MBB: Michigan Intercollegiate Athletic Association

[Knightmare]

Exhibit "A" in support of hot streaks (and corresponding cold streaks).  These are the highlights of Vinnie Johnson 4th quarter of game 4 of the 1985 playoff series against the Boston Celtics.

Take a quick look.  And for good measure "The Microwave" tosses in a double pump perimeter shot for the icing on the cake.

http://www.youtube.com/watch?v=tcKu-U49GA8&feature=related

http://www.d3boards.com/index.php?topic=4596.msg1242754#msg1242754

What you referenced says that Vinnie's shooting percentage wouldn't change during these times, he just shot more frequently after making the first bucket or two.  After watching the video I'd have to say that his shooting percentage changed quite a bit during the 4th quarter and game as a whole.  Does his shooting % change for the season if taking all statistics over the course of 82 games, no, but that is because you are looking at a very big picture that is balanced out by ebbs and flows in his shooting both good and bad that constitutes both cold and hot streaks during the course of a season.  Yes, over the course of time the statistics will regress to a mean or average value but not necessarily over short time frames like one quarter or game or couple game period where a player can be playing better or worse than the expected average.

We may have to just agree to disagree on these theories because in reality that is what both sides of this debate are.  Though this is quite interesting to hear both sides, all good naturedly too which is much appreciated from everyone that it has remained civil.

[KnightSlappy]

http://www.d3boards.com/index.php?topic=4596.msg1242754#msg1242754

What you referenced says that Vinnie's shooting percentage wouldn't change during these times, he just shot more frequently after making the first bucket or two.  After watching the video I'd have to say that his shooting percentage changed quite a bit during the 4th quarter and game as a whole.  Does his shooting % change for the season if taking all statistics over the course of 82 games, no, but that is because you are looking at a very big picture that is balanced out by ebbs and flows in his shooting both good and bad that constitutes both cold and hot streaks during the course of a season.  Yes, over the course of time the statistics will regress to a mean or average value but not necessarily over short time frames like one quarter or game or couple game period where a player can be playing better or worse than the expected average.

I think you're missing the point. It wasn't that the streaks didn't change his season percentage, it was that his percentage when only considering shots taken immediately after a streak began wasn't any different than his seasonal numbers.

So if he was a 45% shooter overall, the researched showed that he also shot 45% on shots following, say, three makes in a row.

Yes, there were instances when he made more shots in a row than would be expected of a 45% shooter, but it didn't happen more than often than random variation would suggest.

[ziggy]

Exhibit "A" in support of hot streaks (and corresponding cold streaks).  These are the highlights of Vinnie Johnson 4th quarter of game 4 of the 1985 playoff series against the Boston Celtics.

Take a quick look.  And for good measure "The Microwave" tosses in a double pump perimeter shot for the icing on the cake.

http://www.youtube.com/watch?v=tcKu-U49GA8&feature=related

http://www.d3boards.com/index.php?topic=4596.msg1242754#msg1242754

What you referenced says that Vinnie's shooting percentage wouldn't change during these times, he just shot more frequently after making the first bucket or two.  After watching the video I'd have to say that his shooting percentage changed quite a bit during the 4th quarter and game as a whole.  Does his shooting % change for the season if taking all statistics over the course of 82 games, no, but that is because you are looking at a very big picture that is balanced out by ebbs and flows in his shooting both good and bad that constitutes both cold and hot streaks during the course of a season.  Yes, over the course of time the statistics will regress to a mean or average value but not necessarily over short time frames like one quarter or game or couple game period where a player can be playing better or worse than the expected average.

We may have to just agree to disagree on these theories because in reality that is what both sides of this debate are.  Though this is quite interesting to hear both sides, all good naturedly too which is much appreciated from everyone that it has remained civil.

It takes a big set of data to prove a big point. The video just proves that it is possible for someone to make a bunch of baskets in a row, that was never in doubt.

[Flying Dutch Fan]

Guys I think we are all in agreement we just are looking at this in two different ways. Hot streaks (and cold ones) do exist and we can cite endless examples.  Some are discussing the probability of the next shot going in which is purely mathematical. Others of us are looking more at the streaks as a whole and have a "problem" with using math only since the game is so much more than that. 

