Suppose you are facing a willpower challenge: You're a regular drinker but you've decided that you will no longer drink every day, but only on Friday evenings from now on.

Now, if I told you, "There is only one way to succeed in this goal, but there are many ways to fail," you could probably make sense of that: The only way to succeed is to make it all the way to Friday without having a drink. But there are many ways to fail. You could have a drink on Saturday. You could have a drink on Sunday. You could have a drink on Monday. (You get the point.)

But here's the thing. I if I said to you: "There is only one way to fail, and there are many ways to succeed," you could probably make sense of that, too: The only way to fail is to have an alcoholic beverage at the wrong time. But when facing temptation, there are many ways to succeed. There are many alternative activities you can do instead of drinking. You can take a bath. You can play basketball. You can have a bottle of sparkling water. You can go for a walk. (You get the point.)

So which is it?

Is there just one way to succeed and many ways to fail? Or is there one way to fail with many ways to succeed?

The truth is, neither of those claims is true or false full stop, but each becomes true or false depending on how you tell the story—how you lump things together and split them apart.

So it's entirely up to you how you frame your challenge.

But here’s the thing: **Seeing many paths to success increases optimism**

In psychology, there is a phenomenon known as the subadditivity effect. It's the tendency to judge the probability of the whole to be less than the probabilities of its parts. So, when we estimate our chances of success in reaching a goal, we can do it two ways. We can consider just one success condition:

A. I will resist drinking alcohol when next tempted.

Or we can consider many success conditions:

B. I will go for a walk instead of drinking.

C. I will take a bath instead of drinking.

D. I will talk with a friend instead of drinking.

E. I will find some other activity to do instead of drinking.

Now suppose we estimate probabilities (P) for A through E. If the subadditivity effect holds, then our estimates will bear this relationship to each other:

P(A) < P(B) + P(C) + P(D) + P(E)

For instance, we might assign probabilities like this:

P(A) = .6; P(B) = .2; P(C) = .2; P(D) = .2; P(E) = .2

So we might give ourselves a 60 percent chance of success if we consider only *one* success condition, but an 80 percent chance if we consider *multiple* success conditions.

It’s important that we don’t explicitly *make* all of the above estimates and compare them in a live situation. If we do, we might break the spell—we might notice the inequality and start revising our estimates so they add up as equal. But we don’t want that. We want to take advantage of the fact that our vague and fuzzy probability judgments will be higher if we don’t look at things too closely.

What we’re looking for is a way to increase our optimism. To do that, it’s sufficient to take our single success condition and break it into multiple success conditions, while doing the opposite with our conditions for failure.

And the reason we want to increase our optimism is because **increasing optimism increases willpower.**

You might protest that using this tactic will produce *false* optimism, that we're *tricking* ourselves into being more optimistic than we objectively should be. But the thing about optimistic probability judgments is that, when you mix them with willpower, they can become self-fulfilling prophecies. That's not just positive thinking mumbo-jumbo. It's game theory.

George Ainslie showed us in* Breakdown of Will* how our tendency to overcome a temptation at one point in time depends on how likely we think it is that we will also overcome similar temptations in the future. If we think it more likely that we will overcome those temptations in the future, then we will be more likely to overcome the temptation now. *(I explain this dynamic in much more detail here.)*

So what's the bottom line?

Well, if you want your willpower to be stronger, it will help if you can raise your expectation of success. If you want to raise your expectation of success, you can take advantage of the subadditivity effect for probability judgments. And that means you should frame your challenge so there is just *one way to fail*, and there are *many ways to succeed*.

Now go ahead, pick your most pressing willpower challenge. Remind yourself that there is only one way to fail. And then ask yourself, "What are all the ways I can succeed?"

See what you come up with.