Poker, Probability, and Financial Modeling with Dan Mayoh

I know you have a horror story over your modeling career. Tell us that.

Every now and then as a model auditor, you just see something that's truly shocking and is almost— can't really be audited well because there's just too much wrong with it and it just needs to die, be reborn in a new form.

We've all opened a spreadsheet before and we're like, what were they thinking?

Meet Dan Mayo, founder and CEO of Fintega Proprietary Limited, an early CFM holder who's turned an actuarial mind toward everything from infrastructure bids to beating Wordle.

I think everybody's a little interested. I know you've done some mathematical modeling of games, but talk a little bit about that, what you're doing.

I remember when Wordle came out, all the craze for a while, myself and Diarmuid Early going back and forth with ideas and building some spreadsheets, building an Excel tool to help you solve Wordle. Just using Excel for analysis of games where there is a strategically optimal way of getting there. Could be other things like trying to write an Excel program to solve Sudoku.

You've done obviously a lot of math modeling, Game modeling, any lessons you would give our audience of how it's helped with kind of decision-making and uncertainty?

The way I like to answer this is—

Financial Modelers Corner is the world's premier modeling podcast. It is brought to you by Financial Modeling Institute, the world's leading financial modeling accreditation organization. Welcome to Financial Modelers Corner. I'm your host, Paul Barnhurst, the FP&A Guy, and in this podcast we talk all about the art and science of financial modeling with distinguished guests from around the globe. The Financial Modelers Corner podcast is brought to you by the Financial Modeling Institute. FMI offers the most respected accreditations in financial modeling, and that's why I completed the Advanced Financial Modeler And encourage you to do the same if you haven't. This week, I'm thrilled to welcome on the show Dan Mayo. Dan, welcome to the show.

Hi, Paul. Thanks for having me. It's good to know you let non-distinguished guests sneak in from time to time.

We get all types, and I would say distinguished. I mean, you came in, was it top 3 in ModelOff, right?

Top 2, actually, in the first year I competed.

I was gonna say, I thought it was second, but I wasn't, couldn't remember for sure.

Yeah, and I was one of the earliest CFM holders in Financial Modeling Institute as well. So this is—

A Chartered Financial Modeler. And, you know, many of our guests are interested to know you do a fair amount of poker playing. You've done some professional poker playing, right?

Yes, I do. I've made some okay money playing poker. It would be nice to do it, I think, full-time professionally, but hard to do that at the moment with the travel it would require and so on. But yeah.

Yeah. As to say, you're in Australia, in case our guests are wondering, and I'm going to guess most of the tournaments aren't in Australia that would allow you to make a full-time living?

There are, particularly if you want to play no-limit hold'em. I think there's enough in Australia that you could certainly play full-time, whether you could earn enough to make a living full-time is another question. But I tend to prefer the mixed games, things that aren't no-limit hold'em, and they're not as popular. And so yeah, for the really big events there, you do have to travel to the US from time to time.

And that makes sense. And I do like how you said you could definitely find enough games to play full-time. That doesn't mean you'll make a full-time living.

Yeah, especially tournament poker. It's very high variance. People can do everything right and still have an edge theoretically, or, you know, have a long-term expected value and still lose $100,000 in a year as well for doing nothing wrong.

There's still a chance component to all this?

Of course. If it wasn't, it wouldn't work as a gambling game. It would be like chess. Ahead of time, who's likely to win the game? Just the better player. And those better players are therefore find it harder to get action. So we'd say getting action would be like finding someone to play against you for money. A good game for that needs the right balance of luck can prevail in the short run, but skill will prevail in the long run. And poker and backgammon are good examples of that.

Yeah, it makes a lot of sense. You want, you need enough of a skill component that people feel confident that over the long run they can come out ahead. There's a chance element that kind of also can be fun too.

Yeah. Otherwise the weaker skilled players won't ever put up their money against the more skilled players if they're going to lose 99 times out of 100 or 100 times out of 100. They need to win, you know, 20 or 30 times out of 100 to make it worth their while.

Let's jump into the episode. We'll talk a little bit more about gaming. I know we have that on the schedule, so hopefully people enjoy that. But before we jump into a couple career questions I want to ask you, we have a career corner section. Why don't you just take a minute and tell our audience a little bit about your background? Just kind of give them the rundown of Dan.

