Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Deep Learning vs Intuition: AI models and venture capital investing
0 Playsin 22 hours
What if the best investment decision is one where no human is involved?
Brant Meyer, partner at Trac VC joins the show this week to talk about the firm’s approach, where algorithms — not partners in puffer vests — make every single call. Over 115 investments to date with zero human investment decisions. An 8.5% loss ratio, orders of magnitude less than traditional VC, would seem to suggest they’re on to something.
George K. and George A. wanted to know, if machines make the decision, what exactly is Brant’s job? But the more interesting conversation isn't about the wins. It's about what the model forces you to confront. We assume removing the human removes the bias — but Trac's algorithms are trained on data with its own biases.
Then there's the psychological dimension. Brant makes the case that most resistance to algorithmic investing is emotional rather than rational. VCs resist algorithms because the discretionary call is the whole point. The juice, as he puts it, is the feeling of knowing. Strip that away and you're threatening an identity.
Which raises the question George K. and George A. keep circling: how did venture capitalists acquire oracular status in the first place? The hit rate doesn't justify it. The pattern recognition, Brant argues, was never really theirs to claim.
And yet , no founder wants to take money from a robot. The relationship still matters. The question is just whether we've been confusing that relationship with the thing it was never actually doing.
Mentioned:
Recommended
Transcript
Traditional VC vs. AI Approaches
00:00:00
Speaker
Joking aside, I think the part that venture capitalists anchor on and say is their winning edge is pattern recognition.
00:00:12
Speaker
and being able to recognize founders and companies at the earliest stages. And it seems absurd to me that in this day and age with the greatest pattern recognition technology ever created and people are still acting like, nope, I'm gonna do it better.
00:00:28
Speaker
um That's just an absurd premise on its face, but that's kind of been the only option to invest in companies at early stages up until now. And so I think in the future, you're gonna see a lot more of this.
Introduction to Brant Meyer and TrackVC
00:00:50
Speaker
Yo, this is Bare Knuckles and Brass Tax, the tech podcast about humans. I'm George K. And George And today, our guest is Brant Meyer. He is a partner at TrackVC, which stands out as being a venture capital fund that explicitly uses deep learning algorithms to make all investment decisions. As you will hear, Brant has no hand in investment decisions, which is...
00:01:17
Speaker
I don't know how many of our listeners pay attention to venture. Really weird. Really out there. It's a really novel concept. It's very bleeding edge. um I do think, you know, like I hearken back to that office space meme from like the 90s. Like, what do you say you do here anyway? Yes. Yes. But yeah, Brent was cool. And you know what? Really smart guy.
00:01:41
Speaker
Really non-traditional background for fin services. And i think I think this is going to turn some heads if people hear this episode because he's on his own.
Deep Learning in VC at TrackVC
00:01:53
Speaker
Brant Meyer, welcome to the show. Thanks so much for having me, Jordan. Appreciate it Yeah, it has been a time. But you are here because you released this video ah basically trying to argue your job out of existence. So for our listeners, you who work for a VC firm called TrackVC, which specializes in using deep learning algorithms instead of the squishy gray matter between human skulls to make investment decisions.
00:02:22
Speaker
um I love the video mostly because the call to action email is the best, but I will link that in the show notes. um But let's take a pause here. Listeners, check out the video if you haven't already, but we'll listen to it right now.
00:02:38
Speaker
Hi, I'm Brandt from TrackVC. What is TrackVC? Well, we're a venture capital firm that uses AI to make every single investment decision. Yeah, every single decision.
00:02:49
Speaker
So, are the algorithms any good? No, our algorithms are great.
00:02:56
Speaker
We've invested $60 million dollars in over 100 companies with a staggering low loss ratio of only 8.5%.
Role of AI in Investment Decisions
00:03:03
Speaker
Two of our first eight investments became unicorns in less than a year.
00:03:07
Speaker
you know how many of those I've picked? Zero. You know why? Because I'm an idiot. You know who's not an idiot? This guy. At TRAC, our team of data scientists create the models that choose every single investment. Isn't that right, Steve? Move. OK.
00:03:26
Speaker
Do you really believe a guy with a puffer vest and three years at Bain can pick winners better than a deep learning algorithm? AI is replacing computer scientists and engineers. But we can't replace this guy? BC's next.
00:03:37
Speaker
Bro! Stop investing in VC firms that are throwing darts at the wall. And if you're a founder, stop taking money from VCs who can't add value. You wouldn't let me on your cap table just because I know Ashton Kutcher, would you?
00:03:49
Speaker
i mean, I don't, but one day, maybe.
00:03:54
Speaker
So how does TRACK help? With data, research, and analytics. We'll price all your future rounds, benchmark you against your competitors, and rank over 400,000 investors with a predictive score. And some of them are awful.
