Subscribe to Annie's Substack

The Science and Strategy of Decision Making with Annie Duke, Decision Strategist and Speaker,

Do you and your team want to get better at making decisions that are not the status quo? Would you like your team to take the right risks and encourage a culture that learns from success and failure? Did you know that being too results-oriented can get in the way of making the highest equity decisions? If so, this is the podcast for you. Annie Duke, professional speaker, and Decision Strategist will share her research and help you learn how to create a process that helps you make better decisions while encouraging learning and risk-taking. She helps people have uncertainty with confidence.
Brain Hacks 4 Leadership


Do you and your team want to get better at making decisions that are not the status quo? Would you like your team to take the right risks and encourage a culture that learns from success and failure? Did you know that being too results-oriented can get in the way of making the highest equity decisions? If so, this is the podcast for you. Annie Duke, professional speaker, and Decision Strategist will share her research and help you learn how to create a process that helps you make better decisions while encouraging learning and risk-taking. She helps people have uncertainty with confidence.

Annie Duke is a woman who has leveraged her expertise in the science of smart decision making to excel at pursuits as varied as championship poker to public speaking. For two decades, Annie was one of the top poker players in the world. In 2004, she bested a field of 234 players to win her first World Series of Poker (WSOP) bracelet. The same year, she triumphed in the $2 million winner-take-all, invitation-only WSOP Tournament of Champions. In 2010, she won the prestigious NBC National Heads-Up Poker Championship. Prior to becoming a professional poker player, Annie was awarded the National Science Foundation Fellowship. Because of this fellowship, she studied Cognitive Psychology at the University of Pennsylvania.

Well, Annie, thank you so much for being on the podcast today, really looking forward to hearing about some of the work that you’re going and starting to think about now around resulting and how it gets in the way of learning. That’s what your last book was really about, and also to the ways that we make decisions more defensively that can impact our productivity, something that is near and dear to the hearts of any leaders.

Thanks for having me on. I’m excited to be here.

Great. Well, tell us first what does resulting mean?

Yeah, so resulting is really something that I take a deep dive on in my book in terms of how it gets in the way of learning, and basically resulting is when we make too tight a connection between the quality of an outcome and the quality of the decision that preceded it. The issue is really we have an outcome, we have the way that the future turned out and that future can end up that way for a variety of reasons, and trying to work backward from the outcome to the quality of the decision that preceded it is really hard because these things are relatively loosely correlated. So, to get this into a concrete example that I think that people will be able to feel pretty deeply and I think that it really gets people to understand what resulting is, I want to take us back to the 2015 Super Bowl where the Seahawks are on the 1-yard line of the New England Patriots.

It’s second down, there’s one timeout left and the Seahawks are down by four with only 26 seconds left in the whole game. So, this is obviously a super important movement because if Pete Carroll can call a play that gets a touchdown, obviously they’re down by four, that’s gonna put them up by two. Let’s assume they make the field goal, but even if they don’t they’re up by two and it’s very unlikely that the Patriots are gonna have any time to get all the way down the field to score again so this is for the game. So, if they can score here this is gonna win them the game. So, with 26 seconds left they’ve got a running back, they’re only on the 1-yard line, remember. They’ve got a running back named Marshawn Lynch who’s an amazing short-yardage running back and everybody’s kind of expecting Pete Carroll to call a handoff so that Marshawn Lynch can just sort of barrel through, hopefully, the defensive line of the Patriots.

Instead, Pete Carroll does something super unexpected which is he calls for a pass play. So, Russell Wilson, the quarterback of the Seahawks, passes the ball and the New England Patriot’s Malcolm Butler intercepts the ball in the end zone. So, let’s agree this is a really, really disastrous result. This is a very, very bad outcome quality. So, obviously the Seahawks lose the game there and when you listen to the in-game call, in other words, the announcers, Chris Collinsworth, he’s really pretty brutal about the whole thing, really saying he can’t believe this call, this is the worst call that he’s ever seen. Then when you look at the headlines the next day for most of the pundits who are doing a nice job here of Monday morning quarterbacking, they are not disagreeing.

So, it’s a lot of this is the most horrible call in Super Bowl history repeated throughout all the headlines and then actually USA Today managed to say it was the worst call in football history period. Now, when Pete Carroll was actually asked about this on the Today Show, what he said was that he would agree that it was the worst result of a call ever, and I think that that’s a really important distinction. So, what we know is that the result of this call was terrible but does that actually mean that the call itself was terrible? The answer is not necessarily. We’d have to go back and we’d have to really look at the mathematics of that call. Just super quickly, the chances of an interception there are somewhere between 1% and 2%.