Yeah, sure. I'm a mathematician at heart. And when I was at university, that came through in, uh, I did a degree in actuarial studies. So I got an actuarial background. And then it was around the time of finishing university and I did a 12-month placement overseas in Colombia, working with PricewaterhouseCoopers in South America. And a lot of the work there, I sort of got exposed to financial modeling. And I think I'd made the decision then I wanted to make my career out of financial modeling. And so when I returned to Australia, I sought out a job in that field. That was 20, 21 years ago now. So it feels like a while. And my first job was, it was a model auditor. And I was working in a team where probably half the people in that team, 10 out of 20, had actuarial backgrounds. I guess there was just something about the skills and the way how we approach logical problems and details orientated and whatnot that made us good model auditors. And yeah, that's where it began. And then do you want the rest of the quick career highlights from there as well? Or should we—

Sure. If you want, go ahead and give us another minute or so of a couple of the highlights and jump into the question.

5 years in that model auditing team, which I was living in Sydney at the time. We did jobs Australia and all around the world as well as probably half Australian projects and half international projects. And then I left that, took an in-house job at what has become Australia's sort of green bank, the Clean Energy Finance Corporation. I got in there at a very ground level. There was a little pilot company called Low Carbon Australia, started a few years before that and got in there at the ground level as their financial modeling guy. Did that for a few years and then branched out into my own consulting business. And I've been doing that consulting business for 12 or 13 years now, essentially building spreadsheet models for corporate clients, um, with a little bit of model auditing and training on the side.

Love the background. Appreciate that. The first one I want to ask is, I'm curious, When did you realize kind of financial modeling could be a career path, right? It's not, it's like it's never been a degree in school. Historically, it hasn't been a career. It's just part of a job you do. But today, really financial modeling, some people will call themselves a financial modeler. So how do you think about that?

Yeah, I'll go. I think it was probably maybe during that year when I was in Colombia with PwC that I mentioned when I did Realized it could be a career. I remember even learning about, you know, actuarial studies as a thing and being an actuary. I was only sort of in year 11 or 12, like immediately before university when I first heard of that and learned about it. And I think it was just playing around with building, you know, spreadsheet models and building You know, either something for a client or something to use in-house, like a, you know, a calculator tool that did something specific, which I would now think about as essentially, you know, you've been tasked with writing a piece of software and you're just choosing Excel as the programming language to do it. Yeah, it was probably at university and that first year after university where I kind of realized Um, okay, you can actually make a career out of this. You can find jobs that specialize in this and you can become really good at it. Um, and you can learn best practices similar to how, you know, software programming would have best practices as well. Um, so then I went looking for jobs in that area. Um, and they were coming back to Australia. They were all in Sydney or Melbourne. And I grew up in Canberra. I was living in Canberra at the time. Um, so I knew I'd be moving to one of those places and yeah, I found a job in, in Sydney. It was actually, um, no, I'm not going to tell you that story. Doesn't matter. Um, so yeah, that was it. And now I look at it now and I, you can absolutely make a career out of it. Um, I think there's a lot of industries like in investment banking, where the modeling is given to people at the very earliest stages of their career when they go into investment banking and they're maybe not taught properly or taught to the same rigor how to do it well, or they're just expected to learn it themselves. And then they'll do that for a few years and then they leave it behind as they move on to other tasks that investment bankers do. And so people that have gone through that kind of track, you know, whilst they certainly know their way around a model, I think aren't as skilled, or they're certainly, it's noticeable to a professional whose whole career is financial modeling. You know, there's noticeable gaps in their skills, whereas someone who has, you know, chosen to specialize in Has advantages. And yeah, most of those people end up in consultancies, I guess, and working for specialized financial modeling consultancies, either running their own businesses or as part of larger businesses, like your Operuses of the world and so on. And then, yeah, I don't know what it's like. Now in the US, I remember, you know, 10 years ago certainly being of the view that, you know, Australia and the UK had quite mature markets for financial modeling professionals. And even with model auditing as well, those markets used formal model audits a lot more as part of their transactions, whereas in the US it wasn't It wasn't as common a thing. Um, and so that's definitely been eye-opening for me, I think, over since ModelOff started really, which was 2012. Um, and then, you know, every year through that I was branching out internationally and I was, you know, meeting people and meeting other top quality modelers from around the world, including from the US, and learning You know, the different ways, different approaches to financial modeling in different parts of the world. And even, you know, FMI, which sponsors this podcast, I've had quite a few debates with Ian in the early days of FMI about, you know, what is best practice and, you know, when I'll Um, you know, working with him on developing the AFM exam and creating cases for the CFM exam, you know, um, and, you know, the, the norm way of doing things in Canada and the US is, is maybe different to what I feel is the normal way of doing things and the best way of doing things. And so it's just interesting to be aware of that.