I just know that once the ball is tossed for the opening tip none of this will matter.

[ziggy]

Guys I think we are all in agreement we just are looking at this in two different ways. Hot streaks (and cold ones) do exist and we can cite endless examples.  Some are discussing the probability of the next shot going in which is purely mathematical. Others of us are looking more at the streaks as a whole and have a "problem" with using math only since the game is so much more than that. 

I just know that once the ball is tossed for the opening tip none of this will matter.

It sounds like we agree. We both want Coach Neil to give looks to a 20% three point shooter just because he started 2-for-2.

[pointlem]

Some are discussing the probability of the next shot going in which is purely mathematical.

Agreed, FDF . . . in the sense that the game's final score is purely arithmetic . . . and that players, coaches, and fans are making mathematical (probabilistic) assumptions when thinking a coach should leave in the game Player X with the hot hand (i.e., that Player X has an increased chance of making baskets after a string of makes).  Implicit mathematical assumptions do affect strategy.

But we've about worked this argument dry . . . so how about a different one (this is fun, and tipoff is a month away):  There's another sports-relevant cognitive illusion (a "temporal contiguity" illusion) that is usually good for a debate.  It's the well-established human tendency to see events occurring just before a final outcome as more determinative of the outcome than are equally important earlier events.  A sports example:  A three point shot in the last minute seems more crucial than the same three points midway through the first half.  Thus rather than maximizing a star player's playing time and points throughout a game, many a coach will sit a player with two first half fouls in order to have the player available "when the game is on the line."  Better to have Star X play 20 minutes, and be available the last 5 minutes, than to play all-out for 30 minutes and foul out with five minutes left . . . because those final points seem to matter more.  In the rest of life as in basketball, temporal contiguity is a source of much superstitious behavior.  (I know, it may be smart to give the player with two fouls a rest . . . but is it more important to play the last five minutes or to maximize minutes?  Should a player with two fouls in the first five minutes be benched for the rest of the half, until that magic 20:00 time warp?)

[ziggy]

Some are discussing the probability of the next shot going in which is purely mathematical.

Agreed, FDF . . . in the sense that the game's final score is purely arithmetic . . . and that players, coaches, and fans are making mathematical (probabilistic) assumptions when thinking a coach should leave in the game Player X with the hot hand (i.e., that Player X has an increased chance of making baskets after a string of makes).  Implicit mathematical assumptions do affect strategy.

But we've about worked this argument dry . . . so how about a different one (this is fun, and tipoff is a month away):  There's another sports-relevant cognitive illusion (a "temporal contiguity" illusion) that is usually good for a debate.  It's the well-established human tendency to see events occurring just before a final outcome as more determinative of the outcome than are equally important earlier events.  A sports example:  A three point shot in the last minute seems more crucial than the same three points midway through the first half.  Thus rather than maximizing a star player's playing time and points throughout a game, many a coach will sit a player with two first half fouls in order to have the player available "when the game is on the line."  Better to have Star X play 20 minutes, and be available the last 5 minutes, than to play all-out for 30 minutes and foul out with five minutes left . . . because those final points seem to matter more.  In the rest of life as in basketball, temporal contiguity is a source of much superstitious behavior.  (I know, it may be smart to give the player with two fouls a rest . . . but is it more important to play the last five minutes or to maximize minutes?  Should a player with two fouls in the first five minutes be benched for the rest of the half, until that magic 20:00 time warp?)

Let's say I put a large sum of money behind one of three doors. I ask you to pick between door 1, 2, and 3.

You pick door number 1 and I then open door number 3, revealing an empty room.

I then give you the opportunity to reconsider your selection of door number 1. Only a fool would not reconsider.

The initial selection was made among three possible outcomes, giving you a 33.3% chance at "success".
Reconsidering gives you the option to choose among two possible outcomes, giving you a 50% chance at "success".

[KnightSlappy]

Some are discussing the probability of the next shot going in which is purely mathematical.

Agreed, FDF . . . in the sense that the game's final score is purely arithmetic . . . and that players, coaches, and fans are making mathematical (probabilistic) assumptions when thinking a coach should leave in the game Player X with the hot hand (i.e., that Player X has an increased chance of making baskets after a string of makes).  Implicit mathematical assumptions do affect strategy.

