2026-09-02
Agenda
Agenda
The Expected Utility of an outcome is the value of that outcome multiplied by its probability:
The current cash jackpot for the “Mega Millions” lottery is $68.9 million and the odds of winning are 1 in 290,472,336. A ticket costs $5. Is that a good deal?
The probability of winning is:
\[\frac{1}{290,472,336} = 0.00000000344265\]
So the expected utility is about 24¢:
\[68,900,000 \times 0.00000000344265 \approx .237\]
Paying $5 for an expected utility of 24¢ is a bad deal.
To break even on a $5 ticket, the jackpot needs to be about $1.45 billion:
\[1,452,369,541 \times 0.00000000344265 \approx 5.00\]
(This ignores taxes, the chance of splitting the jackpot with other winners, the chance of winning smaller prizes, and the diminishing marginal utility of money.)
You should choose the action that maximizes your expected utility. To figure out what action this is, you need to consider:
| win 50% |
lose 50% |
Expected Utility | |
|---|---|---|---|
| fight it | 0 years | -15 years | ? |
| take the plea | -5 years | -5 years | ? |
You have been arrested and charged with a crime that carries a 15 year sentence. You can take a plea deal and get that reduced to 5 years, or you can fight it in court. Your lawyer tells you your chances in court are a toss up. Calculate the expected utilities. What should you do?
You want to watch the 1967 live-action film The Gnome-Mobile. You can stream it for $4. Or you can download a pirated copy. There is a one in a million chance you will get caught, and you live in a country with draconian anti-piracy laws, so if you do get caught, you will be fined $100,000.
Create a decision matrix and calculate the expected utility of each possible action. What should you do?

| God (\(p_1\)) | No God (\(p_2\)) | Expected Utility | |
|---|---|---|---|
| Believe | \(\infty\) | \(f_1\) | \(\infty\) |
| Don’t | \(f_2\) | \(f_3\) | \(f_4\) |
Argument
| God (\(p_1\)) | No God (\(p_2\)) | Expected Utility | |
|---|---|---|---|
| Believe | \(\infty\) | \(f_1\) | \(\infty\) |
| Don’t | \(f_2\) | \(f_3\) | \(f_4\) |
Objections:
A yacht would make me happier. A second would also increase my happiness, but by a bit less.
Let \(y\) be the boost from the first yacht, and suppose each additional yacht boosts my happiness by half as much as the one before:
\[y + \frac{y}{2} + \frac{y}{4} + \frac{y}{8} + \frac{y}{16} ... = 2y\]
The benefit of infinitely many yachts is not \(\infty\), but just \(2y\).
| God (\(p_1\)) | No God (\(p_2\)) | Expected Utility | |
|---|---|---|---|
| Believe | \(\infty\) | \(f_1\) | \(\infty\) |
| Don’t | \(f_2\) | \(f_3\) | \(f_4\) |
How do we assign probabilities \(p_1\) and \(p_2\)?
| God (\(p_1\)) | No God (\(p_2\)) | Expected Utility | |
|---|---|---|---|
| Believe and Pray | \(\infty\) | \(f_1\) | \(\infty\) |
| Believe and Don’t Pray | \(f_2\) | \(f_3\) | \(f_4\) |
| Don’t Believe or Pray | \(f_5\) | \(f_6\) | \(f_7\) |
| Generous God 25% |
Stingy God 25% |
No God 50% |
Expected Utility | |
|---|---|---|---|---|
| Believe | \(\infty\) | \(f_1\) | \(f_2\) | \(\infty\) |
| Don’t Believe | \(f_3\) | \(f_4\) | \(f_5\) | \(f_6\) |
| God1 20% |
God2 20% |
God3 20% |
No God 40% |
Expected Utility | |
|---|---|---|---|---|---|
| Believe in God1 | \(\infty\) | \(f_1\) | \(f_2\) | \(f_3\) | \(\infty\) |
| Believe in God2 | \(f_4\) | \(\infty\) | \(f_5\) | \(f_6\) | \(\infty\) |
| Believe in God3 | \(f_7\) | \(f_8\) | \(\infty\) | \(f_9\) | \(\infty\) |
| Don’t Believe | \(f_{10}\) | \(f_{11}\) | \(f_{12}\) | \(f_{13}\) | \(f_{14}\) |
The book argues that, even though belief isn’t voluntary, you can still adopt the strategy of trying to get yourself to believe: go to church, avoid spending time with atheists, pray, read religious texts, etc.
Of course, if you try, there is a chance you will succeed and a chance you will fail:
| God exists and you succeed 25% |
God exists and you fail 25% |
God doesn’t exist 50% |
Expected Utility | |
|---|---|---|---|---|
| Try to Believe | \(\infty\) | \(f_1\) | \(f_2\) | \(\infty\) |
What is the expected utility if you don’t try?
Agenda
This is not an essay. This is a thinking-through-writing exercise. I want you to write about whatever it is that you find most puzzling or confusing about the Argument for Betting on God. Try to put your puzzlement or confusion into words, as if you were explaining it to a classmate. If there is something you are struggling to understand, think about what it might mean or how it might work, and explain those different possibilities, and think about which, if any, seems promising.
Think about whatever it is that you find most puzzling or confusing about the argument. What do you find puzzling or confusing about it? Are there ways you might try to understand it? What are those? Why are they not quite working out?
Turn to someone, and share your thoughts.
This is not an essay. This is a thinking-through-writing exercise. I want you to write about whatever it is that you find most puzzling or confusing about the argument. Try to put your puzzlement or confusion into words, as if you were explaining it to a classmate. If there is something you are struggling to understand, think about what it might mean or how it might work, and explain those different possibilities, and think about which, if any, seems promising.
Swap papers with someone. Read each other’s papers, then discuss.