Yeah, you bring up a great point. It definitely, you know, It's, and it's different in profession, you know, how project finance goes about it is different than maybe what best practice is for M&A or best practice if the model's going to be used again and again versus a one-time transaction. And so there are, I mean, there's, there are certain things that should apply to every model no matter what, but then you get into the, okay, well, here I have to do it. Other places I probably shouldn't do it. Like there are some places where you might have to have some circular references. There are others where It's like, you know, the general is not at all, or just different things. So I totally agree with you. It's interesting to see how different it can be depending on how somebody was brought up and what area they focus on.

Yeah, but ultimately, if you're writing the model for someone else, um, yeah, you got to bend a little bit to what they want and what they expect and how they're going to use it, you know.

Correct, to your audience and their skill level. And there's so much that goes into that. So I have one more, one more question I kind of want to ask in Career Corner, and then we'll, we'll move kind of the main part of the interview. So What's one skill over your career that you've developed as a modeler that you think most people would be surprised by, that they don't typically think of as a skill modelers have to develop?

Yeah, I actually, I don't know. I read this question, you know, earlier today, knowing you would ask it and struggled to really think of an answer. Maybe the thing that comes to mind right now is just The similarities between modeling and software development, I think, and just thinking about, you know, a model being like a, you know, a computer program that you're developing. I mean, Excel is just the language that you're writing it in. And, you know, some people have never thought of Excel that way, and that it is really just the most accessible programming language in the world. It's a great open sandbox for anyone who wants to play around with it. But then you can also write professional models in it that underpin transactions in the tens of billions of dollars as well. And it is still the tool that the industry uses at the big end of town for those things, you know, and it's— I could go on and on about, you know, Excel here and that it just, you know, people, a lot of businesses might start using Excel for one kind of thing and then they grow to a certain size and they outgrow using Excel for that thing and they need a better, you know, solution and database type work, CRM type work is an example here. And then they, you know, make the mistake of assuming that because Excel is not appropriate for that task when you get too big, that means Excel's not appropriate for any task when you get too big and important. And that's absolutely not the case because when it comes to financial modeling, it is, yeah, it is still the best software that is You know, still used on the biggest transactions.

So there's a reason a billion people use it. So I like that computer programming, you know, that you learn programming skills is not something people often think about, but you're right. I mean, Excel is the most accessible used programming language in the world if you want to think about it that way. Yeah.

And I think designing a financial model on paper before you start building it and all the components it'll have and how they'll link to each other. you know, can be quite similar to, you know, when you want to write, you know, a piece of software that does something and designing the flowchart of the logic of the software on paper as much as you can as well before you go and start building it as well. There's similarities there. So that's probably the best answer I can give you to that specific question.

That works. All right. I know you have a horror story over your modeling career. Tell us that.

Yeah, that was the other question that stumped me. Tell me about your worst financial model ever. And if I split this into models that I've kind of built from scratch versus just models that I've seen or been tasked with as an auditor or been tasked with to rewrite sections of as a consultant, I think they're quite different answers. Models that I've you know, build from scratch. Yeah, not many horror stories there.

That's good.