00:04:06
Speaker
You know who's a horrible investor? Seriously? I can't
00:04:13
Speaker
I get it. Living in the past can be fun, but you can't get exceptional results from VCs that are doing the same old thing. Be part of the AI revolution in venture capital that starts with TrackVC. Was
00:04:29
Speaker
that puffer vest? I can't with this guy. All right. Brant, the video and you are pretty blunt about it. The algorithm picks the winners, ah not the the humans. So for lack of a better place to start, what exactly is your job? i am i'm a professional. i'm I'm basically a salesman for our money.
00:04:55
Speaker
That's it. um On the, you know, joking aside, I think the part that venture capitalists anchor on and say as their winning edge is pattern recognition.
00:05:11
Speaker
and being able to recognize founders and companies at the earliest stages. And it seems absurd to me that in this day and age, with the greatest pattern recognition technology ever created, and people are still acting like, nope, I'm going to do it better.
00:05:27
Speaker
um That's just an absurd premise on its face, but that's kind of been the only option to invest in
Challenges and Biases in VC
00:05:34
Speaker
companies at early stages up until now. And so I think in the future, you're going to see a lot more of this. So then in terms of what my job is, given that, if it's not sourcing the deals, it's not picking the deals, what is my job?
00:05:48
Speaker
um And a lot of that is after the fact and how we service our portfolio companies. And so that's also kind of a new model, which is, you know, if you're a lead investor and you've been on boards of companies, the advice you're giving these companies could be good and it could be applicable.
00:06:05
Speaker
We're not a lead investor, so we're writing smaller checks. We don't sit on boards, um as I'm sure we'll get into you know Before I came into venture capital, I was in the FBI for 10 and a half years. You do not want me telling you how to run your business, how to hire an engineer, or and how to go to market. None of that is going to be helpful. But what I can do and what Trac does really well is we are like your outsourced data research and analytics team.
00:06:30
Speaker
So we will price all your future rounds. We will give you competitive intelligence on your competitors. We can give you a list of, hey, here's the investors that give off the most predictive market signal. But it's basically, we're giving you data so that you as the CEO can make better decisions. And then my job when I'm actually talking to the companies and really what the point of the video was,
00:06:51
Speaker
was just to give you the context of our process, how we work, and how we make decisions. And um the founders really seemed to resonate with that. VC is not so much, but the founders really liked it. Oh, you don't you don't say. um Before I turn over to George, I will say the pipeline of former ah high-placed government folks turning to VC and giving advice...
00:07:22
Speaker
You know, you you said you don't want me. That hasn't stopped anyone in the government from from sitting on boards and trying to give advice anyway. And, you know, we were having a conversation yesterday about a company that didn't pass our algorithms, um but had some pretty interesting external metrics.
00:07:41
Speaker
And we've made 115 investments and no human has ever made an investment decision. And one of the partners was like, who does the fundraising side was like like, why can't we make an exception? And I said to him, i was like, if you take the guardrails off and let me start making decisions, my ego will fill this entire room in two seconds and I will turn into a maniac. So it's a good thing for me. I'm no different than those folks. um It's just that I'm in a context where I'm protected from my kind of worst instincts ah to think that I know more than I do.
00:08:21
Speaker
Well, it's like, it first of all, it's pleasure to meet you. It's fascinating, right? Because... you know George and I have bob a bunch of different things that we do, and more and more of those things are are having AI become part of that process.
00:08:35
Speaker
um But you know human the loop is always still something that's like you know critically important. like i you know i'm I'm in the middle of doing some policy updates at my day job, and when I'm looking at AI governance, it's human in the loop is a critical step.
00:08:49
Speaker
So it's fascinating to me that you... have a value prop where you're taking the human out of that decision process. And then, you know, I look at it like a little bit like Moneyball too, where like, you know, if you remember that movie, um, where, uh, was it the Brad Pitt's character, right? The, the Oakland, uh, Oakland A's manager or yeah, it was a GM when he went to like, instead of going, uh, intuitive scouting, which is a traditional way of doing like baseball or sports scouting, he did everything off stats.
00:09:17
Speaker
Right. So that to me is a model that basically AI would do. um It's just interesting because money. Like you talk to VCs, you talk to angel investors. People are very,
Human Interaction vs. Algorithms in VC
00:09:29
Speaker
very apprehensive about letting a machine dictate where their money's going.
00:09:33
Speaker
So i'm I'm kind of curious from your path. Like you went from analyzing threats at the FBI to analyzing startups at TRAC. Like on the surface, those feel like completely different jobs. But I'm guessing, you know, the underlying skill set is still the same. And I'm kind of wondering, like, what is it? Because you're doing this in a completely revolutionary way. Sure.
00:09:53
Speaker
So for your last question, kind of what's the crossover between being an intelligence analyst and a early stage investor? The crossover is you have incomplete information, a lot of uncertainty, and you have to make probabilistic assessments about potential future outcomes.
00:10:12
Speaker
And you you don't know the future. And so the only thing that you can really control is the rigor of your analytic and decision-making process.