So, once you know that fact that 99% of the time either the ball’s gonna be caught for a touchdown or it’s just gonna be dropped, which will stop the clock and allow them to go for another play, once you know that fact of what’s the interception rate there I think we can agree that probably declaring it to be the worst call in the history of football is probably a little bit of an over exaggeration. I think that we can see how much the quality of the outcome, which was so bad, is yanking us around in our ability to see through to what the quality of the decision is and we can do that pretty easily by just imagining this.

Imagine that he calls the pass play and the ball is actually caught for a touchdown in the end zone, and I do this in front of audiences all the time. I ask them to actually just close their eyes and imagine this, and everybody can feel it immediately. What are the headlines gonna look like the next day? Everybody just says immediately it’s brilliant, it was one of the most brilliant calls in football history, and we can actually see this because Philadelphia this year actually called a similarly unusual play but in this case, it worked out and everybody hailed the coach of the Eagles, Doug Pederson, to be a genius. So, there’s clearly something weird going on here, right?

You have the decision. The decision’s the decision regardless of which future happens to occur. There’s a variety of futures that can occur, and yet we can feel it very deeply that when the outcome is really bad we think the decision is really back, and when the outcome is really good we immediately think that the decision is really good and yet based on one time that shouldn’t actually affect in reality what we think about that decision, and that’s just because in life the quality of an outcome and the quality of the decision are only loosely correlated. I can give you a super example. I’ve run a red light in my life, not gotten a ticket. I haven’t gotten in an accident. I’m not trying to re-do that. I don’t think that that was a good decision because the outcome happened to be good. You can see how this is a really big problem for learning.

So, what I’ve been thinking about recently is how is this a problem when we’re results oriented? This is a very common thing that people say, we’re results-oriented. So, I’ve been thinking about that because we have this problem which is well what do you mean by results? Because clearly, you can see what happened with the Seahawks, that people said oh here’s this result and I’m gonna make this results-oriented decision and what ends up happening is that Pete Carroll is getting excoriated for a decision that was actually mathematically pretty good. By the way, if anybody wants to geek out on it they can go look at Benjamin Morris on 538 to see some really good analysis of the play.

But, now what I’ve been thinking about is well what does that do to people’s decision making if they know that they’re gonna be evaluated based on the result? If they know that what’s gonna happen is there’s gonna be a collective outcry about the quality of the decision if you have a bad result? And, what are you doing to people in that situation? What I’ve been realizing is you really force people into a very defensive crouch when they’re making decisions. In other words, you’re gonna get it so that they’re not making necessary decisions that are going to create the biggest value. They’re going to make decisions that are going to defend against being judged for a bad outcome.

So, if we go back to the Pete Carroll situation, I can give you an example of how this might happen and then we can take a deeper dive. If you go look at Benjamin Morris, there’s a very good argument that the highest return play is the pass, and obviously, Pete Carroll felt that way. He felt like the highest return play was gonna be a pass. Now, the lower return play would be a handoff to Marshawn Lynch. This also happens to be the status quo play. It also happens to be the expected play. This is what everybody expected him to do, but it’s actually a lower return. You can imagine that if a coach knows that when you do this unexpected thing when you do the pass play, which happens to be higher return, that if it doesn’t work out you’re gonna be yelled at, then you can imagine that as a defensive maneuver they might just hand it off to Marshawn Lynch. Why? Well, we can do that thought experiment as well.

Let’s imagine that you hand the ball off to Marshawn Lynch and Marshawn Lynch just happens to fail to score. What do the headlines look like the next day? I’ve done this thought experiment with people that I work with and they all say the same thing, oh the Patriots were too good. So, now they’re not blaming Pete Carroll’s decision making any more because it’s just the status quo choice, but it happens to be a lower equity choice. It has a lower return. It’s just that it happens to be a choice that defends against people yelling at you for the choice because it’s a choice that’s more agreed upon. People have talked for a long time about how in the NFL they’ve been very, very slow to come around to these more creative fourth down plays like going for it on fourth down, for example, and if you wanna know why even though the math has been out there, I think that this problem with defensive decision making is really what tells you what it is.

If a coach whose job is on the line knows that they’re gonna be evaluated on the result of one play or one game or one season where you don’t have enough data to really say anything about what the decision process is, that they’re gonna be just evaluated solely on their short-term result then what are they gonna do? They’re gonna make decisions that defend against being evaluated on those short-term results. In other words, they’re gonna stick with the status quo.

Which is definitely not something that you want when you’re looking at innovation, trying to compete or beat your competition, right?