But we've about worked this argument dry . . . so how about a different one (this is fun, and tipoff is a month away):  There’s another sports-relevant cognitive illusion (a “temporal contiguity” illusion) that is usually good for a debate.  It’s the well-established human tendency to see events occurring just before a final outcome as more determinative of the outcome than are equally important earlier events.  A sports example:  A three point shot in the last minute seems more crucial than the same three points midway through the first half.  Thus rather than maximizing a star player’s playing time and points throughout a game, many a coach will sit a player with two first half fouls in order to have the player available “when the game is on the line.”  Better to have Star X play 20 minutes, and be available the last 5 minutes, than to play all-out for 30 minutes and foul out with five minutes left . . . because those final points seem to matter more.  In the rest of life as in basketball, temporal contiguity is a source of much superstitious behavior.  (I know, it may be smart to give the player with two fouls a rest . . . but is it more important to play the last five minutes or to maximize minutes?  Should a player with two fouls in the first five minutes be benched for the rest of the half, until that magic 20:00 time warp?)

Points are points and minutes are minutes, but minutes late in the game are usually of the high leverage variety.

I would guess you would want to maximize the time that your star player is on the floor, but keep the game leverage in mind as you do so. I think coaches probably tend to "save" their players a bit too much, but I think their general idea is probably right.

Let's say the probability of Team A and Team B winning a game is 50% at the beginning. If team A makes the game's first shot, the win probability shifts in their favor only a fraction of a percentage. If the game is tied with only 10 seconds left and Team A hits a basket, the win probability shifts heavily in their favor. Both baskets are worth two points, but they aren't worth the same in terms of win probability added.

It's similar to when you bring in your closer in a baseball game. Do you always save him for the ninth inning, or do you bring him in the eighth? It's all about leverage index, although LI is much easier to quantify in baseball (in my opinion) with it's discrete, 24 base-out state, nature.

[Mr. Ypsi]

Some are discussing the probability of the next shot going in which is purely mathematical.

Agreed, FDF . . . in the sense that the game's final score is purely arithmetic . . . and that players, coaches, and fans are making mathematical (probabilistic) assumptions when thinking a coach should leave in the game Player X with the hot hand (i.e., that Player X has an increased chance of making baskets after a string of makes).  Implicit mathematical assumptions do affect strategy.

But we've about worked this argument dry . . . so how about a different one (this is fun, and tipoff is a month away):  There's another sports-relevant cognitive illusion (a "temporal contiguity" illusion) that is usually good for a debate.  It's the well-established human tendency to see events occurring just before a final outcome as more determinative of the outcome than are equally important earlier events.  A sports example:  A three point shot in the last minute seems more crucial than the same three points midway through the first half.  Thus rather than maximizing a star player's playing time and points throughout a game, many a coach will sit a player with two first half fouls in order to have the player available "when the game is on the line."  Better to have Star X play 20 minutes, and be available the last 5 minutes, than to play all-out for 30 minutes and foul out with five minutes left . . . because those final points seem to matter more.  In the rest of life as in basketball, temporal contiguity is a source of much superstitious behavior.  (I know, it may be smart to give the player with two fouls a rest . . . but is it more important to play the last five minutes or to maximize minutes?  Should a player with two fouls in the first five minutes be benched for the rest of the half, until that magic 20:00 time warp?)

Let's say I put a large sum of money behind one of three doors. I ask you to pick between door 1, 2, and 3.

You pick door number 1 and I then open door number 3, revealing an empty room.

I then give you the opportunity to reconsider your selection of door number 1. Only a fool would not reconsider.

The initial selection was made among three possible outcomes, giving you a 33.3% chance at "success".
Reconsidering gives you the option to choose among two possible outcomes, giving you a 50% chance at "success".

Ah, you must be a fan of Marilyn vos Savant (who did this identical puzzle a few years ago).  Though she never acknowledged it, I disproved her then, and I can do it to you now!