00:10:25
Speaker
And in the world of intelligence analysis, um whether analysts use these techniques or not, but you know you get taught all these basically structured analytic techniques to basically put up guardrails against your biases, come up with alternative hypotheses, um and really look at all possible future outcomes and then make an assessment and say, here's the one that I think is more most likely. And so...
00:10:50
Speaker
in early stage investing, it's very similar. um You don't know what's going to happen in the future. of them, VCs act like they do. um And so you can really only control your decision-making process. And, you know, if you're investing early stage, we invest seed to series B, but probably, you know, 30% of that capital is is at seed. So it's really early.
00:11:10
Speaker
So we're making predictions, but, um, If that company goes on to become a unicorn, most of the things that led to that will happen long after I've made a decision about investing in them.
00:11:24
Speaker
right And so i think I think there's also coming from intelligence analysis, I have a little bit of a different mindset looking at business. I mean, I didn't even study business in college. I was an Islamic studies major. So I am a clean slate when I came into the world of venture capital.
00:11:41
Speaker
But what that means is I think some investors look at companies and they have it in their mind that one of them is a future unicorn. It's predetermined. And I just have to identify which one it is.
00:11:52
Speaker
And that's not the case. There's many future outcomes for each company that are possible. um And I just have to, over time, have a
TrackVC's Competitive Edge
00:12:05
Speaker
little bit of alpha in my decision making such that I'm making a little bit better decisions and I have a little bit more exposure to the ones where the things could go right.
00:12:13
Speaker
But it's not like it's embedded in that. Before I go, for George, I just have two questions for the audience sake. One, how long has Track been doing this, been around? Sure.
00:12:24
Speaker
like how are you accept Yeah. So 2020 was our first fund. So we're coming up on ah May 2020 was our first investment. So five and a half years. I want know to what you can share.
00:12:38
Speaker
How many of these investment decisions have led to return on investment? Because usually the typical game is we throw money at 10 companies and maybe one of them does well. Yeah. So that's, so our, both our decision-making and our portfolio construction is unconventional. So fund one, we had 46 companies. Fund two was in 2022. We had 77 companies and we're investing out of our third fund right now. So over 115-ish companies we've invested in so far.
00:13:08
Speaker
um The first number that pops out is the false positive rate. How many of our investments have lost money? So in fund one, only 11% of the portfolio have lost capital. In a traditional early stage venture fund, it'd be tripled out. and Yeah, know that's bananas.
00:13:28
Speaker
And then fund two, it's 5%. And I mean, that'll that'll grow a little bit. But
Future of AI and Human Intuition in VC
00:13:34
Speaker
so in terms of setting money on fire, we have stopped that on the venture side.
00:13:39
Speaker
um So I think that that I can say, hey, we can plant the flag that we're we're better at picking companies that survive. Now, the next question is, but how many of those that survive are going to be these power law, huge winning companies? Because you still have to have those for the portfolio to make sense.
00:13:54
Speaker
um two of our first eight investments um we didn't invest at a seed stage one was a series b and the other one was like a late seed those became unicorns in about 12 months um and the and we have returns in both funds like actual cash back to investors but we haven't returned both funds yet we're only five and a half and three years in to separate funds. So it's going to take time in terms of like, what's the overall return on the fund, but the markups are great.
00:14:25
Speaker
um and But in terms of DPI dollars paid back to investors, we're top decile funds in both vintages. So, so far, so good. The thing that's also a little bit different about our portfolio construction is to use your money ball example.
00:14:41
Speaker
We still have so many runners on base. um And so it's going to take some time to see kind of where we land with some of those big power law winners. But to belabor the money ball, you have the runners on base. We have the runners on base. Yeah, exactly. And I think, and the other part of it too is, and fund two is significantly outperforming, starting to outperform fund one.
00:15:07
Speaker
So the technology, technology is iterative, but apart from that is, you're an investor in a venture fund, which is a questionable thing to begin with, to invest dollars into which is, that that's a whole other podcast. That's a different podcast. That's a different podcast. So I'm not, I'm not the apologist for investing in venture, but if you are going to invest in venture and you investor, you have to pick the right fund, the right manager within that firm, the right vintage,
00:15:40
Speaker
And then even if he strikes gold and gives you your huge return, he's got to do that again in a different vintage and another fund. And so the repeatability of venture returns is almost impossible.
00:15:54
Speaker
And I think that also lends to, you know, this partner made that investment. He left for the next fund or, hey, this guy was great when the market was good, but he's not when it's bad. the whole The whole value proposition of our approach too is the systemization of it.
00:16:08
Speaker
and the repeatability of it and trying to professionalize the asset class.
00:16:15
Speaker
Over to you, George. That's awesome. Yeah, so let's talk a little bit more about the process. um And ill let me set it up for a little bit. I want to understand, like, what is the application process to the decision process? Because I think anyone who has any inkling of venture, right, it's usually...