Well, yeah, so what you’re really trying to encourage in the people that work with you is you want them to be trying to find out what the truth of the matter is, what it is that can be learned from outcomes good or bad, what innovations might be there to be had. You’re asking them to figure out what the highest return decision is gonna be. What you don’t want them to do is figure out what the decision is the least likely to incur somebody’s wrath is gonna be, and there are all sorts of ways in which you, through evaluating people on short-term results, you can force them into defensive crouches and that defensive crouch can exhibit itself in a variety of ways. So, one of the ways and this relates back to what we were just talking about, is that people will very often try to make a decision such that if there’s a bad outcome it is a reasonable explanation. It appears reasonable that it could be due to luck. In other words, to factors that are outside of the person’s control.

Here’s a simple example that’s not from business and then we can talk about how this might manifest itself in a business situation, but I think that this will help people to see very clearly where the problem comes from, and this example comes from Kevin Zollman, who’s a game theorist actually at Carnegie Mellon. Let’s say that you as a leader offer somebody a plate that has two cookies on it and one of the cookies is very big and one of the cookies is small, and the person feels this, that they know if you choose the really big cookie that they will be judged by the leader for being greedy, for being super greedy and a pig. And, if they choose the small cookie they’re gonna be judged as a ridiculous virtue signaler. So, no one situation. If the outcome is I’ve chosen the big cookie, then I’m gonna be judged for that. If the outcome is I’ve chosen a small cookie, then I will be judged for that. So, the solution there is for them to flip a coin. If you flip a coin, now notice that the judgment goes away.

Why is the judgment going away there? Well, it wasn’t in my control. I know I ended up with the big cookie, but you can’t say that I was a pig because I flipped a coin. That’s how I ended up with the big cookie. Or, I know I ended up with the small cookie and you can’t say that I’m a ridiculous virtue signaler because how could I have controlled it? I flipped a coin. There was nothing I could do. So, this is one of the ways that people will get into a defensive crouch. So, they’re not choosing the cookies based on what their values are or what they think the best result for them is or what they think the best result for you is. Instead, they’re just sort of deflecting and saying well let me pick up a coin and I’m just gonna leave it to chance. Obviously, this is not something that you want to have happened.

There are different ways that you can do this, by the way, that don’t involve coin flipping. Let me ask you this, have you ever been in a situation where say someone that you’re on a date with is trying to figure out what restaurant to go to and you insist that they choose?


Right, so why are you doing that? Well, because then it’s not on you if it’s not a good experience.


Shouldn’t you be choosing? It’s very rare that you don’t actually have a preference. You’re not going in and examining what your preference is and balancing that with what you think their preferences might be and trying to come up with the most rational choice for the both of you, which is gonna balance out both of your happiness. Instead, you essentially pick up a coin to flip by saying no, no, no, you choose.

Yeah, happens a lot.

That could take the responsibility off of you and you can see how that happens in the workplace a lot, that sort of you choose. No, you choose. It’s your decision. We’re offloading decisions and we’re not really thinking clearly about what our own preferences are, what we think is gonna balance out to everybody’s preferences in the room because by saying you choose now whatever the result is it’s like well it’s not on me, I didn’t choose. Notice that saying you choose is a choice in itself. It’s interesting because obviously picking up the coin and choosing to flip is still a choice in itself, we just don’t think about that because we’re just like oh, okay, now luck is a reasonable explanation so we’re gonna go with that. And we can see this in the Pete Carroll example which is why is it that as the collective public we feel so much better if he fails by handing the ball off and having the run play happen? Which remember, that’s the lower return play. It’s mathematically not as good a play.

So, why is it that we feel so much better about that? Why is it that if he had made that decision we would all be like poor Pete Carroll, nothing he could do? And the reason is that that choice of hanging the ball off is the consensus choice. It has a tremendous amount of consensus around it just like not going through red lights has a lot of consensus around it. It’s a decision that people, they’ve already made an evaluation of whether it’s a good decision or a bad decision, it’s just the status quo choice and when you have a status quo choice, people say well clearly that’s a good choice because that’s been agreed to as part of the collective wisdom, and so therefore if there’s a bad outcome from it must just be due to luck because that clearly was the best choice because we’ve all agreed to it.

This is the first thing that we see happen when we put people into a defensive crouch is that they tend to go with status quo choices. They tend to go with what the legacy choice is and not poke around to see if there’s a better way because if they go with the legacy choice and it doesn’t happen to work out, well there’s already consensus around that. So, they know that people are gonna shrug and say well what could you do? Because this is the way it’s always been done, but what you really want from the people that you’re working with is the willingness to poke around, to push the boundaries, to see if there isn’t a better way. This is how progress gets made, and this is certainly how we defend against people coming in who are creative, people coming in who are trying different things. If we continue with the status quo then what happens is we obviously get caught unprepared for when there are big changes and we’re not pushing the boundaries and we’re not moving things forward. We’re not really looking around for that pass play anymore.

