G.L.Piggy [at] gmail.com
Recent Comments
- Hank on This is not The Jungle
- Bob Esponja on This is not The Jungle
- Bob Esponja on This is not The Jungle
- Bob Esponja on What’s Lacking in Women’s Basketball?
- Camlost on This is not The Jungle
- a_peraspera on This is not The Jungle
GLP Approved
- Captain Capitalism
- Dalrock
- Dennis Mangan
- Econ Log
- Elusive Wapiti
- Foseti
- Hawaiian Libertarian
- HBD Bibliography
- Heartiste
- Lion of the Blogosphere
- Marginal Revolution
- Matt Forney
- Randall Parker
- Razib Khan
- Reihan Salam
- Roosh V
- Simon Grey
- SOBL
- Steve Sailer
- The Last Psychiatrist
- The Spearhead
- Thrasymachus
- Ulysses
- Vox Day
Tweet Tweet
Error: Twitter did not respond. Please wait a few minutes and refresh this page.
They say that what we consider streaks are mostly just the positive segments of random chance. I really don’t see how this is true. I fully believe in streaks. Streaks are completely about confidence….
————————————————————
I wonder whether the pointy headed academicians who aver that ‘hot streaks’ are just random chance have ever shot a basketball, or thrown a baseball, or served a tennis ball, or swung a golf club….
I don’t think anybody would deny that technique (physical motion which is influenced by confidence level) plays a role in each of these activities and that the quality of technique can wax and wane day-to-day, even at the highest levels of competition.
Point being, if hot streaks are merely the ‘positive segments of random chance’, then likewise cold streaks are just the ‘negative segments of random chance.’ Accordingly, poor technique does not lead to worse results and good technique does not lead to better results….
….but anybody that has ever shot a basketball knows this is absurd — technique matters.
Go to the pdf article, ‘The Hot Hand in Basketball’, in the article linked above, and search for where the authors control for whether any particular shooter is letting his elbow fly out versus keeping it pointed at the bucket, or for whether a particular shooter is getting his legs into the shot. It’s no surprise that they don’t control for these sorts of things, but the impossibility of quantifying these sorts of variable does not make them less relevant.
So instead of saying that hot streaks are just the ‘positive segments of random chance’, I think the appropriate conclusion is that hot streaks and cold streaks are endemic — they are the norm, hence it’s exceedingly difficult to statistically isolate the norm from the ….. norm.
Similarly, a lot sports nuts / stat geeks like to say that there is no such thing as ‘clutch’ (John Hollinger, for example — he plays the role of a statistician on ESPN, but not very well imo). That’s nonsense, for if there is no such thing as ‘clutch’ then there is likewise no such thing as ‘choke’. Anyone that has ever crapped their pants in a pressure situation can tell you otherwise…
@2
GLP, So in the face of a vast sea of empirical evidence against “streaks” or, “the hot hand,” you counter with your… faith… in streaks. “Gamers” will respond with some crap about “inner game” and confidence what not, but it’s just people blowing smoke up each other’s asses.
This is like people insisting on blank slatism with regard to HBD despite the body of empiricism against it.
@alexamenos
“Similarly, a lot sports nuts / stat geeks like to say that there is no such thing as ‘clutch’ (John Hollinger, for example — he plays the role of a statistician on ESPN, but not very well imo). That’s nonsense, for if there is no such thing as ‘clutch’ then there is likewise no such thing as ‘choke’. Anyone that has ever crapped their pants in a pressure situation can tell you otherwise…”
I’ve crapped my pants in a pressure situation. I’ve also done some amazing things in pressure situations. I’ve also done a variety of things in between that cluster around average… it’s almost like… there’s a normal distribution at work.
However, I agree on Hollinger. John Hollinger is the “everyman” of sabermetrics. Smart enough to be impressive to Joe Six-Pack, but an infuriatingly a simpleton to those above average in intelligence.
@myself
Last bit should read as “but an infuriatingly simpleton”… how ironic.
Miley:
“GLP, So in the face of a vast sea of empirical evidence against “streaks” or, “the hot hand,” you counter with your… faith… in streaks. “Gamers” will respond with some crap about “inner game” and confidence what not, but it’s just people blowing smoke up each other’s asses.
This is like people insisting on blank slatism with regard to HBD despite the body of empiricism against it.”
The article itself mentions that bowlers experience hot hands and that Joe DiMaggio’s 56 game streak had a component of non-random chance to it. What, only bowling and DiMaggio ever had streaks and streaks stopped occurring outside of those isolated incidents? That’s absurd. Besides, why do you believe that there is a “vast sea” of evidence supporting the claim that streaks are merely random chance? You’re going to place your “faith” in this fledgling research?
They mostly leave out the variable of confidence and yet sports are greatly influenced by the confidence of the participants at any given moment. The better research question might be: does confidence impact performance?
“Besides, why do you believe that there is a “vast sea” of evidence supporting the claim that streaks are merely random chance? You’re going to place your “faith” in this fledgling research?”
Probably years of following the NBA when I was younger, and seeing research piece after research piece (some done informally by interested and not unsophisticated parties; some more academic in nature, e.g. the Burns piece) finding no evidence of “the hot hand”? In fact, the arrow of confidence pointed the other way. Some research detected the opposite: players got overconfident during a streak and start taking poorer shots, and thus missing more than they made relative to no streak.