00:16:37
Speaker
Companies, founders pounding the pavement or getting connected to their network, doing pitches, pitch after pitch. And then you're just, it's it's very people oriented, right? So this is a ah new mind model. But the reason I ask, I think is because...
00:16:52
Speaker
There are some nuances, just to make the listeners aware, you've intimated, right? Like, track is not a lead investor. There are firms that firmly are. That's their thesis. Yep. they're only early stage. And then there are some that, like, I don't touch anything before Series B. And some are late stage. There's all sorts of things.
00:17:11
Speaker
Yeah. But a lot of those are also predicated on being able to establish two intangible things that i I feel like could test the data hypothesis. The first is if you're using pattern recognition, are you missing like the weird outlier disruptive technologies, right?
00:17:37
Speaker
Maybe. The second intangible is there are for better, for worse, there's a reason there are serial founders, right? They exit and then they come back around to the same group of people and they're starting a new company.
00:17:54
Speaker
And VCs tend to trust those teams over the idea because they're like, i have just seen what these people can do. Sure. They're operators. They can weather the storm. I saw them pivot like three times and they still came out on top, right?
00:18:09
Speaker
So those two things feel like they are resistant to the data hypothesis. So if you could take it in two parts, just like quickly, let's describe TRAC's application process. And then how do you account for those intangibles?
00:18:21
Speaker
Yeah. So our application process is if if you're a Series A or B company, um I typically will already have some data on you. And I will have a good sense of whether you pass my algorithms. So for for all of those companies, it's really just outbound on my end. So it's like, i already know whether your company we could likely invest in. And so by the time I've contacted you, like you already score really well.
00:18:43
Speaker
um And then with those companies, I'll have a 30 minute call with them, which is really just to explain our process and give them context for us to see if they want to go through a process. And then if they do, I just shoot them a form. It's like 10 questions. It takes them five minutes to fill out. And it's nothing, it's typical investor stuff. Yeah.
00:19:01
Speaker
And then that data is just the ground truth that I compare against the data I have from all these different sources I've bought data from. And as long as that matches up and they still pass the algorithm, then we invest. And sometimes it doesn't.
00:19:13
Speaker
And then I'm like, hey, you know, you don't pass and and we can't invest. Seed stage is a little bit more of a black box in the sense that I probably have to do more. Not enough data. Right.
00:19:25
Speaker
In terms of like knowing who to talk to and who to like have come through our process. so And we're trying to... to create other algorithms so that we can go earlier with other data so that it's not like, we're so it's not me deciding who to talk to at seed stage. um And so we've like kind of worked around that to your question about pattern recognition, which is one we've gotten before, which is, you know, something along the lines of, well, if you're building an algorithm based off of historical data, how are you going to identify some new technology or trend that doesn't exist in the data?
00:20:00
Speaker
And I usually turn the question back on the other person and I say, that's a great point. And how do you expect Joe Puffervest to identify that? And so if it's identifiable, so there's one question is, is it actually identifiable at all?
00:20:17
Speaker
Or is it just luck? Right. If it's just luck, then, you know, hope you're lucky. um If it's not luck and it can be discerned in some way and you're an investor in a venture fund, which fund would you rather invest in?
00:20:31
Speaker
Joe Puffer Vest or the greatest pattern recognition technology ever built? So that's a personal question for you. I think, but, you know, that's a... But I think what we're overlooking there is they've worked with Joe Puffer Vest before. Joe Puffer Vest has been successful. And so that's a fair that's a fair ah way to deploy your capital.
00:20:55
Speaker
But in the abstract sense, I think the argument falls apart a little bit just in terms of, well, someone's got to recognize that pattern. um And, you know, is it is it possible that our technology will score a company company negatively because they're doing something that's never been done before? Yeah, and I'm sure it's possible and I'm i'm sure we've done it. so um But I'm not so much as worried about that as being able to build a model that's able to consistently identify and deploy capital in the venture space in a way that's going to give our investors a consistent return.
00:21:32
Speaker
So that's kind of what I'm more worried about. And then the second part of your question, what was the second part of your question Like the human element? Yeah, the team, you know, like the know-how and the personalities. Yeah. So like I was saying, as we're trying to go earlier on seed, we have a ah sixth algorithm that we're beta testing that is trained just on seed and we're making smaller investments out of that.
00:21:54
Speaker
And it measures 61 dimensions and about 70% of those are founder-related metrics. And so we're basically trying to see like, We've seen some combination of predictive factors of kind of founder data that are somewhat predictive, but we just don't know yet how predictive is that. And is it too, are we overfitting the model to previous outcomes that is not really indicative of future outcomes? So there is some part of that.
00:22:25
Speaker
um and one of those very One of those variables should be, did they appear on the Forbes 30 under 30? 11% chance they're going to prison. Yeah, exactly. That's the fraud predictive ah technology, you know?