So, what is it that a leader can do personally to look at their own decisions and test to make sure that they’re not just going for status quo and that we’re really making a decision that is the best decision?

I think there’s a variety of ways to do that, and I think this actually ties back into how do you get the people that you’re working with to not be in these defensive crouches and defending against bad decisions? I think that these are tied together because what you want to be doing is always examining to try to make sure that you have the best process going as opposed to really signaling that you’re results-oriented. So, how can you do that? There are a few ways that you can do this. Let’s say that an outcome has already happened. So, way number one is to really work with the group when somebody is coming in to try to do the post-mortem on an outcome, to have it look more like they’re doing it before the outcome occurred to try to reinforce that there’s a very loose relationship between outcome quality and decision quality. So, how can you do that? Well, when you’re describing a decision to people or when anybody is coming in and describing a decision and asking for advice from a group that you’re working with in order to try to figure out whether you would want to repeat that decision or what parts of the decision might be due to luck or what might be due to skill, have them describe the decision that they were struggling with only to the point that they need help.

So, in other words, if you’re in sales describe what the relationship is with the person that you were negotiating with. Obviously in this particular case you know what the product is because you’re in the same group with the people, so really work with this is what I knew about the person that I was negotiating with, this is what my past experiences with them had been, this is what my goals were in the negotiations so I went in and we’re negotiating and we get to this point where he says X and I needed to know what to respond there, and then stop. Our tendency is to want to tell the rest of the story. Our tendency is to want to say and so I did this so this was my choice, and then this was his response, and then so on and so forth. Notice what you’ve done there is you’ve let the person know first of all what your choice is, which you don’t want to because they’re gonna try to in general, the first thing is you don’t want to let them know your choice, particularly as a leader because they’ll try to make your choice make sense.

So, you’ve now infected them with a bias which is they’re gonna try to attempt to make your choice make sense because that’s the sort of thing that you do when you’re in a room with someone who’s in a leadership role. But then once you’ve actually told them what the response was of the other person, let’s say that the response of the other person was good, now we’re in the Pete Carroll problem. They’re gonna try to make the good response make sense. So, they’re gonna talk about your decision as if it were a good decision because you got a good response from the person that you were negotiating with. If the response of the person was bad, then they’re now gonna try to figure out what was wrong with your decision.

It’s gonna change the way that they view what you actually decided to do. What you want to do is essentially quarantine all of that off from the group and really describe the decision problem up until the point that you have to make a choice and no further, and then start to elicit advice from people in the room. What you’re doing is essentially you’re getting them to think about the problem as a prospective decision as opposed to a retrospective analysis. That’s one thing that you can do. Now, sometimes the decision is already known. In that case, you have to be asking really good questions. So, if everybody in the group already knows what the outcome was, the outcome is already known, the result is already known, what you can do is say okay, we know that the outcome was say bad so you’ve told me all of the reasons why you think that this was a bad decision because the outcome was bad, but now I want everybody to really argue the other side.

How are we gonna do that? So, let’s go back through this decision and imagine it was a good result. Why do we think we would have gotten a good result from this decision? So, now you’re just asking them to do what’s called a counterfactual, which is imagine that things had been different. So, just say I get that the ball was intercepted but let’s imagine that the ball was caught for a touchdown. Now, let’s go back and analyze that decision in light of the ball being caught for a touchdown and just see what happens there. So, notice that you’re allowing for the fact that there were many different futures that could occur. One future happened to occur, but let’s think about what would have happened if other things had happened and get people to really start working with that, and what will happen is that they’ll start to focus more on the good parts of the decision.

You can actually query particular things. Okay, this is all great so what do we think was due to luck, though? We know that some stuff was due to luck. Let’s try to identify what the things were that were due to luck. What do we really think was due to the decision quality? So, that’s another thing that you can do with a group. Another exercise that you can do with the group is to divide the group into two and with one group go and tell them the decision, so this would be if the outcome is not known by the group, with one group go sequester them off. Do break out groups. You’ve got two breakout groups. One group, you actually describe the decision and you give it a good outcome.

In the other room, you describe the decision and you give it a bad outcome and now you have the groups come together and discuss their analysis of the decision so that you can get them to see how much the outcome quality is really affecting the decision quality, and now you can interpolate between the two analytics processes to try to get down into the truth. Now, what this does is it helps you with two problems at once. It’s a little killing two birds with one stone. It’s certainly helping you analyzing your decision process. So, it’s helping you with that retrospective problem of, you have an outcome you’re trying to figure out whether a decision was good. How do we now get people to give us good advice on the decision quality?