Yeah, there's a lot where the goalposts move, you know, in the middle of the project or near the end of the project to adapt. Um, and there's some models that I would have written, you know, 15 years ago that if I would open up now and have a look at it, I would, I'd just be shocked at, uh, that I was capable of producing something of, you know, such low quality as I would do now. But I guess that's good as well because it shows up being, you know, learning and advancing my skills. But nothing truly horrible there. But from an auditor point of view or from a, you know, a consultant where, you know, a client might ask you to You know, take their model and, you know, maybe take their template and use that as the basis for writing something for a certain transaction, or, you know, take their model and just rewrite one part of it, but, you know, not touch the rest of it. And that sometimes doesn't really work. There's just been some bad ones that with more experience now, I would just learn to say no. You know, to the, to the, um, to the brief much earlier in the process. Um, and then from a model audit, the worst ones would just be every now and then as a model auditor, you just see something that's truly shocking and is almost, can't really be audited well because there's just too much wrong with it, you know? Um, and it just needs to die and Be reborn in a new form. And yeah, but from the auditing point of view, I can't, I mean, I know things like that have existed, but I can't remember specifically what they are. From the, here, take my model and, you know, use that as a starting point for the work we want you to do. Yeah, a couple come to mind, but I don't want to name names of certain clients that have done that and so forth. I'll just say that with more experience now, I know not to get myself into those corners to begin with.

Good way to put it. There are times when it's like, I've learned, I'm steering clear of that one. Not surprised to hear you say when you get some models that you're auditing, it was like, this thing just needs to be, needs to die and reborn. We've all opened a spreadsheet before and we're like, What were they thinking?

It's a challenge as a consultant, particularly working for myself, where you might be at a, yeah, just a point where, you know, you're desperate for the work and desperate for revenue. And so you'll say yes to something that, you know, you should have said no to, or, you know, you'll agree to do it the client's way when you should have pushed back harder, you know, to do it. different way or something. Um, and then, yeah, you just get yourself into a real hole. Um, and yeah, you either painfully get through it or you, you know, admit defeat far later than you should have and, uh, pull the plug on the thing and just wasted everyone's time and money for a little bit. So it's, uh, it, you know, my learning from that is to Um, yeah, listen to the little voice in your head that tells you you should say no to something, even if you're really desperate for the work.

You know, that's one of the most important lessons you have to learn as a business owner is when to walk away, when to say no. You're like, but the revenue is good. Yeah, but it's just not worth it in the long run sometimes. I'm not surprised to hear you say that because that's definitely one I've had to learn. All right. So you started your career in auditing. I'd love to dig into that a little bit. What did you learn from auditing models that you've taken with you today that's shaped the way you build models?

Yeah, just, you know, being exposed to lots of different styles of models, you learn pretty quickly what works well and what doesn't in terms of design principles for the most part. And, you know, the exact syntax of a formula or exact choice of a function used Is, I think, far, a far lesser part of the learning than just, you know, how to lay out the model, how to structure the navigation around the model, how to structure your error checks, how to, you know, nicer ways to, you know, build a table of contents that can, you know, dynamically update easily and Um, you know, nicer ways to just lay out each sheet to, to be able to find your way around the model and find the part you're looking for. And, um, you know, even in terms of, you know, what you do with the first, you know, 12 columns of each sheet for, you know, different labeling and helper cells and things like that, you know, before you get into your timeline and, and just what works well.

That makes sense. There's just, there's a lot of value in being able to see a lot of different ways of doing things.

Yeah.

Yeah.

But then, you know, an auditor is a bit unique in that they have to, you know, pretend to be a user of the model as well as have to, you know, really investigate in every cell of the model. So they really need to learn what the model's doing and drill into that. And so you pick up on every little thing that makes that easier, you know, to do. And And then you can put your developer hat on and, you know, when you need to make adjustments to the model or you need to update one section of the model or some goalposts move and you've got to amend one thing or you've got to add a new feature to the model, there's definitely design principles you can follow that make those development tasks easier versus harder as well, you know, like sort of And I could get into that, but it's, um, you know, it's beyond the scope of what we're talking about today.

That makes, makes a lot of sense. Appreciate that. So I want to ask one question. You competed in ModelOff, right? You mentioned that. I think you came in 2nd in the world. You mentioned that really helped give you confidence to start your consulting business. So maybe you can share a little bit about that.

