Will the Blue Jays make the playoffs? A statistics professor discusses the odds
After an impressive 11-game winning streak that thrilled longtime fans and attracted new ones, the Toronto Blue Jays faced the Yankees on August 14 with great expectations from the stands.
They lost the game but is it possible to predict how the rest of the season will go?.
Jeffrey Rosenthal is an award-winning professor in the department of statistics renowned for his ability to teach quantitiative reasoning to students who lack confidence in math and the bestselling author of Struck by Lightning: The Curious World of Probabilities. U of T News spoke with Rosenthal about the likelihood of the Jays making the playoffs and other useful applications of statistics in the classroom and beyond.
Are you a baseball fan?
Well, I'm a "fair-weather fan" – I don't follow baseball as closely as I used to, but when the Blue Jays are doing well, then I get excited along with everyone else. I can still remember the thrill I felt when Joe Carter hit his World-Series-winning home run back in 1993.
Is it possible to calculate the likelihood of Blue Jays making the playoffs?
Yes and no. There are so many variables that it's impossible to take them all into account. But by making various simplifying assumptions, you can at least get a "rough" idea of what the odds are. But even rough estimates can be useful and insightful.
It's kind of like repeatedly flipping a coin. You might get a few heads in a row, but you probably won't get too many heads before tails comes up. With the Jays, even if they have, say, a 60 per cent chance of winning each game (which is quite good), the probability of winning the next nine games in a row is only one per cent, i.e. very unlikely. That's why long winning streaks are very rare, even for good teams. And indeed, on Friday, the Jays finally lost, after 11 wins in a row.
So, what are the lottery jackpot odds?
Since the success of my general-interest book, I am often contacted by the media to calculate all sorts of probabilities, from lotteries to opinion polls to crime statistics and games and contests and more.
In some cases you can be very precise, like for lottery jackpot odds or simple gambling games. In other cases you have to be more creative and make more assumptions as you go, like for homicides or the Blue Jays.
For the Lotto 6/49, to win the jackpot you have to choose the same six numbers between one and 49 as the lottery people do. The problem is, there are nearly fourteen million different ways of choosing those six numbers. So, your probability of winning the jackpot with one Lotto 6/49 ticket is about one chance in fourteen million – extremely unlikely.
That's why I've never bought a ticket.
What are some other applications of statistics in the real world?
There are so many, which is what makes the subject so interesting. In addition to sports and gambling, statistics are important for understanding crime rates and trends, demographic changes, medical studies, genetics and diseases, polls and marketing, physics and chemistry, financial markets, and so much more.
Whenever we have data available, whether from an experiment or a survey or commercial transactions or computer usage or anything else, we can use statistical analysis to understand it better and make future predictions.
How does understanding statistics make students more informed citizens?
Much of our news and political debate involves quantitative items like debts and deficits, public opinion, demographic predictions, revenues and costs, and so on.
Some citizens aren't equipped to understand such things, which lessens our ability to make intelligent decisions and to hold our leaders to account.
The more numerically literate we become as a society, the better we can navigate the demands of our modern world. In fact, I teach a course, STA201: Why Numbers Matter, which is designed precisely to get non-science students thinking about these sorts of issues and becoming more comfortable with them. (Read more about the course.)
How do you do that?
I've tried to come up with examples of quantitative reasoning which aren't too difficult, but which are fun/different/engaging. For example: how objects and weights (from Godzilla to Mini-Me) scale with differing sizes; gambling probabilities; exponential growth of anything from investments to diseases or even the incomes from Jane Austen novels; the mathematics of musical notes; margins of error in public opinion polls; Fibonacci sequences and the Golden Ratio; and more. The student response has been very positive, so that's great.
(See Professor Rosenthal discuss the probability of the Jays making the playoffs on CTV News.)