So, that’s certainly helpful and that’s gonna help your process more because you’re gonna be better at identifying the luck elements and the skill elements, but what it’s also doing is it’s signaling to the whole team that you’re interested in process, that you’re not so interested in the way that things turn out. You’re really interested in whether the decision itself was a good one, and that’s what you care about and that you’re gonna be poking around at whether something was status quo or not. You’re gonna be poking around at what was luck, what was skill, what could we have done better? How do we think that we could have improved this? You aren’t that kind of results-oriented leader, at least not when we mean results in the short run based on just a few outcomes.

When you start signaling that you want to poke around into the decision process and that the quality of the outcome is not so important to you because you trust those good quality outcomes will come from good quality decisions, what you’ve done is you’ve freed that group up. You’ve sent a really big signal to them that what you care about is really pushing the boundaries of what a good decision looks like.

That’s a great example. I can see how that could really not only impact the decisions you’re making in that group but to your point earlier you’re gonna change the culture and you’re gonna change people’s actions by showing what you reward.

Yes, exactly. Here’s another way that you can signal this. In a room, if you create a really good scenario plan which involves here’s the decision under consideration. This is what we’re thinking about. Let’s identify what we think the possible outcomes of this decision are. So, let’s take a super simple example. So, I actually did this with a nonprofit that I work with called After-School All-Stars. They are involved in a lot of grant writing. There’s a very simple example of how you can do this and how you can do scenario planning. Obviously, when they write a grant, let’s simplify this to there’s two outcomes. They either get the grant or they don’t. What they had done before was there was some award amount for the grant. Let’s say they were applying for a $20,000 grant or a $100,000 grant and they just had a list.

Here are the grants we applied for, here are the grants under consideration that we’re thinking about applying for, and here are all the award amounts. So, the status quo, the way that they dealt with this situation before was that naturally, they wrote the bigger grants first and they prioritized those and if they were gonna hire an outside grant writer, they would tend to hire it obviously for the biggest award amounts and then they would go down the stack. So, that was just the status quo way that they did that. Then, they judged how things went by whether they got the grant or they didn’t, so that sounds like a pretty natural process but this creates this problem of resulting. We’re really just looking at what’s the result of the end point of the grant process.

So, when I went in I said what I’d really like you to do is really think about these grants in terms of expected value, which just means let’s think about our decision-making process in terms of we have the award amount of the grant, say $100,000. Don’t think about that grant as $100,000 grant. Think about that grant as what percentage of the time do you think you’ll get it multiplied by the potential award. If you think you’re gonna get that grant 80% of the time and the potential award is $100,000 then that grant is worth $80,000 because you’re gonna get it 80% of the time that you’re applying over time to $100,000 grant. If you think that that grant you’re gonna get 20% of the time, then that grant is worth $20,000 in the long run.

I said I want you to walk through all of your grants and I want you to take a stab at what the expected value is, what the percentage is. Now, they said to me well how are we supposed to know what the percentage is? I said well, you know better than anybody else does because you’ve been writing grants for a long time and you’ve dealt with grants like this before and so go and try to do your best. Just take a stab at it. Now, what’s important about the willingness to take a stab at it is let’s say that you take a guess at it and you say I’m gonna say 60% because I think we’re gonna get it somewhere between 40 and 80% and the sort of mid-range of that is 60, and what I say to them is well that’s better than defaulting to I don’t know, it’s just worth $100,000 so at least you’re taking a stab at it, and in taking a stab at it, what you’re thinking about is how well does this grant fit? What has my track record been in terms of the grant writing process? What kind of things are out of my control?

What are likely just gonna be due to luck? What is likely gonna be due to the kinds of things that we do? You’re actually going to be thinking about the skill and luck elements of getting that grant, but now what I did was I had them go through and did that. Now they had a percentage of the time that they took a stab at that, a percentage of the time that they thought that they might get the award, and now they multiplied that by the grant awards. Not only does that get you thinking about really what are my chances at the grant? But, notice it changes your work stack now. So, you’ve upended the status quo a little bit because before they would have paid much more attention to a $100,000 grant than a $50,000 award amount just because that’s what they would have done.

They would have said well the $100,000 one is worth more. Look, it’s $100,000 but now what I revealed to them is that sometimes that’s not the case because if you have say a $50,000 grant that you think you’re gonna get 80% of the time, that is worth $40,000, and if you have $100,000 grant that you think you’re gonna get 20% of the time, that is actually worth $20,000 in the long run. So, notice what you figured out is that the $50,000 grant is actually the more valuable grant to you because you’re gonna get it more often so that means you really want to make sure that you’re getting that one done. So, the first thing it does is actually gets you to restack your work.

