Here’s a drinking game for you the next time you tune into any sports broadcast, be it motorsports, golf, rugby, boxing or tennis. Drink every time you hear the word ‘momentum’. You can play the game with pre-shows, post-shows, and even with elections or the stock market. Momentum seems to be a fundamental principal of competition.
But is there even such a thing as momentum?
What is Momentum?
In sports, We use the word ‘momentum’ to describe the concept that winning makes you more likely to win. It’s the idea behind the ‘hot hand‘, ‘hot streaks’, being ‘on a roll’ or ‘in the zone’, or having a ‘run of luck’.
Momentum vs. Inertia
In physics (and, let’s face it, we had the word first), momentum is the product of mass times velocity.
Momentum is also related to force.
These equations tell you that the force an object exerts depends on how much momentum it has and how quickly it loses its momentum. An object going fast hurts more hitting you than the same object going slow. A heavy object will hurt more than a lighter object at the same speed.
Does it seem like this definition doesn’t match with how the term ‘momentum’ is used in sports? There’s a reason for that. The sports ‘momentum’ is more like inertia.
Newton’s First Law states that an object at rest will remain at rest unless acted upon by a force and an object moving in a particular way will keep moving in that same way unless acted upon by a force. The problem is that inertia has acquired a negative connotation associated with resistance to good change, as in the inertia of a big company.
Psychological Momentum (PM)
Physics doesn’t depends on people. Regardless of what you think about gravity, it’ will keep pulling things to the ground at a rate of 9.8 m/s2. There is nothing in the laws of physics that supports the idea that winning begets winning.
This is why academics who study sports use the term ‘psychological momentum’ or PM.
psychological momentum: The positive or negative change in cognition, affect, physiology, and behaviour caused by an event or series of events that affects either the perceptions of the competitors or, perhaps, the quality of performance and the outcome of the competition.Oxford Dictionary of Sports Science & Medicine
In other words, PM changes the way a person thinks or behaves, and it can be positive or negative. But the definition contains a wiggle word when it comes to whether PM affects either the quality of the performance or the outcome of the competition: perhaps.
When I investigated whether the concept of momentum translated to NASCAR, I learned that the majority of people who study this phenomenon don’t agree on whether it exists in whole or part.
Self-efficacy is a person’s belief in his or her ability to do something. There is no doubt that self-efficacy is a requirement for success. People who are told before a test that they are smart and can do well on the test score better than people who are told nothing, or who are told that this is a very hard test and they probably won’t do well. The driver who wins three races in a row goes into the fourth feeling pretty confident.
You can go into a competition feeling unbeatable and still get beaten. Self-efficacy is a necessary condition for good performance, but can you will yourself into winning? Many sports psychologists make that argument. Are they right?
Can You Will yourself to Win?
A lot of studies in the last thirty years have tried to answer this question. They’re difficult studies to do because there are so many variables. Most studies focus on basketball and hockey because there are a large number of teams, a relatively small number of people per team, and a large number of games per season. Football has large numbers of players and few games. Most individual sport players also play small numbers of games each year and a single event like an injury that causes them to miss a competition can skew the data badly.
Most of the studies look at data across a game or a season (or multiple seasons) in search of evidence of streaks that can’t be explained by other factors (like random chance).
But if a team plays three low-ranked opponents in a row, they’re more likely to rack up wins than if they’re playing the division leaders. But were they on a hot streak? Or were they just better than their opponents?
People don’t understand probability. Our own mental prejudices can fool even those of us who intellectually understand probability. The gambling industry makes millions of dollars off people’s misunderstanding of chance.1The gambling industry works because people don’t understand statistics. Some people believe that they must keep playing if they are winning. Conversely, people also convince themselves that bad luck can’t continue indefinitely and keep playing when they are losing. One study claims point spreads are determined not by the anticipated scores, but so that the bettors split 50-50 and thus ensures the house makes money.
I toss a coin nine times and get tails, heads, tails, and then six heads. Is the probability that the tenth flip is heads greater than 50% or less than 50%? Do you say greater because we’re coming off six heads in a row? Do you say less because you know that, on average, there will be equal numbers of heads and tails?
The probability of heads for that tenth flip is still 50%. The laws of probability only hold in groups of very large numbers. Ten is not a very large number.
The Value of Large Numbers
The graph below shows the results for a series of tests of how many times heads came up in 32 coin flips. That is, they flipped a coin 32 times, recorded the number of heads, then repeated it 49,999 more times. They did this on a computer because that’s a total of 1.6 million trials. If you did it in real life, it would take you eighteen and a half days if you could flip one coin a second, 24 hours a day.
The result is the typical normal distribution: most of the results were around the 50/50 mark, but that doesn’t mean there weren’t some streaks. If you look at the raw data, you’ll see that:
- There were 2 times when there were 28/32 heads
- There were 0 times when there were 27/28 heads
- There were 6 times when there were 26/32 heads
The statistics are similar if you look at the no-heads side of the chart.
If you ran enough trials, you would eventually get a case in which all 32 flips were heads (or one in which no flips were heads). The probability for getting no heads or getting all heads is 1 in 86,206.
Even in a perfectly random process, it is possible to get what we might intuit as being an exceptional result.