Yeah, so this was in 2012, which was the very first year ModelOff was run. And I learned about it. I think I was a member of an online sort of learning platform called Excel Hero Academy, which was a paid course run by a guy named Daniel Ferrer. And then, you know, one of the sort of the bulletin boards on that website, you know, I learned about ModelOff, the competition. Signed up out of curiosity to what a financial modeling competition would be like. He ended up qualifying for the finals in New York. They flew me to New York and I was one of 16 from around the world. Came second in that. So that was, I guess, the first eye-opener that, you know, previously I'd only ever known I was like, the best Excel guy in my office, you know, and now I had a benchmark against, you know, a much wider group from around the world on that. So that, yeah, I guess gave me a— made me aware maybe of the level of my skills, but more so it was actually the organizers of ModelOff that I met And, you know, 3 names, John Persico and Johan Odu, who, you know, were behind the Model Competition, and Joseph McDade, who wrote the questions for the Model Competition for the first year. I'm sure a lot of your listeners might be aware of Joe McDade. He's since gone on to work for Microsoft in Redmond as one of the Excel program managers and Brought us dynamic arrays and all those things. Um, but Joe at the time, you know, was just working for his own little Excel consultancy out of Melbourne with a couple of business partners.

Appreciate you sharing that and kind of, you know, it's amazing how often the connections help us. Not so much, I mean, I'm sure you learned a lot from ModelOff itself, great experience to compete, but it's the people you met that really kind of helped you seeing, oh, I can do this. There's a lot of, there's others doing it. Kind of helping you get it set up because I think— Yeah, absolutely.

It shows you how to do it, shows you that it's possible, it's not that hard. Um, you know, and whilst I, I guess I had, you know, the, the technical skills, um, you know, to, to sell them out as a consultant, um, you know, yeah, the result with Model F made me aware, you know, that You know, I was, yeah, maybe more elite than I was aware of on that technical side. But yeah, it was really just the people that showed me how to follow that path. Yeah, I'm still friends with them all to this day. I remember, you know, conversations with Joe, you know, over Teams prior to dynamic arrays coming out. You know, I'd signed an NDA with Microsoft so he could talk to me about these things and chatting with him about certain features of dynamic arrays and how they work and what syntax we'd use. So things like the@ symbol and the hash symbol and what they do with DAs and all these conversations where I'm bouncing ideas around with Joe. It's good fun.

It's funny when you mentioned dynamic arrays, I was teaching that yesterday and it never ceases to amaze me how many people still have no clue that Excel can spill.

Yeah, I can.

Yeah, I mean, you and I know it, but just, it's, it's pretty amazing how many people are still unaware that it's dynamic by nature now. Still operate always in single cell, and there's times you need to. They did a great job of allowing it to do both.

Yeah, they did. And, but you're right, even just that little shift of, um, you know, not needing the Ctrl+Shift+Enter anymore because every familiar with, by default behaving as an array. And, you know, the day that happened, you know, which has led to also the spilling, it's— yeah, I get that people who aren't power users, you know, and don't keep up on, you know, reading about Excel developments might never be aware of that. And, you know, I remember even when Excel 2007 came out, which was, you know, the start of the larger grid, you know, we could go beyond column IV, you know, and beyond row 16, 65,000, you know, and at the time I was working in this model auditing team. And so, you know, there's 20 of us and we were all Excel professionals, but it was still quite useful for, you know, one or two of us to deep dive into it about everything new that Excel 2007 could do. And then we ran a little— you know, training session on a certain Monday for the rest of the team. So they would be aware, you know, and I guess if even power users still needed that, you know, I get that how, yeah, a lot of casual users might be oblivious to the developments in the last 10 years of Excel.

It's easy to do when you don't use it as often as power users do. Great Great point there. So I want to talk a little bit, I think everybody's a little interested. I know you've done some mathematical modeling of games. So talk a little bit about that, kind of what you're doing.