The second thing it does is it gets you thinking well what can I do in order to improve the probability that I’m gonna get this grant? So, if it’s a lower probability grant you’re actually thinking really clearly about what can I do to push the probability out? It also allows you to see when it’s worth it to hire an outside grant writer because if you say well I think an outside grant writer is gonna increase the probability that we get the grant by X, you can actually look at that, figure out how many hours, so let’s say you think it’s gonna increase it enough such that the grant is worth $5,000 more, what you can do is you can say well is the outside grant writer gonna cost me less than that? And as long as the outside grant writer is gonna cost you less than that, then it would be worth it to hire the outside grant writer so it tells you when you should be hiring independent contractors, for example.

But, then also what happens is that now whether you get the grant or not, you’re not going back and blaming anybody because it’s now in the stack. What you’ve said, if you say I think I’m gonna get the grant 20% of the time, what you’re recognizing in advance is that 80% of the time you will not get the grant, and that has been agreed upon in the room through a process, through a memorialized scenario plan such that nobody’s pointing a finger saying that’s your fault that we didn’t get the grant. Instead, people are saying okay, well, we didn’t get the grant. That was one of the outcomes. We recognized that in advance. What we cared about was the process in the writing of the grant and how we stacked our work, so now let’s just take that to make sure that we call up.

So, what you do now is because you’re so focused on refining the estimates of how often you think you’re gonna get the grant, you call the foundations both where you don’t get the grant, which would be our natural tendency is just to call up and say why didn’t I get it? But you also call up the foundations where you did because you’re gonna be querying on particular things like well how much luck was involved? How much was it because I wrote the grant? Were these things that I could have done better that would have even increased even more the probability that I got the grant? And you’re actually querying the awards when you actually get it versus when you’re denied also.

So, you’re querying on both sides because you’re focused on process now. And, you can see all these really nice signals that it sends to the team there.

That’s a great example. Nice process and then also, as you said, the reinforced learning on what did work and maybe what was missing from a learning perspective.

And notice that it helps you to not, what we don’t want as an organization, not only do we not wanna be defensive in our prospective decision process, which is I’m just gonna stick with the status quo, but we also retrospectively once we have outcomes occur, we don’t wanna be reactive. We really don’t want to overreact to something. So, there are two disasters that can happen. One is that you have a result that’s mainly due to luck and you start mucking around with your decision process because you’re overreacting to the quality of the outcomes, so you have some bad outcome that’s just mainly due to luck and now you start changing your decisions just because the outcome was bad so you feel like you need to do something.

And, then it also stops you from reinforcing decisions that shouldn’t be reinforced, in other words, good outcomes that are mainly due to luck where you start reinforcing those decisions when actually it was due to luck. Sports is such a great place to see this. There’s a really great study that was just done. People can find it if they look on Behavioral Scientist is where it was reported, where somebody looked at NBA teams and what happened when there was a very close win versus a very close loss. So, let’s take the example of an NBA team wins by one point or loses by one point. When there’s a one point win or loss, let’s assume that the quality of the decision making that led up to the one-point win is probably, I think we should just make an assumption no different on average than the quality of the decision making that leads up to a one-point loss.

I mean, that’s clearly within the margin of error. A one point win or one point loss, it’s like did the last team get their shot off? What happened with the shot clock? Who had the last possession? These are gonna be things that are gonna be relatively out of your control, and yet what they found was that lineup changes occurred much more often after a one-point loss than a one-point win. So, in other words, when the team won by one point they tended not to be changing the decision process that led to that one point win. They’re not really changing the lineup at all. But, when the team lost by one point then they were going in and reacting to that and changing the lineup, and that is not what you want in your business.

You do not want to be changing your lineup all of the sudden when you have a one-point loss and not changing it with a one-point win. You want as much as possible to react to those same events identically.

That’s a great point, great example. I agree with the sports analogy, too. I can really see that. How have you applied it to yourself?

So, let me, first of all, say that these biases that we have around results, these biases that we have around outcomes, like any bias, whether it’s confirmation bias or availability bias or hindsight bias, any of the biases that we have are very, very built into the way that our brains work. So, I just want to say I do apply this to my own life but I’m bad at it. And I want to say that really clearly because I think that once we are aware that these biases exist, we can actually be very judgmental of ourselves when we fail to avoid them. So, one of the things that we need to recognize for ourselves in that they’re hard to avoid because I can’t look at the color red and not see red. I can’t just say oh, I know that there are particular wavelengths and my brain happens to be perceived that that is red so now I’m just gonna have that not happen.