To dimiss the research as “fledging” because you don’t agree with it out of subjectivity is disingenuous. I may come off as hostile only because I enjoy reading your other stuff, and it’s disappointing to see how you handled this.
‘The article itself mentions that bowlers experience hot hands and that Joe DiMaggio’s 56 game streak had a component of non-random chance to it. What, only bowling and DiMaggio ever had streaks and streaks stopped occurring outside of those isolated incidents? That’s absurd.’
I’ll give you bowling, maybe, but the DiMaggio example is reminiscent of Malkiel’s coin-flipping thought-experiment.
“To dimiss the research as “fledging” because you don’t agree with it out of subjectivity is disingenuous. I may come off as hostile only because I enjoy reading your other stuff, and it’s disappointing to see how you handled this.”
The research *is* fledgling. Which is fine. All research topics must go through such a stage. Do you believe that everything that can be known about hot streaks has been learned through this small amount of research? I doubt it. I think you’re placing way too much emphasis on what I’ve written about it. You speak of how I’ve “handled this”. What am I handling? I made a comment about the topic on a facet of the argument which the researchers don’t seem to take into much account.
I’ve played sports and I know that confidence is important. When you shoot a 3, sometimes you just know it is going in. Sometimes you know that it isn’t. Of course, there are also instances where you think that the shot is going in but it doesn’t, and you believe that it isn’t going to go in but it does. But there’s a research question right there and one that I don’t think researchers have modeled. Basically, researchers want to boil players down to automatons or robots who are not influenced by human emotions. So how do you model human emotion?
“But there’s a research question right there and one that I don’t think researchers have modeled. Basically, researchers want to boil players down to automatons or robots who are not influenced by human emotions. So how do you model human emotion?”
You don’t need to “model human emotion” in order to come to the conclusion, that, empirically, the “hot hand” does not increase the probability of the next shot going in. In fact, the tendency is for the opposite: streaks decrease the probability of the next shot going in.
Your view on being unable to “model human emotion” sounds strangely Luddite, like the naysayers of Moneyball in the nascent days of Billy Beane.
“I’ve played sports and I know that confidence is important. When you shoot a 3, sometimes you just know it is going in. Sometimes you know that it isn’t. Of course, there are also instances where you think that the shot is going in but it doesn’t, and you believe that it isn’t going to go in but it does.”
A gambler could say “I’ve gambled and know that the sixth sense exists. Sometimes as a dealer is flipping that card, you know it’s going to be a BlackJack. Sometimes you know it isn’t. Of course, there are also instances where you think it’s going to be a BlackJack but it isn’t, and you believe that it isn’t going to be a BlackJack but it is.”
Human behavior does not need to fully understood in order for the outcomes of it to be modeled and measured.
Look, I’ve also played a lot of sports as well. I recall the times where I was sure I made the game-winning lay-up only to have it clank off the back of the rim, or where I was sure it was going to be a winner down the line only to have it sail wide. I recall the times where I was just hoping and praying and scored a go-ahead goal in soccer from 35 yards out or made an impossible return in tennis. Then I think a bit harder and there’s tons of instances in between.
Miley,
I’m at work, so I can’t give very long responses, but it occurs to me that it’s incorrect to think about streaks the way they were discussed in the OP. A streak in baseball is entirely different than a streak in basketball. Batting in baseball (and pitching) is a dominance game. Shooting a basketball is not a dominance game; it is more about chance, and is mostly just one guy against one basket with one ball.
I’ve crapped my pants in a pressure situation. I’ve also done some amazing things in pressure situations. I’ve also done a variety of things in between that cluster around average… it’s almost like… there’s a normal distribution at work.
—————————————————————————–
Would you argue that all people have an equal propensity to crap or not crap their pants in pressure situations? It seems to me that once it’s admitted that some people don’t perform as well under pressure as others (relative to their own non-pressure abilities) then the reality of ‘choking’ is well established. Once it’s acknowledged that some people have a greater propensity for choking than others then meaning of ‘clutch’ becomes clear — a clutch performer is one not negatively affected by pressure.
A streak in baseball is entirely different than a streak in basketball. Batting in baseball (and pitching) is a dominance game.
——————————————–
I would suggest that golf is a good game for considering the subject.
To say that there is no such thing as a ‘hot putter’ is to say that the quality of a golfer’s putting stroke doesn’t vary from time-to-time, nor does his stroke occasionally get dialed in to such a degree that he can hit his desired line and speed consistently for several holes in a row.
That is an almost Druge Report worthy juxtaposition of the last sentence of the description of link 2 and the 3rd link. Bravo.
Neither walks nor sacrifices count towrdas at bats. If a player was intentionally walked 4 times, he would not have had any at bats, therefore, if he were to get a hit in the next game, his streak would continue (but would not count the game where he had 4 IBBs).In the same manner, a player could get a hit in 14 games and then takes one game off because of a small injury.Then he comes back in the next game and gets a hit. His streak would still be 15 games (eventhough 16 games had been played) because he did not have an at bat in that game.Airline Pilot