00:22:37
Speaker
um So there is that. And I think what ends up happening is, even when we talk to LPs, kind of along the lines of the questions that you asked is, keep kind of the the what's behind your two questions a little bit is like,
00:22:56
Speaker
can AI really do everything that's necessary as part of the picking job? And I think on the sourcing and the picking, I think it absolutely can.
00:23:07
Speaker
And I think it's going to do it in ah so in a superior way. Much in the same way, like um if you've ever been in a Waymo taxi, the self-driving cars, no like if you look at the data and you just think about it intellectually,
00:23:21
Speaker
Those cars should be way safer than a human being driving a car. Yet, the visceral emotional experience of being in the car when the wheel's turning or it goes down the street you don't think it should go down is very uncomfortable.
00:23:36
Speaker
yeah and it And it is for that reason that a lot of VCs will never kind of go down this path. It's psychologically, it's uncomfortable. um Now, the part that's I don't think ever going to be replaced on the VC side is, even though we invest capital this way and source companies this way, I still have relationships with these founders.
00:23:58
Speaker
And no founder wants to take money, i don't think, from a robot. I mean, maybe they would, but but likely not. So I think where the future venture is gonna land is somewhere between where it is now and what it looks like when you apply for a mortgage loan, right? Like when, back in the day, you wanted to go get a business loan or a mortgage loan, you gotta put on a tie, go sit in front of the banker and give them the whole spiel and a song and dance. Yeah, just all subjective prejudice and impressions. And now it's like, what's your FICO score? How much money do you make? Here's a check.
00:24:37
Speaker
And so tracks more on that end of the thing, which is like, we don't need to do this dog and Tony show. Like here's some capital. We're going to make a lot of bets. um if If a company passes our algorithms, we typically say they have like a one in six chance of becoming a future unicorn.
00:24:55
Speaker
But I don't pretend to know which in my portfolio are going to be those ones. Right. And when even LPs ask, like, who do you think the big winners are going to be? I'll tell them who's doing the best, but I don't even engage in trying to fool myself that I know what's going to happen.
00:25:12
Speaker
Okay. i I really appreciate that. I also, you know, my largest hope would be the algorithm helps break a lot of the prejudice that already exists, you know, a very low percentage of yeah founders of color, women founders yeah getting funded. And you yeah know that there's just so much bias built into the system.
00:25:32
Speaker
And I think, you know, honestly, I think that's a question we talk about and struggle with because we're also training on data. with that bias embedded in it. And so we have to kind of constantly interrogate the data and the context with which um it existed in, for sure.
00:25:52
Speaker
So like I do find this fascinating, right? Because, you know, you bring up the Waymo case. And I remember when I wrote my first Waymo at RSA last year, and I was like, this is insane. um They did recently come out with ah some pretty negative headlines about their support teams over in Asia actually operating those vehicles. So that's not a good headline for you when you're trying to run self-augmenting vehicle company.
00:26:21
Speaker
After the break, we get into what value VCs actually bring to founders. What is the psychological safety element that's at play here in investment decisions and even pitching decisions?
00:26:34
Speaker
And then also we just ask the obvious question. If TRAC is building something automated at scale, does it just kind of eat the VC market? I don't know. gets really meta, but stay tuned for the second half of the show.
00:26:54
Speaker
I think what's interesting is I appreciate the decisiveness of your process as you explained it. Because in my experience in dealing with VCs, they will waste all your time in the world. Yes. And not anything. thoseate but They'll take free options as long as you're giving them, you know? They'll they'll take all your free ideas until they they'll you know they're blue in the face and just do nothing for you and you're just left waiting there.
00:27:20
Speaker
yeah So I do appreciate that you at least can provide like an decisive end to that conversation. yeah And I have to say though, like an 8.5% loss ratio is like genuinely mind blowing, but that just impressive as hell.
00:27:36
Speaker
But here's the skeptical question, right? Sure. Is that the algorithm working or is that a function of, ah you know, the stage that you're investing at? Which sectors and the market conditions in the last few years?
00:27:47
Speaker
How do you separate those things? Yeah. So I think, so for both funds, by by capital deployed, They're all about a third seed, series A, B. um By number of companies, um they're mostly seed just because they're smaller checks.
00:28:08
Speaker
um So they're like 50% or 60% seed companies. And so um i think I think two things are going on there. I think one, we deploy bigger checks at later stages.
00:28:23
Speaker
And we also have more data. at later stages. And so the kind of efficiency of the model is probably best at like series A. you know At series B, a lot of the data that we have, probably a lot of other investors or are better at picking off of that data as deals get later.
00:28:48
Speaker
um But the problem is at Series A is also those are the hardest deals to get into. and so it's easier to get into deals at seed and then follow on um at Series A. So to answer your question, the loss ratio, I think like, yeah, we haven't lost a lot of our capital because um those big checks that we write, we typically like, are the they just have lower attrition rates and and we know more about them.
00:29:11
Speaker
So that's part of it. As far as sector though, like I think, i mean, i don't know the number offhand. I mean, we've got, i mean we've got to have damn near 60 70 seed stage companies. So our loss ratios should be way higher um than that.