The biases are very similar to that as well, that while we might know, for example, that confirmation bias exists or resulting exists, that doesn’t mean that just knowing it is gonna make us stop doing it and as much as we try to put processes in place I think we need to recognize that we will still be falling prey to that on a regular basis. Otherwise, I think that we can get into a stance that is not particularly compassionate to ourselves or the people around us as we hold ourselves to a higher standard than is actually realistic given the way that our brains are built. So, I want to be very clear about that, that I do very poorly at this, but that being said I do better than I would have if I didn’t know that this kind of things existed and if I hadn’t set up guardrails in order to help me to do this less.

The good news is that like compounding interest, small changes make big differences in the long run. If you can cause people to be defensive less, if you can fall prey yourself less to, say, confirmation bias, if you can be more process oriented and less outcome-oriented less, those things will have very big returns over the long run. So, I just want to say that first.

So, in my own life, one of the ways that I’ve really gotten to be able to change these kinds of processes for myself is that I’ve really trained myself and worked with other people to enforce this within myself, so a lot of this, notice that what we were talking about is how the group is behaving. Obviously you’re creating culture as a leader, but you’re reinforcing this as a group that I’ve really set up where the people around me are holding me accountable to these kinds of processes and I am trying to think and express myself in a more probabilistic way, and I think that that’s probably the biggest thing that I’ve done in my own life that has helped me to be a little bit better at this kind of stuff.

So, let me give you an example. If I were to express to you some sort of opinion or prediction that I had and you were talking to me, you would notice this thing that I do which is I try to actually assign a percentage to it. So, when I was working with the nonprofit, After-School All-Stars, and I was getting them to assign a percentage to the chances that they thought that they would get the grant. I actually do that in my personal life. So, here’s an example. If you asked me if I could go to dinner on Friday, I might say to you I’m 80%. So, I’m not saying yes or no. I’m giving you some sense of probabilistically how likely I think that that is to occur. So, here would be another one as far as expressing something in my own life. So, there’s a lot of talk right now about whether say the Democrats are gonna win the house, and what I hear from pundits on TV is they seem to be either in the yes or no camp, but if I were talking to you about it I would say well, at the moment Democrats are pulling way ahead, but November is still a very long way away and a lot of things can intervene between now and November.

I’m not enough of an expert to understand how the districts are drawn to know how I’m supposed to translate necessarily that the Democrats say plus six in the generic vote. So, if I had to say do I think that the Democrats are gonna win the House, I would say 65% of the time. Now, that’s not my actual prediction. I’m just throwing that out right now to make it clear, but I do that about things like that and then I also would do that about belief, for example. There are different ways that I might express uncertainty about a belief. I might give you a percentage, I might give you a range. So, I might say to you Elvis was somewhere between 40 and 47 when he died.

So, notice I’m giving you a range, I’m not giving you an exact number there but that builds my uncertainty into it. It tells you how certain I am of that belief. I might say I think Citizen Kane won best picture, but I’m like 63% on that. I’m always trying to think how sure am I? What’s my level of certainty or uncertainty is a better way to put it around this belief or prediction that I have? The reason why I think that’s just super helpful in terms of overcoming this is that first of all I’m wrapping in on the front end the possibility that the thing that I believe is not true or the thing that I’m predicting will not happen so that when it doesn’t happen, I don’t consider that a bad outcome. I consider that an outcome that I had already expressed so it’s neither bad nor good. It’s just one of the possible things that could happen.

The other thing I think it does for me that’s really, really valuable is that it’s constantly reminding me that my beliefs are under construction, that they’re in progress, that they’re not logged so much as true or false but as probabilistically open and that what my goal is is not so much to confirm the things that I already believe, which would be engaging in confirmation bias and to have my view of the world be correct, but rather to always be trying to refine these percentages. So, in the same way that once you’re putting a number on the chances that you think you’re going to get a grant, what that does is now it causes you to be really, really hungry for information where you’re trying to figure out well, what are the factors that will help me figure out what this percentage is so that I can be better at that?

Once I get the grant or don’t get the grant, I’ve gotta circle back and find out what I did well, what I didn’t do well, what I could do better. How much luck was involved? So that the next time that I try to put a percentage on this that I’m more accurate. So, by expressing myself this way, I keep myself in the zone less of I know this to be true or I know this to be false and more this belief is under construction, it’s in progress, and so, therefore, I wanna be really open-minded to all the information that pertains to this belief that’s out there. I wanna think about why I might be wrong. I wanna find out what other people know that I don’t know, what other people’s opinions are so that I can then wrap that in to become better and refine those percentages more so that I can start approaching a more accurate representation of what the objective truth is as opposed to just reinforcing the things that I already believe.

I think that that way that I talk about things does actually help me. I think it does help me to incorporate other people’s opinions a little bit more. I think it does push me toward open-mindedness and the people around me enforce that. They really help to hold me accountable to that because they know that this is a goal of mine and I’ve expressed that to them. So, sometimes if I do express something with certainty, somebody around me who’s in on it, because not everybody is, but the people who are in on it who are around me will say to me well, how sure are you? How often do you think that’s gonna happen? They’ll actually remind me that I’m supposed to be expressing probability when I say that.