IF you were opening a door at random (such that #3 could have the treasure), your conclusion would be correct.  Since you (presumably) KNOW where the treasure is, you could ALWAYS open either #2 or #3 to reveal an empty room - you have added NOTHING to the information available to the contestant.  While it is true that their initial pick was 1/3 for any room, their current pick is 1/2 FOR EITHER REMAINING ROOM.  Switching or not switching is entirely arbitrary.

But I'm unclear how this relates to the 'time of the game' puzzle.  A score with 15 minutes left will (can) change the strategies for the rest of the game.  A score with 3 seconds left might well be the final crucial difference.

[KnightSlappy]

I don't want to speak for him, but I don't think ziggy meant that to be taken in a serious manner.

[ziggy]

Ah, you must be a fan of Marilyn vos Savant (who did this identical puzzle a few years ago).  Though she never acknowledged it, I disproved her then, and I can do it to you now!

IF you were opening a door at random (such that #3 could have the treasure), your conclusion would be correct.  Since you (presumably) KNOW where the treasure is, you could ALWAYS open either #2 or #3 to reveal an empty room - you have added NOTHING to the information available to the contestant.  While it is true that their initial pick was 1/3 for any room, their current pick is 1/2 FOR EITHER REMAINING ROOM.  Switching or not switching is entirely arbitrary.

I think you're right. Plus in my view, the decision to or not to reconsider is making a defacto choice between two options thereby gaining the probability of "success" by default.

But I'm unclear how this relates to the 'time of the game' puzzle.  A score with 15 minutes left will (can) change the strategies for the rest of the game.  A score with 3 seconds left might well be the final crucial difference.

I doesn't. I more or less paraphrased a scenario presented in the movie "21" about the MIT guys that counted cards and swindled Vegas.

Good night, all.

[goodknight]

Calvin's 2010-2011 varsity and junior varsity rosters are posted on the college's website:

http://www.calvin.edu/sports/mens/basketball/roster.htm

KVS must like his recent recruits -- five sophs and four freshmen on the varsity. 

[calvin_grad]

Let's say I put a large sum of money behind one of three doors. I ask you to pick between door 1, 2, and 3.

You pick door number 1 and I then open door number 3, revealing an empty room.

I then give you the opportunity to reconsider your selection of door number 1. Only a fool would not reconsider.

The initial selection was made among three possible outcomes, giving you a 33.3% chance at "success".
Reconsidering gives you the option to choose among two possible outcomes, giving you a 50% chance at "success".

Ah, you must be a fan of Marilyn vos Savant (who did this identical puzzle a few years ago).  Though she never acknowledged it, I disproved her then, and I can do it to you now!

IF you were opening a door at random (such that #3 could have the treasure), your conclusion would be correct.  Since you (presumably) KNOW where the treasure is, you could ALWAYS open either #2 or #3 to reveal an empty room - you have added NOTHING to the information available to the contestant.  While it is true that their initial pick was 1/3 for any room, their current pick is 1/2 FOR EITHER REMAINING ROOM.  Switching or not switching is entirely arbitrary.

But I'm unclear how this relates to the 'time of the game' puzzle.  A score with 15 minutes left will (can) change the strategies for the rest of the game.  A score with 3 seconds left might well be the final crucial difference.

Sorry, Mr. Ypsi.  Ziggy's right.  Only a fool would not switch.  You'll win 2/3 of the time if you switch doors.

Time for some hoops.   

[sac]

Calvin's 2010-2011 varsity and junior varsity rosters are posted on the college's website:

http://www.calvin.edu/sports/mens/basketball/roster.htm

KVS must like his recent recruits -- five sophs and four freshmen on the varsity. 

I looked at that and thought there's no way Danny Rodts can be a Senior already.  Geez time flies.

[KnightSlappy]

Calvin's 2010-2011 varsity and junior varsity rosters are posted on the college's website:

http://www.calvin.edu/sports/mens/basketball/roster.htm

KVS must like his recent recruits -- five sophs and four freshmen on the varsity. 

I looked at that and thought there's no way Danny Rodts can be a Senior already.  Geez time flies.

I'll take the extra year if you're giving it. 

Danny spent a year on JV, so he wasn't "here forever" like Veltema and Veldhouse were.