00:29:29
Speaker
um And sector wise, I mean, it's pretty broad. We have space companies, rocket companies, media companies, electric motor companies, consumer.
00:29:41
Speaker
it's It's pretty well across the board. You know, certainly... Most is software companies, um which is not a typical of any other venture firms. But yeah, so I think there's... So in terms of how to explain it, I think, hey, there's the alpha of our algorithms, but you know there's also the alpha of...
00:30:04
Speaker
ah It's not my bias. It's not the other partner's bias. It's not because I went to school with them. It's not because, oh, this guy also went into the FBI and I you know have an affinity for him.
00:30:15
Speaker
There's a consistency that you just can't match with a human decision-making process. Now, are there biases in the model? um Absolutely. um But they're pretty consistent. So I don't know if I answered your question, but my sense is just the way that we're picking and the...
00:30:35
Speaker
discipline to never veer and take things of our own hands. Like there are there are founders that I fall in love with, companies I love, and they don't pass and we don't invest.
00:30:47
Speaker
Also, there's plenty of companies where I'm like, I do not get it at all. That founder was not honestly impressive and we give them money because the algorithm said to invest. And that has two also side effects, which is,
00:31:03
Speaker
One, it keeps my ego in check, right? Like if if this fund crushes it, you can't put me on the Midas list. I didn't pick them, bro. like i'm not i don't that doesn't It doesn't imbue me with anything, you know?
00:31:20
Speaker
At the same time, if the company completely traps the bed, it's not like a personal dagger. to my heart, you know, because I didn't like pound the table that like we've got to do this deal. And so it does help us to be a little bit more objective and detached when we're making these decisions.
00:31:39
Speaker
I really appreciate that last point because two things are true in Silicon Valley and elsewhere. When you have gotten lucky enough or you have enough high-profile unicorns under your belt, yeah these VCs sort of take on this like oracular status. Philosopher king. Yeah. And it's like...
00:32:02
Speaker
Nah, there's a rich guy, had a hunch, got lucky, right? Because you're like, oh, they invested in Facebook early. Yes. But what are all the, where are all the dogs in the graveyard that didn't make it? Right. No one talks about that. So there's definitely like a confirmation bias. Yes. um But I also think to your point, when things go south, people who have that reputation,
00:32:26
Speaker
Do weird things to make it seem like it wasn't their quote unquote fault either. the i don't know. There's like lots of machinations to make up for it because you you have status to lose. Whereas you're saying like, I don't know, machine did it. Yeah. Right. i Well, I think a lot of venture, and this is, this is, I'm not venture batching. There's plenty of people of venture batching. And when you're not in venture and then you get into venture, you start realizing all the dumb things, you know, that you think people do from the outside of why they do them. Like there's incentives and things in case, but because the timelines and the feedback loops are so long and so few venture investors have any way to know if they're any good of it.
00:33:03
Speaker
What, what venture ends up becoming is competence theater. Yes. I love that term. Yes. So it's like, I'm not going to be able to give you results for a long time.
00:33:15
Speaker
And the only way you're going to even know the results is if one, you've invested in my fund or two, I personally tell you. And so that's going to drive the behavior you're talking about, which is I'm going to shout it as loud as possible from the rooftops yeah when i score and I'm gonna spike the football on everybody.
00:33:36
Speaker
um Or I'm just going to keep quiet when things go badly or like kind of point the fingers. And it's not because I'm like, necessarily a horrible, narcissistic, grandiose person. Though we've got plenty of those yeah as as any other, but it's just, it's this weird psychological effect that being in this business has on you.
00:33:57
Speaker
One example is like, um and I think this has helped me in my venture career, but when you, you know, you're working on these things, you're trying to be good at them. You're trying to educate yourself.
00:34:10
Speaker
You know, you finally make a deal and, you know, you've kind of got a lot riding on it to pay off because you don't get a lot of shots on goal. And so you're kind of amped up for this thing to work. Well, the analogy I draw is when I was in the FBI, i was working a drug case and there was this subject and he would rent a car every three days on the dot for a year.
00:34:32
Speaker
He spent $35,000 a year in rental cars. And it was like clockwork. And drug dealers are not known to be very like... historically like On schedule. On schedule, logistically, whatever. So I'm the intelligence analyst. I've got my FBI goggles on.
00:34:49
Speaker
And I'm thinking, right, let's intelligence analyze the hell out of this thing. So I'm doing big, you know, big-ass charts and heat maps. And I've got things going on. And I came up with this long list of like...
00:35:02
Speaker
possible explanations. Like one was, well, you know, he doesn't want to get robbed driving through the neighborhood. So he changes cars every three days. Or, hey, if he gets pulled over with drugs in the car, it's a rental car and there's couple of deniability. And like I had all these things and I thought I was like the smartest guy that ever lived.
