I believe it was Stephen Gould (no not Eliot) who calculated that on average, the term “hot hands” (designated to a player who is in the “zone” or his peak performance of play) does not naturally exist. He took the number of shots scored for a basketball team (I forget which one), and calculated that (with allowance for player ability) the number of shots actually made during the supposed “hot hand” phenomenon was actually as likely as a statistical coin-flipping either head or tails. Sports streaks are normal, and a part of everyday statistics.
To read more (and discover my fallacy with the aforementioned authors name which I can’t be bothered to change, go here. There’s also an article in Scientific American about it, but I can’t remember the issue).
But there are exceptions to the rule. Bob Beamons record long jump. Joe DiMaggio’s 56-game hitting streak.
But taken from a statistical standpoint, which was the most greates sporting achievment of all… and I don’t mean in your opinion, or else this would be in IMHO. By the numbers, which was the most difficult feat ever accomplished in sporting history… and do we even know of it? Which one sporting thing defied all the conventional and outside odds to take its place in the annals of sporting legend? Is there such an event?
Johnny Vander Meer threw back-to-back no-hitters in June 1968. Eighteen consecutive innings without so much as a gork falling in. A few pitchers have managed two no-nos in a season, but never, ever back-to-back. Not even close. I’d call that feat untouchable.
Sports aren’t exactly a laboratory, and you’ll have a great deal of difficulty in trying to treat them like one. How could we ever reach an indisputable conclusion to this question? There’s no formula for comparing basketball shooting percentage to hitting streaks in baseball, to long field goals in football, to consecutive games played, etc etc etc. Hell, even if there was, how could it compare players in different eras? Would Wilt Chamberlain score 100 in the current NBA? Could DiMaggio hit in 56 straight in the 1880’s? There’s just no one definitive approach to answer these questions. The best you’ll get is conjecture by “experts” in each field, and even then, you can’t get them to agree across the board. And then, supposing you could get a statistically valid approach to determine how hard it is to, say, hit in 56 straight games, you’d have to be able to translate that approach into all feats in all athletic endeavors to be able to compare. In short, I think this question really is more of an IMHO, because there isn’t a factual answer to be had. I would say Cy Young’s wins records, for what its worth.
Hmmm… I’m not quite sure what you’re asking but it sounds like any achievement based on a streak would be considered a statistical accident and thus eliminated. In other words given enough at-bats eventually everyone would have a 56-game hitting streak, and given enough opportunities every pitcher would eventually pitch two no-hitters back to back. The same would apply to touchdowns in a game or season, points scored etc.
That leaves us with feats like Bob Beamon’s. Given an infinity of at-bats even I would eventually have 100 hits in a row, but I would never be able to broad jump - what was it - 26 feet.
So if you eliminate any feat based on a streak or any amount of luck, you’re not left with much. In football the longest run for a touchdown wouldn’t count but longest field goal might. I can’t think of anything in baseball or basketball that would count. In fact you’re pretty much stuck with individual sports like track and field, which brings me back around to Bob Beaman.
I think you’re misinterpreting the research. What the study shows is that, despite the perceptions of fans and commentators, it’s not necessarily wise for a coach to go with the player with the “hot hand.” The “hot” player can go cold at any time - these streaks never last forever - and you can never predict when that will happen. The conventional sports wisdom is more superstition than anything else, since statistically it doesn’t exist. I think that was the point of the study.
I would dispute the idea that “given enough games, anybody could have a 56-game hitting streak,” since it’s only happened once. A hit in baseball is a combination of a huge number of things, including position of the fielders, wind, bat speed, the pitcher, the pitch he throws, and thousands of other things. The fact that the “hot hand” phenomenon is meaningless does not mean DiMaggio’s streak was luck or a fluke of probability. I think it’s probably the greatest single achievement ever in sports.
I think that Xavier is trying to place a rough figure on certain sports feats. For example, I’m thinking Wilt’s game was one single example in all of the professional basketball games ever played, which should be calcul-- calclu-- which should be able to be figured out or guessed real good.
Somehow I think if you have to get that technical, guys like Craig Breedlove are going to enter this conversation.
I believe the “hot hand” phenomenon is not always random. Sometimes there is a physiological reason for the streak. When a jumpshooter is hot, it could mean that he has temporarily programmed the perfect stroke into his mind and is temporarily transferring this mind into muscle contractions perfectly. The hotness indeed does end, but only because this temporary channeling leaves the shooters’ mind naturally. The task of performing the same action again and again is too monotonous for a person and the streak has to end. The streak was wrought by a superior athlete channeling his mind well, but this occurrence is random. It doesn’t necessarily happen after a long day of practice the day before. It’s the product of possessing the ability long enough for the randomness to occur. But the origin is skill.
Well, there are limits to that also. Some shots are easier than others, for example. Being defended makes a difference. I think some of the preceding posts sound like a comparison to the Shakespeare/infinite number of monkeys thing - if you take enough shots, eventually they’ll fall in in the right order. I think that’s inaccurate.
On June 22, 1947 Cincinnati’s Ewell Blackwell got close. After no hitting the Braves on June 18 he went 8 and 1/3 innings against Dodgers before giving up a broken bat single in the ninth.
You can say with absolute certainty that Herschel Walker’s 99-yard TD run will never be beat. It is the longest possible distance. It seems that this is the only record (that I can think of) that CAN NOT be beaten. This is different from WILL NOT be beaten: Vander Meer’s feat COULD be equalled, in theory, but it probably never will be, for a variety of reasons. Ditto for Ripken’s record (which I’m not overly impressed with, but that’s another thread). Double ditto for the record for complete games pitched (I forget who holds this one).