Yeah, look, I guess that just comes from having this consulting business and having an interest in analyzing games and the actuarial skillset. And Excel being my programming language of choice, you know, I've done quite a lot of, yeah, building models of games that might have a random element or have a find the optimal strategic way of playing this game element to it sort of thing where you're kind of modeling out the game tree. And I think some of the first examples I did of this was when I was still working as a model auditor. Um, and there was a time in 2008, I spent about 3 months living in, in Colombia again. Um, I went back for 3 months and I was just doing my Sydney model auditing job remotely from Colombia for a few months. Um, but I, Befriended a, the owner of a small poker room slash casino in Bogotá. I was playing a lot of poker there. This is an American man who owned this little casino. And he was, you know, offering some sort of variations on blackjack rules or some side bets on his blackjack games. And, you know, kind of wanted to know a bit more about the mathematics of what he was offering and if they were exploitable to card counting and things like this. And Um, he had already banned me from playing blackjack. He'd eased that wish away at this point. Um, but I was friends with him, so I was happy to help him out. Um, and yeah, so build up some models to run some numbers, you know, on these questions he was asking and, um, gave him his answers. Um, and that's kind of where it started. And I guess I just realized I had a bit of a knack for it and quite enjoyed it. Um, And yeah, I've done a lot of it since then, you know, whether it be actual work for a client or something I'm just doing for myself because I'm interested in it. And it could vary from, yeah, casino-based examples and wanting to, you know, know what's the best way to play, you know, a certain casino game where you have decision points during the game and what the edge is if you do that. you know, to things like hypothetical questions like, you know, if you had this extra little bit of information, which, you know, you're not meant to have, but like if you could see that, you know, this dealer's card here was a, you know, if you knew it was a high card or a low card and, you know, if you had that information, how does that change the edge of the game? Just for fun, things like that. Or it could just be, you know, then other You know, non-gambling related games. I remember when Wordle came out and, you know, it was all the craze for a while. Myself and Diarmuid Early going back and forth with ideas and building some spreadsheets on kind of use, you know, building an Excel tool to help you solve Wordle, for lack of a better word, you know.

Why am I not surprised you work with Dim on that? That does not surprise me.

I mean, Dim has a similar mind to myself. Um, I met Dim through ModelOff, um, when he competed in ModelOff, uh, you know, before he became the, uh, superstar he is now. Um, and, you know, Dim is a strong mathematician as well. And so we, yeah, we had a lot of shared interests.

Yeah, I can definitely see where there's translation from the math and the actuarial to the game side. I mean, right, there's a lot of, you're trying to narrow options down, a lot around uncertainty, reducing that uncertainty. So when you're playing the game, there's a lot of what you're modeling. You got to—

Yeah, it can be that and it can just also be building simulation models, you know, and running Monte Carlo simulations, or, you know, depending on the problem at hand, building a deterministic solution that rather than just does simulations actually looks at every possible way the game can unfold with precise probabilities attached to each, you know, to give you exact answers on certain things. And those sorts of skills have been useful in You know, paid consulting work, which I think was in one of your questions as well around, you know, abatement modeling for, you know, say if I'm working on, you know, I've done a lot of work on infrastructure transactions and people bidding for things like, you know, the concession to run a public transport franchise, you know, to run a train network in a city or something like that. And, you know, as part of that bid, you know, which is really this consortium I'm working for, submitting a price that they want in order to run the network, you know, there'll be elements in the contract that say, well, if you don't meet certain performance metrics, you know, you have too many trains canceled or late arrivals or things like that, you know, then you get these financial penalties. And at the bid stage, the consortium needs to be able to budget an allowance for that or to quantify how much do we actually expect to incur in these penalties or earn in bonuses or whatever it might be that are performance related. And they would have some data from their experience or any other networks, for example, around the frequency of events like, you know, a light rail vehicle hitting a car or, you know, a train line being shut down due to a power failure or whatever it might be. You know, they would have, you know, some probability around the frequency of these events and, you know, then I would turn that into— You know, dollars at the end of the day, you know, to budget an allowance along with not just an expected value, but a probability distribution to sort of show at different confidence intervals, you know, how it might look. And, you know, the way a lot of those abatement contracts work is you might budget, you know, over a 20-year contract. Yeah, let's say for argument's sake, a $10,000 a month loss, but it turns out that, you know, 95% of the time you actually lose zero and the 5% of the time you lose massive amounts, you know.

Yeah, you lose a million dollars. Yeah, yeah.

You know, and, and, but, but you didn't, you know, they want to understand that and, you know, and it's the same sort of skills, you know, that I talked about with the mathematical modeling that can You know, that I can use to, to do that sort of modeling for them as well.