That’s great. Telling people where you’re focusing on and having them give you reinforcement is extremely powerful for learning. So, Annie, this has been wonderful. So, helping us understand resultancy and really thinking about it from a perspective of as leaders in business we want to get results. Of course you want to get results, but when we’re so focused on results we can get people to take that status quo decision, the decision that might seem safer so we’re not being judged, but what we miss out is a great process and a way to really think through the best decision, think about criteria for that, and really get some consensus from a team perspective.

Right, exactly. And actually, I’m really happy that you used that word consensus because I do think that one of the things that we have to be careful of, and I do think this comes through this idea of creating these processes, particularly I think really valuable is that the two breakout groups where one group is told that the outcome is good and one group is told that the outcome was bad and you then see what comes out of that, because what you want to be really careful of is creating false consensus. What is the thing about consultants? They never tell you to eliminate the department that hired them. So, consultants are a way that you can create false consensus.

Let’s say that there isn’t a status quo decision that’s already agreed upon, that there’s no red light, green light situation, and you’re trying to work through a decision. One of the ways to create false status quo is hiring outside consultants are used this way a lot where they’ll come in and just reinforce what the person who hired them wants to do, and so that’s creating false consensus. Well, they agreed with me so therefore we can move through with this or to push the team toward consensus.

So, here’s an example of a way to get false consensus is let’s say that you have four people interview a job candidate. If you have them all discuss it in the room with each other, you will get false consensus around a candidate. So, it’s whoever expresses a strong opinion first, the other people are gonna tend to coalesce around whoever expressed that strong opinion first as opposed to offering up with what their real views are. Now you have some false consensus toward the decision. So, what would be a good decision process there, the same thing is quarantine. So, have each of the four people write a report on how they view that candidate before they get to talk to each other because once they get to talk to each other, you might as well not have had four people interview the candidate.

So, that is one thing that I do wanna stick a pin in is just be careful of false consensus because that’s a different way to get to the status quo.

Yeah, that groupthink.

Yes, exactly.

Well, thank you so much. Is there any other last tip to share or how about this? What is one thing that they can take away and put into action after they listen to this call? One small thing?

Yeah, I think that the main thing that’s really easy to implement is to start trying to express uncertainty in the way that you speak. Now, that doesn’t mean, I don’t think that you’re supposed to go around being like well I don’t know. I don’t know, you choose. Notice if I say something like I think that it’s 65% that the Democrats are gonna take the House in November, I can express that with a tremendous amount of confidence and it’s actually feels very good and believable to the people who are listening to you because by saying that I think that it’s 65% that the Democrats are gonna take the House, I’ve signaled a whole bunch of stuff to you, that I’ve thought about it, that I’m bold enough to actually assign a number to it, that I’m willing to say that I’ve thought about this enough that I will put a percentage on this.

I think that you can express a lot of confidence in expressing uncertainty as long as you express uncertainty in the right way. I think it can make you be a much more believable communicator by saying I’m 65% that this will happen, I actually invite you to share your information with me because you know that when you do that we’re not gonna be in a fight. If all I say in the Democrats are gonna take the House in November, if you disagree with me you may not open your mouth for fear that we will be in a fight because we disagree. But, if I say it’s 65%, now I’ve opened the door for you to do that so you’re now gonna share helpful information with me, which is actually really important to my process.

If I’m a leader, I want people to feel free to express their concerns or their contrary opinions in the room. It’s really important to process, so this is literally one of the simplest things you can do is just start trying to talk this way. Start thinking when you say I think that this is the candidate we should hire, instead of saying that, say and my level of certainty around that opinion is 77%. If that’s all that you do, you will change the culture of your workplace, of your team in a way that’s very positive.

Well, again, thank you so much for coming. I’ve really enjoyed your topic. I’ll be very interested in hearing about your next book and as you start really thinking through more about some of these decisions and how we can become more effective as leaders, leading organizations, leading our teams and helping people have uncertainty with confidence.

Oh, I love that. Can I use that? That’s a great turn of phrase. I would like to be able to use that. I will credit you.

Okay, you can have it!

Thank you.

Thank you.


I hope that you have enjoyed this and can start using some of these great ideas to start making smarter decisions. Make sure to subscribe to be alerted to ongoing podcasts. I work with leaders and their teams to apply these concepts, grow themselves, their teams and their business. Schedule a free 30 minute consultation here to see if I can help you, your team or your organization. You can reach me, Jill Windelspecht, directly by email at [email protected] and visit my website at