So I guess I’d say there are three categories of sports records:
Those that absolutely cannot be broken, such as a 99-yard TD run, or, let’s say for an example, a 1.000 batting average.
Those that can in theory be broken, but won’t, due to changes that have occurred in the culture of the sport, such as a pitcher throwing 30 complete games in a season or a consecutive games played streak.
Those that are just darn impressive, but there’s no reason that they can’t or won’t be broken, other than their impressiveness, such as a 56-game hitting streak or a 20-strikeout game.
“Buy a newspaper in an American city and see your own country disappear”…
It’s amazing how much SportsCenter ingrains Americans with the idea that there are no sports outside the US, unless an American athlete happens to be involved (like Lance Armstrong, for example)
If you want some truly impressive sporting achievements, here’s a brief synopsis of whats been going on in the rest of the world:
-Jim Clark won the Monaco Grand Prix, Le Mans, and the Indy 500 in the same year (3 entirely different racing disciplines- 1967 I believe)
-Denmark won the European (football) championship in 1992, despite only getting into the tournament because Yugoslavia was banned (underdogs don’t come much smaller)
-Pele became the youngest World Cup scorer ever (and his record stands) in 1958, at the age of 17…
-NZ’s Jonah Lomu broke the record for most Rugby World Cup tries at the age of only 23… 23!
-Carl Lewis won the long jump in track and field competitions 65 (!) times in a row… and nobody cared
-Gary Sobers hit six sixes in one over (the cricket equivalent of 6 home runs, not in 6 at bats, but in 6 pitches)
-And why can’t Herschel Walker’s record be beaten? The CFL has 105 yard fields…
With all this talk of Bob Beamon, I just wanted to point out that Mike Powell currently holds the long jump record at 29’ 4 1/2" (8.95m), and has since 1991. I remember watching him break Beamon’s record and about crapping my pants. I always assumed it would be Lewis that accomplished that feat, but Powell proved me (and countless others) wrong.
The main model is that a given player has x% chance of attaining a certain feat (hitting a basketball shot, starting a game, getting a hit during a game, etc.). So you say a player has a (i dunno) 50% chance of getting a hit in a game. Then the odds of getting a 56-hit streak in 56 games are 1 in 2^56. Then you multiply that by how many players have every played, how many 56 games there are in a baseball season, etc. If youre final answer is something like 1 in 10000, then you figure you’ve got something amazing. If it’s 1 in 3, you conclude that it’s just chance.
The problem is that you have no way to assign that initial probability. Wilt Chamberlain’s odds of hitting a shot are huge, and he takes as many shots as he wants. Joe DiMaggio had a much higher chance of getting that streak than (say) Bucky Dent. I just don’t see how you can separate out the component of skill. Some players are better than others. Is Lance Armstrong’s feat really statistically interesting if he is a much better biker than everyone else?
Beamon’s jump is probably in a different kind of model… you awesome the distribution of jumps is more or less normal, and try and figure out how much of an outlier he is. I remember seeing a presentation which showed that Dennis Eckersly was not only a good reliever, but literally a 1 in a million reliever.
Unfortunately this is one of those questions that you could spend weeks defining just what the actual question is, and only then get around to answering it.
One particularly illuminating aspect of the work is that even when presented with the hard, statistical evidence, players and coaches alike refused to believe that hot hand did not exist.
I’m familiar with the study mentioned in the OP via Carl Sagan’s Demon Haunted World. The point of the study is that given enough chances impressive streaks will occur due strictly to chance. If I flip a coin a billion times it is very likely that I will get a streak of, say, 100 heads. Sports streaks are usually not that strictly defined. In batting a streak of .800 (80%) would be amazing.
Here is how someone with a little stronger statistical skills and better baseball knowledge could replicate the study. Let’s assume the following, though a lot of it is wild guessing:
Batters get hits about 1 out of three times at bat.
There are about 300 professional starters in a season.
These guys each get about 1000 at bats in a season.
Instead of having at-bats let’s just roll dice and say that 1’s and 2’s are hits and everything else is an out. That gives you your 1 in 3 odds. Each “player” rolls dice 1000 times and the next “player” rolls 1000 times until you get to 300,000 rolls. It’s not hard to imagine that some pretty impressive streaks will happen purely by chance with that many rolls. (And no cheating, now. A streak that crosses the boundary between one player and another doesn’t count.)
You need to appreciate the power of very large numbers. Given enough chances, the improbable becomes commonplace.
Nope, that’s exactly accurate. Let enough monkeys play in enough games and eventually one will score 100 points.
I think we need Xavier back here to clarify his OP.
**It’s amazing how much SportsCenter ingrains Americans with the idea that there are no sports outside the US, unless an American athlete happens to be involved (like Lance Armstrong, for example) **
Americans care about American sports; so what? Certainly, people in other countries are the same with their own countries’ sports. Let’s not start getting nit-picky about how America-centric America is.
**-And why can’t Herschel Walker’s record be beaten? The CFL has 105 yard fields… **
No, the CFL has 110 yard fields
The record cannot be broken because it is an NFL record. The CFL is a different version of the sport, with different rules. You might as well say that Carl Lewis broke Walker’s record because he regularly ran 100 yards at a time. To make any comparison valid, you have to play by the same rules. That’s why Arena Football QBs aren’t that impressive when they throw for 7 or 8 TD’s in a game, while that would be remarkable for an NFL QB.