So I want to ask one more question here, then we're going to kind of move on to rapid fire, which I'll lay out how that works. So you've done obviously a lot of math modeling, game modeling. Any lessons you would give our audience of how it's helped with kind of decision-making and uncertainty? Because, right, there's a lot of uncertainty in any of that you're doing.

Yeah, there is. Um, and yeah, I saw this in your list of questions as well, and I think it was related to, you know, the poker background as well and how that helps with making decisions in uncertainty. And yeah, I think actually, like, the way I'd like to answer this is the experience playing poker has probably helped even more than just the pure modeling of it, although one reinforces the other. Which is to say that, um, yeah, I've learned, you know, what uncertainty is and that you can't be outcome orientated. You know, all you can do is be confident you made the best decision with the information you had available at the time. Um, and then And when there's an uncertain element to it, well, you can't really judge yourself based on the outcome that happened then. Um, you can only judge yourself based on, you know, did you make the right decision with the, the information that you had? And sometimes you win, sometimes you lose, um, even when you make the right decision. And that's fine, you know, and I think you With enough time playing poker, you learn to accept that quite easily and move on from that quite easily.

I think a lot of times people think, you know, I worked in FP&A and we always had to submit a budget with a number. And, you know, a lot of people would just do the number and it's like, help people understand the risk around that because you're never going to hit that exact number. You know, how confident are you in it? And so I think what you shared there is a lot of what I had to learn in You know, in a different way, in my career is trying to help people understand, okay, well, this, this, and this have to be true to hit that number. How likely are these to be true? We've never done one, we've never done two, and we've never done three. So let's reassess that number, you know, those type of things.

Yeah, I think that's a different, it's a slightly different thing, like covering up your downside and looking at, you know, risks. And, you know, this comes up in the infrastructure bidding for contracts as well. Like you identify all these risks and you might quantify them, you know, what's the chance of this risk happening? And then what's the cost if it does happen? And, you know, you add all that up over 100 risks, you get a, yeah, a risk register and a risk allowance that you put in, as well as wanting to know, you know, what are the really bad ones that could actually lead you to default on the contract? But I think that's a skill where modeling helps, but it's a slightly different thing to just making decisions with imperfect information and being comfortable to do that. Like, you know, it can be as simple as like, you know, is this a go or no-go on some capital expenditure decision, you know, and or on an M&A acquisition or something. And just, yeah, I guess being comfortable with uncertainty and probability and, you know, and experience in things like playing poker. Yeah, you become much more comfortable in just understanding that, well, you'll never know everything. You'll never have That's a good example.

And I think we could, we could spend a whole episode on this, this whole thing you're talking about here. But, you know, I'll say 2 things, then we'll move to rapid fire. I, I loved how you mentioned, you know, it helps you just be able to make decisions with imperfect information, which is something everybody needs to be able to do. It's a core skill. And then, you know, the whole idea of the randomness impacts the penalty. And there's a lot of situations, drunk driving is a great example Because there are some people that it's a— it feels like a slap on the wrist, and there's others that it's much more serious because of the outcome, even though they both, they both made the same stupid decision at the end of the day. But because of a random chance, one ended up in a lot worse of a situation. And so yeah, there's a lot of examples like that, and it's an imperfect world. So you bring up some, some good points.

Yeah, I mean, that can happen on the flip side with like Executives being rewarded for outperformance, and sometimes it's just random chance and not nothing they did as well where they get it. And as a poker player and as someone who tries to get better as a poker player and look back at the decisions you made and to learn from them, we try to very much take an opposite view where we're not— if you made a poor decision, But then got lucky and won, you know, well, you still made a poor decision and you should still learn from that, you know, even though you might've won the tournament. And vice versa, you might've made the right decision and got unlucky, you know, and lost. But when you look back and self-critiquing, you know, how you did, you say, well, I still did everything right and I made the right decision here, you know? And so.

All right. So rapid fire on this one. Here's how it works. You have to pick a side. No, it depends. No elaboration, just kind of quick answers. And then at the end, you can elaborate on the one that you feel like there's the most nuance because we know there's exceptions to everything. It's more just kind of fun. If you had to pick a side the majority of the time, how would, what would you recommend?

Sure.

So first one, we'll start here. Circular references in models. Yes or no? No. No.

VBA? Yes.

Lambdas in financial models?

Yes, they can be used.

