that Minnesota goes winless in conference - with 18 chances - is greater than rolling snake eyes.
Click to expand...
Let's look at it this way- the odds of winning a game are additive of the odds for each game so with 18 remaining games, if our odds of winning any single game were 5% our odds of winning one game all year (or more) would be 90%- thus in that case the odds of going winless would be 10%.

We are going to have some games where are odds are 30% and some where our odds are 2%. Until we get better- which I expect we will, it is tough to imagine us being favored in a BigTen game as we are today.
 

bga1 said:
Let's look at it this way- the odds of winning a game are additive of the odds for each game so with 18 remaining games, if our odds of winning any single game were 5% our odds of winning one game all year (or more) would be 90%- thus in that case the odds of going winless would be 10%.

We are going to have some games where are odds are 30% and some where our odds are 2%. Until we get better- which I expect we will, it is tough to imagine us being favored in a BigTen game as we are today.
Click to expand...
Well, not quite on the first paragraph. If the probability of losing each game was exactly 95%, the chances of going winless over 18 games is 39.7% (.95 to the 18th power). So, it's not quite that bad. It's more like the Gophers have a 16% chance of winning, on average, any of the remaining B1G games, though, which is still really bad.

According to t-rank, the game against Nebraska at the Barn on 1/7 is currently a 39% chance. That's currently the best odds of any remaining game.

Sorry for the stats class.
 

formerlybis said:
Well, not quite on the first paragraph. If the probability of losing each game was exactly 95%, the chances of going winless over 18 games is 39.7% (.95 to the 18th power). So, it's not quite that bad. It's more like the Gophers have a 16% chance of winning, on average, any of the remaining B1G games, though, which is still really bad.

According to t-rank, the game against Nebraska at the Barn on 1/7 is currently a 39% chance. That's currently the best odds of any remaining game.

Sorry for the stats class.
Click to expand...
You are probably right. It's been about 45 years since my last stats class so I will bow to you.
 

formerlybis said:
Well, not quite on the first paragraph. If the probability of losing each game was exactly 95%, the chances of going winless over 18 games is 39.7% (.95 to the 18th power). So, it's not quite that bad. It's more like the Gophers have a 16% chance of winning, on average, any of the remaining B1G games, though, which is still really bad.

According to t-rank, the game against Nebraska at the Barn on 1/7 is currently a 39% chance. That's currently the best odds of any remaining game.

Sorry for the stats class.
Click to expand...

How much of the probability model is subjective data? The "odds" of winning any given game with 100's of variables do not exactly equate to the 1/36 chance of getting snake eyes, which is an exact mathematical equation, with only six variables per die. Modeling isn't exactly the same, especially in something as subjective as games played by teenagers and young adults, who can also be quite unpredictable. My suggestion is that it is impossible to suggest an exact probability in basketball like you can with dice and cards.

For fun, I wonder what probability models might have given to Villanova to beat Georgetown in what is likely the biggest upset in NCAA championship history.
 

Holy Man said:
How much of the probability model is subjective data? The "odds" of winning any given game with 100's of variables do not exactly equate to the 1/36 chance of getting snake eyes, which is an exact mathematical equation, with only six variables per die. Modeling isn't exactly the same, especially in something as subjective as games played by teenagers and young adults, who can also be quite unpredictable. My suggestion is that it is impossible to suggest an exact probability in basketball like you can with dice and cards.

For fun, I wonder what probability models might have given to Villanova to beat Georgetown in what is likely the biggest upset in NCAA championship history.
Click to expand...

The number he captured was accurate if you assume independence from one game to the next and assume the underlying probabilities for each game are accurate (95% in this case). Then it truly is the product of the probabilities. Of course, it doesn't take into consideration what a win might do for the next game, unknown injuries, etc. Too many variables to count, but fun nonetheless if you are a stats dork like me.
 

golfing18now said:
The number he captured was accurate if you assume independence from one game to the next and assume the underlying probabilities for each game are accurate (95% in this case). Then it truly is the product of the probabilities. Of course, it doesn't take into consideration what a win might do for the next game, unknown injuries, etc. Too many variables to count, but fun nonetheless if you are a stats dork like me.
Click to expand...
Yes. The question of course is the assumption of the underlying probabilities which are much more subjective than the analogy to dice and cards, which are fixed and known. My own editorial position is that people are tempted to equate the probability certitude of snake eyes to things like winning games. Modeling has a subjectivity to it that typically isn't addressed in a 300 level stats or QA class, or I suspect on a public message board. Subjectivity notwithstanding, it will be a disaster if the Gophers go winless in the B1G.
 

Holy Man said:
How much of the probability model is subjective data? The "odds" of winning any given game with 100's of variables do not exactly equate to the 1/36 chance of getting snake eyes, which is an exact mathematical equation, with only six variables per die. Modeling isn't exactly the same, especially in something as subjective as games played by teenagers and young adults, who can also be quite unpredictable. My suggestion is that it is impossible to suggest an exact probability in basketball like you can with dice and cards.

For fun, I wonder what probability models might have given to Villanova to beat Georgetown in what is likely the biggest upset in NCAA championship history.
Click to expand...
I love talking this stuff. *None* of the data are subjective. But the underlying variables are subject to sometimes large variation - random or otherwise - in single games (that's why they're called "variables").

I think of it this way. Start with the idea that a team has a floor and a ceiling for performance. Then put those on a scale of 0-100. Some teams are more consistent - their range will be narrower than those who aren't. So let's use Georgetown and Villanova as our example. This is completely made up. Georgetown may have been a range of 85-100 while Villanova was 70-90. Georgetown would be a big favorite, but there was the possibility of Villanova giving an 87 performance when Georgetown was 86. It happens.

What kenpom and t-rank are doing in giving their exact probabilities for future games is perhaps a bit misleading, but only slightly. You don't know the standard error (like when you see political polls say "the race is candidate A x% and candidate B y% +/- 3%"). But, those kenpom and t-rank numbers are the midpoint of those ranges, and thus, valid in my eyes.
 

Holy Man said:
How much of the probability model is subjective data? The "odds" of winning any given game with 100's of variables do not exactly equate to the 1/36 chance of getting snake eyes, which is an exact mathematical equation, with only six variables per die. Modeling isn't exactly the same, especially in something as subjective as games played by teenagers and young adults, who can also be quite unpredictable. My suggestion is that it is impossible to suggest an exact probability in basketball like you can with dice and cards.

For fun, I wonder what probability models might have given to Villanova to beat Georgetown in what is likely the biggest upset in NCAA championship history.
Click to expand...
They build models for things for more complex than basketball. Analytics and a lot of gambling odds are often built from those things in the medical and financial industries. There will always be disagreement on things to put in the model (at what weight), but the models do exist.

But you're right, they aren't going to be as accurate as purely objective things like coin toss or dice. However, a lot of people make a lot of money utilizing similar kinds of models for complex situations.
 


Bob_Loblaw said:
They build models for things for more complex than basketball. Analytics and a lot of gambling odds are often built from those things in the medical and financial industries. There will always be disagreement on things to put in the model (at what weight), but the models do exist.

But you're right, they aren't going to be as accurate as purely objective things like coin toss or dice. However, a lot of people make a lot of money utilizing similar kinds of models for complex situations.
Click to expand...

Weather has entered the chat
 

You must log in or register to reply here.



Top Bottom