00:35:20
Speaker
We finally arrest this guy. We proffer him. So he he volunteers to talk for a reduced sentence. And at the end, I'm like, hey, man, I just got to know what's with the rental cars. he was like oh, you know what?
00:35:31
Speaker
My boy Swizzy used to rent cars and it looked cool. And that was it. And I just deflated completely.
00:35:45
Speaker
There was no reason. There was, yeah, there was no reason. I was just like, when you put FBI goggles on, those long you're going to FBI the heck out of things. And so the crossover with Venture is...
00:35:58
Speaker
Because you the these things take so long, you become so invested and it warps your vision of things to the point that when you see the behavior of them blaming whatever, I think a lot of them truly believe it.
00:36:12
Speaker
It looks absurd to the outside, but it just sucks you in so much that you're like, the only kind of way to preserve your ego is to be like, yeah, well, that's... I knew it you know way back when. But if you go look at the deal, no it's like, dude, you love this company.
00:36:26
Speaker
So I think it's just a very tough psychological space to live in. this kind of you You never have resolution and you have a lot of uncertainty. I mean, the psychology is very real. Founders are very...
00:36:38
Speaker
invested emotionally, right? There's a lot in there. So um you guys say in the video, you know, stop taking money from VCs who can't add value. I think investors also have a sense of psychological safety. Like if they can get A16Z, they feel like they're on top of the world or whatever like that, right? So all very human stuff.
00:37:00
Speaker
How do you make the case when you're doing that outreach to founders? Yeah. If they have a choice between track and I don't know, somebody at Sequoia. Yeah. You should definitely take Sequoia.
00:37:10
Speaker
If it's between the two of us, just to be clear, you should definitely take Sequoia. I would take Sequoia if it was mine, but here's why we're not a lead investor.
00:37:21
Speaker
I don't think, and mean, maybe a firm in the future, this will work. I don't actually think our model works as a lead investor. Interesting. Um, in the sense that Sequoia and a 16 Z, if they're on your cap table, like you can recruit better.
00:37:38
Speaker
You can like it, they, they do add value. Now, is it because they're great at picking companies or are they more king makers? Like once they pick you, it then endows you with something. Yes. There's, there's, that's very subjective for sure. Who knows, but it's a, the effect is real.
00:37:53
Speaker
Correct. it You know, so I'm not like poo-pooing that. Um, so so So the first point, I'm not competing to be the lead check in a deal.
00:38:04
Speaker
So I'm playing a different game. So I don't have to compete with Sequoia or A16Z or whatever. I'm not taking a huge chunk of the route. So for every every, whether it's seed A or B, There's typically one seat for a lead investor, but there may be four to 10 seats for follow-on investors.
00:38:24
Speaker
So really, I'm only competing with those folks for allocation kind of after lead. um And also, I think, you know, the video, I was very strident um on purpose to be provocative and and funny, but there are many models in venture that make money or there are a few models in venture that make money. Most models don't, but every firm's not going to look like us. So anyway, in terms of how we win the deals.
00:38:53
Speaker
So I'm, I'm competing for a follow on check, a smaller check to fill out their round. And when they hear our story, We close 95% of the deals where the founder takes my call, which I think is about as high as anyone I've ever heard. And it's not because it's me. If my other partner, just 70 years old, when he does it, it's the same rate.
00:39:15
Speaker
So it's not because I'm like some great salesman or like have a weird background. It's because... our story and kind of what the video touches on speaks so clearly to founders who were like, kind of like like what what George was saying, like, these guys are giving me calls, wasting my time. They don't know what they're talking about.
00:39:33
Speaker
Whereas we're like, it's algorithmic, fill out the form, I'll give you yes, no answer tomorrow. And then we'll give you a suite of research data and analytics to help you throughout that. And no other, the other part is no other venture firm really is giving them that pitch.
00:39:49
Speaker
And so, you know, they're having meeting, and meeting, meeting, and then there's pattern interruption, which is us. And it kind of like makes them take notice. um yeah And so the other thing, the the last thing I think, though I have, i have, have zero data to support this. So this is, I'm tagging, this is hot take. um So pick it for what it's worth. But because the founder does not have to sell me, it's the algorithm that's going to decide.
00:40:18
Speaker
we're able to actually really have a much more human conversation and build rapport because it's like, hey man, and I'll tell them. Like my opening pitch to founders is, Shrek, we're a completely algorithmic-based investor and all that means is I don't make any investment decisions. I'm just the face of the machine.
00:40:36
Speaker
So let's just have a conversation. Right. And that kind of like gets their attention. Um, so I think then when they hear kind of what we bring to the table and we're just kind of like weird and different, and most founders think that they're weird and different.
00:40:49
Speaker
I think there's a lot of resonance there. Nice. That's, uh, this is, I gotta be a real. I didn't expect this conversation be so cool. Um,
00:41:01
Speaker
We had really low expectations, Bram. Thanks. Bro, this is my life. Low expectations and you just step in over that low bar, you know? Like, I just got to ask this, right? So, track is building something that, like, if it works at scale.