Okay. External workbook links?

No, if possible, and under very specific, you know, ways of using them if required. Yeah.

Merged cells, are they ever acceptable?

See circular references.

Do you think financial modelers should learn Python in Excel?

I haven't needed to yet. Okay.

I'll take that as a no. What about Power BI?

Again, I haven't needed to yet. It's not to say they shouldn't, but you can make a career in financial modeling without either of those 2 things.

What about Power Query?

Same. You don't need it. Okay. It's useful to have a big toolbox, but no, it's not needed.

I'm curious what you say on this one. Should every financial modeler be able to build a fully integrated 3-statement model?

Yes.

Okay. Will Excel ever die?

The short answer, no. The literal answer, well, yes. You know, we're not going to be using it in the same way.

Everything dies in the long run. Yeah. What financial statement do you think is most important for modelers? Income statement, balance sheet, or cash flow?

Income statement.

Okay. What's your favorite LLM? Do you like Claude, Copilot, ChatGPT, or something else?

I'll defer to my wife. She likes Claude. I haven't used them enough to kind of know, but my wife uses them a lot. She says Claude. Love it.

If you could pick only one for all your models, you could only do one, would it be sensitivity or scenario analysis?

Can you elaborate?

So sensitivity in the sense of, you know, you have this output, if you change it, here's what happens. So you're running sensitivity tables. Scenarios is you're creating different scenarios. I got my best case, my worst case, or Off the model.

I would say it's the same thing. The scenario is just letting you change 10 sensitivities at once.

I guess you could argue that. I get what you're saying.

That is how my mind thinks about it.

You're the first one that's asked me that way, so I like it. All right. So we'll give you that. So you can do your scenario and make your sensitivities from there. All right. Do you believe financial models are the number one corporate decision-making tool?

No, but they probably should be.

That's not an uncommon answer. What's your lookup function of choice? Which one do you like?

Index.

All right, you're an index guy. I like it. And then last question, favorite Excel shortcut?

I love putting things in the Quick Access Toolbar. And so I can just, you know, a shortcut for me will be like Alt+6, but that's going to be different for someone else.

Correct. Your favorite are the ones you create yourself.

Yeah, but to give you a better answer, the one I use really quite a lot is just inserting rows, which Alt+I+R works. Yeah, I was just going to say, old legacy shortcut.

Alt+I+C, Alt+I+R.

Yeah, yeah, yeah. But it is also just Alt+4 for me on my quick access toolbar. Inserting rows, I use a lot.

Okay. All right. Thank you for joining me. I've enjoyed the conversation. If you want to share anything about your business or any last thoughts before we wrap up, anything you want to share before we go?

Nothing comes to mind. You're gonna have to prompt me a little bit with a question or something.

Well, I'll give you a final question. All right. So people want to learn more about you, Dan, about your business. What's the best way for them to do that?

Okay. So my business is fintega.com. The website needs an overhaul. It kind of got corrupted at one point. And so it still exists, but it's a lot of like lorem ipsum type text on there. By the time this airs, perhaps I would have fixed it. But that's where you can reach me if you want any consulting work done. Otherwise, learn more about me. I've taken a backseat in the last few years of being an active member of the global modeling community compared to sort of what I was 5 years ago. And so have less time these days with my family.

There's a lot of truth to that one.

Yeah. Or I've just decided I'm too old for, to be a regular participant in industry debates and things like that. So I don't know. I guess I've chosen to be on this podcast. I'm on the, you know, the FMI Global Leaders Council. So there's certainly still a part of me wanting to remain an active member of the global community in financial modeling. That's it. Otherwise, just, you can send me an email, dan@fintega.com, or find me on LinkedIn and reach out if anyone wants to follow up with anything. Why don't I respond?

Thank you for joining me, Dan. I appreciate it. Great chatting with you. I'll go so you can enjoy the rest of your evening, but thanks, Dan.

Thank you, Paul. Enjoy your day.

Financial Modelers Corner was brought to you by the Financial Modeling Institute. This year I completed the Advanced Financial Modeler certification and it made me a better financial modeler. What are you waiting for? Visit FMI at www.fminstitute.com/podcast and use code podcast to save 15% when you enroll in one of the accreditations today.