Randomness and Momentum
Many researchers studying momentum compare winning and losing streaks with the streaks possible given the rules of chance: in other words, if such a thing as momentum exists, it has to create streaks that can’t be produced for some other origin.
Early research done on individual racquet sports shows how difficult this is. 2Iso-Ahola, S., and Mobily, K. 1980. Psychological momentum: A phenomenon and empirical (unobtrusive) validation of its influence in sport competition. Psychol Rep 46: 391-401. The player who won the first game often went on to win the second game and the set; however, if the first and second game split, then neither player had an advantage in terms of winning the set. Does the first situation (Player 1 wins the first game and goes on to win the set) imply the existence of momentum or does it just mean that Player 1 is better than Player 2?
The Research on Momentum
Studies focus on either teams or individuals, and on single games or on series of games. Here are the highlights of the research as I understand it.
- In baseball, Allen St. John analyzed the 2007 World Series and found no impact on the playoff games based on how teams’ records coming into the series.
- The behavioral economist Toby Moskowitz came to the conclusion that “there is a much stronger belief in momentum than is warranted by what we see in the data”. (Moskowitz has a great book called Scorecasting that analyzed all facets of stick-and-ball sports.)
- MIT researchers Johnson, Stimpson and Clark compared football drives that followed big defensive plays to drives that didn’t and found no significant association between a big defensive play and and subsequent offensive performance.
- The canonical paper in the field by Vergin points out that the 1996 MLB season has 1112 winning streaks of 1 to 12 games in length. Under a random-chance scenario, you would expect 1,116 streaks.
- In Feb 1987, the LA Lakers scored 29 consecutive points against the Sacramento Kings; however, even a run that long is possible within a random process.
- Just last month, Schilling analyzed the full NBA 2016-2017 season and found that the lengths of the runs in NBA games are no longer than those that would occur under a random process.
To be fair, there are some studies that claim to a small momentum effect if you control for ability; that is, these papers use complex models that factor in the relative strength of the players. But the effects are small (a few percentage points).
A number of studies show that better competitors are more likely to have longer streaks and more likely to bounce back from failure.
Real games aren’t handicapped for ability, so how much those conclusions apply to reality remains questionable.
Momentum and NASCAR
If football is too complex for researchers, NASCAR is a nightmare.
- It’s not two teams against each other, it’s 35 to 40. You have to track positions.
- The elimination format of the current playoffs introduces chaos. Circumstances that have nothing to do with talent level can eliminate drivers.
- A basketball game has ten players on the court. In a NASCAR race with 40 drivers, there are 200 pit crew members and 600 engineers, crew chiefs and other support personnel for a total of 840 people. All it takes for a race to go into the toilet is for one of those people to screw up.
- An engineer miscalculates the tach and you get a speeding on pit road penalty.
- A pit crew member lets a tire get away.
- A driver gets caught up in someone else’s accident.
- NASCAR also has much more variability than other sports
- track temperature
- track condition,
- what the other thirty-something cars are doing
It’s a really tough thing to study in any systematic way.
But extrapolating what we know about momentum from all of the other studies, the place where a study would be most likely to show the impact of momentum is in laps times. The lap is the analogy to the free throw. Drivers (usually the lead driver in the lead) hits his marks consistently, gets in a rhythm and makes good lap times. He’s ‘in the zone’.
But all that goes out the window when they encounter lap cars. Or they lose the lead. Or there’s an accident.
But NASCAR Has Streaks…
Three drivers dominated in 2018. Kevin Harvick, Kyle Busch and Brad Keselowski each won three races in a row. Winning three races in a row is low probability, and three drivers each winning three races in a row is even smaller. But it’s not non-zero. If we saw this happen three years in a row, it would be different.
But is that due to momentum?
As I’ve discussed before, the quirks of the 2018 season can be attributed to the non-competitiveness of previously dominant Chevy teams. This isn’t momentum: It’s ability.
In NASCAR, technical advantages are far more important than momentum. Penske is onto something this year. Stewart-Haas is struggling. You don’t hear anyone talking about momentum when trying to explain Kevin Harvick’s performance this year with last year.
So… No Momentum in NASCAR?
From my years of listening to drivers on the radio during races (one of the great features of NASCAR), I think I can make a case for negative momentum. There are some drivers who enter into a downward spiral at the first negative event, like a bad pit stop or a minor accident. They talk themselves into a bad finish. It happens to crew chiefs, pit teams and everyone else associated with a team, too. Some people make a mistake and it inspires them to do better. Others are haunted by the mistake.
What’s in people’s heads determines a certain amount of the race . That’s why many pit crew coaches combine physical and psychological training, and drivers hire sports psychologists to help them manage their emotions.
You can’t psych yourself into the championship, but you can definitely psych yourself out of it.
So Why Do We Hear So Much About Momentum Then?
Sportscasters, players, coaches and fans talk about ‘momentum’ as if there was no doubt of its existence because we’ve done it so much we’ve talked ourselves into believing it. As I noted with respect to whether people would notice cars being 10-15 mph slower on intermediate tracks, what you want to believe determines a large part of what you believe. The more something is repeated, the more people repeat it, often without bothering to think aboot whether it’s true.
Be careful what you put into your head because once it’s in there, it’s hard to get out. Scientists constantly questions their assumptions and predispositions, especially in research that involves peoples.
It’s not a bad thing for all of us to do.