00:41:17
Speaker
Essentially, you know, it argues that most of the people in your industry are overpaid and unnecessary. And, yeah ah you know, there's a lot of founders that feel the same way about that, even without knowing you exist.
00:41:30
Speaker
Do you ever feel like you're building the thing that eats your own profession? Well, I don't, I mean, that's the funny thing is i don't really think of it as my profession. I'm just a money salesman.
00:41:42
Speaker
You know, and I'm ah the evangelist for like, hey, there's got to be a better way to do venture. So I've never actually done venture in the conventional way. So i don't I don't really worry about it. But I did think it was hilarious. I don't know if you saw this like six months ago. Mark Andreessen was on a podcast and he goes, he's basically saying everything's being replaced by AI and then says with a straight face.
00:42:05
Speaker
But I think venture might be one of the last things to be replaced by AI. Because it's hard. And I'm just like... Come on, bro. I'm like, come on, bro. Like, get out here. Yeah. And to be clear, if I was Marc Andreessen and had made the money he made, I would be saying wild stuff all the was just for me I would be a maniac. So I'm not like throwing chill at him, but I'm just like, come on, man. That's not like, like let's get serious. But but i think I think on the like sourcing side and on the picking side,
00:42:38
Speaker
I think absolutely, it's going to be hard to argue that. um But ironically, I think on the human relationship side and being able to get into the deals, um there's you still have to have a human that the other human wants to have a relationship with. Otherwise, because these hot companies, they they've got money thrown at them left and right. And so I don't think that's going to go away. And I think...
00:43:07
Speaker
I have embraced the human side of my job, which is like building relationships with these founders, getting to know them. And just basically like, I'm just here to like, make sure you guys know we exist and know what our process is so that when I, hey, have my list of 150 companies this year I'm trying to get a hold of, you at least have some context for who I am before I reach out. And if you don't want to go through a process, that's fine, you know?
00:43:30
Speaker
um But I think the folks that think that they are smarter than everyone else or like... I mean, it's the classic, you know, stock picking versus exactly invest in the index fund, know. A hundred percent, you know? And the problem is...
00:43:47
Speaker
from an emotional and psychological perspective, indexing doesn't give you the juice. yeah And people want the juice. yes And I don't blame them for that, you know? yeah So they're going to fight against it because they want the juice.
00:44:00
Speaker
um Yeah. Wow. I really appreciate it. Man, we could go for hours. i have a lot to say about VC. I've seen amazing VCs that really go to bat for their companies. I genuinely want you to succeed. And I've seen VCs that just look like loan sharks. Like, I gave you the money.
00:44:15
Speaker
Give me my returns. Like, like 3 million for 30%. Yeah. That's a real thing. Oh yeah. So, um but I really appreciate the time, Brant. Thank you. ah Really also appreciate the class exhibited in the video.
00:44:30
Speaker
ah And um yeah, this is, this is really interesting. This is ah like a perfect VC conversation for this podcast. So thank you. Awesome. Thanks for having me guys. i really appreciate it. All right. We'll talk to you soon. Okay, man.
00:44:45
Speaker
right, let's leave you with some questions to take forward. Touched on, i think, pretty much like psychological safety felt like the big theme of that in this weird investment startup landscape. And so I guess the question I would pose to our listeners is Getting at the heart of algorithmic decision making, where are places where you would want human interaction and where are places you want machine interaction? Like Brandt brought up, mortgages used to be super subjective. And if they just didn't like the way you walked into the bank, they could deny you the ability to get a home. And now we don't have that. Thank God.
00:45:21
Speaker
ah What does that mean for technology investment? Is there something similar there? I think that's interesting. Yeah, I look at it like, you know, what role does human intuition still have in the world of business?
00:45:34
Speaker
Because at the end of the day, you know, like a big part of why we'll say VCs are one thing, but high net worth individuals and angel investors are also a huge, huge resource, especially for a lot of startups that aren't in the circles they need to be socially to get VC funding.
00:45:51
Speaker
And a lot of those investment picks are made off of the intuition of do they like the founders, right? Because those higher net worth people, they're not technologists. They usually they make their money other ways and they're they're picking based on how they feel.
00:46:04
Speaker
Well, is there still going to be a place for that kind of investment are all these folks and investors just going to buy some track type software at scale to just do the job for them? Yeah.
00:46:16
Speaker
Things to consider. We'll leave that with you and we will see you next week.
00:46:23
Speaker
If you like this conversation, share it with friends and subscribe wherever you get your podcasts for a weekly ballistic payload of snark, insights, and laughs. New episodes of Bare Knuckles and Brass Tacks drop every Monday.
00:46:36
Speaker
If you're already subscribed, thank you for your support and your swagger. Please consider leaving a rating or a review. It helps others find the show. We'll catch you next week, but until then, stay real.