Segments in this Video

Power of Prediction (05:03)


Mathematics and data combine to help bring innovation and success. Talithia Williams visits the Orange County Fair to see if the "wisdom of the crowds" theory is true. Crowd estimates are more accurate than the vast majority of individual guesses.

History of Predictions (02:15)

Romans studied the flights and cries of birds while Chinese read the results of "oracle" bones. Statistics is a framework to provide the rules and language by which we perform science; it predicts the likelihood of future occurrences.

Chance and Percentage (04:09)

Casinos profit using probability. Gerolamo Cardano's theory became the "law of large numbers." Divide the number of shots made by the number of shots taken.

History of Probability Theory (03:10)

Blaise Pascal and Pierre de Fermat exchange letters about imaging possible future outcomes of an incomplete game. Individuals can calculate with precision the likelihood of events occurring.

Weather Prediction (06:34)

The National Weather Service releases weather balloons daily, recording data every ten meters. Computers and scientists analyze the information at the National Center for Environmental Prediction in College Park, Maryland using numerical forecasting. The apply fluid and thermodynamic equations to evolve the atmosphere through time.

Making Accurate Predictions (06:11)

Scientific theories make predictions that must be tested before they are proved. Ronald A. Fisher establishes guidelines for designing experiments using statistics and probability as a judging method. "P-hacking" occurs when researchers massage data to obtain a p-value of .05 or under.

2016 Presidential Election (04:29)

Although Hilary Clinton was the overwhelming favorite according to pollsters, Donald Trump won. Use randomness to find a representative sample. The margin of error is the maximum amount the result from the sample can be expected to differ from the whole population.

Possible Polling Errors (05:08)

Question wording, emotions of respondents, and future behaviors cannot be calculated. Response rates have declined due to caller I.D. and answering machines. Trump's victory relied on few votes spread across Pennsylvania, Wisconsin, and Michigan.

Baseball Statistics (03:12)

Sports analytics uses predictive models to improve a team's performance. In "Moneyball," Brad Pitt portrays Billy Beane who innovated the practice of using stats to choose players.

Search and Rescue (06:28)

The U.S. Coast Guard's Sector Boston Command Center initiates a search for a paddleboat that has gone missing. Thomas Bayes begins with initial probability based on what is known. The Rescue Optimal Planning System (SAROPS) creates probabilities based on water drift and other known information.

Artificial Intelligence (05:19)

In "deep learning," trial and error guides the computer's knowledge. Stanford University conducts an experiment to see if a machine-learning algorithm can diagnose skin cancer as well as a board-certified dermatologist. Probability and prediction using mathematical analysis can help guide the future.

Credits; Prediction by the Number (00:46)

Credits; Prediction by the Number

For additional digital leasing and purchase options contact a media consultant at 800-257-5126
(press option 3) or

Prediction by the Numbers

3-Year Streaming Price: $169.95



Predictions underlie nearly every aspect of our lives, from sports, politics, and medical decisions to the morning commute. With the explosion of digital technology, the internet, and “big data,” the science of forecasting is flourishing. But why do some predictions succeed spectacularly while others fail abysmally? And how can we find meaningful patterns amidst chaos and uncertainty? From the glitz of casinos and TV game shows to the life-and-death stakes of storm forecasts and the flaws of opinion polls that can swing an election, Prediction by the Numbers explores stories of statistics in action. Yet advances in machine learning and big data models that increasingly rule our lives are also posing big, disturbing questions. How much should we trust predictions made by algorithms when we don’t understand how they arrive at them? And how far ahead can we really forecast?

Length: 53 minutes

Item#: BVL169058

Copyright date: ©2018

Closed Captioned

Performance Rights

Prices include public performance rights.

Not available to Home Video, Dealer and Publisher customers.

Only available in USA and Canada.