Hey miners, welcome or welcome back to the channel, MCN Money, the official home of power mining analysis. In today's episode, Anthony Power and I are joined once again by Jason Les. He's the CEO of Riot Platforms. They just came out with their Q4 and full year earnings, and that's what we're going to be talking about in today's presentation. Before we get into it, take a second, smash a like button, guys. Big help to myself on the channel. If you're not already subscribed to MCN Money, feel free to join and let us know in the comment section below if you're currently holding shares of it, what you thought about their Q4 results, and your outlook for the remainder of 2025. Now, with that being said, let's get into today's interview.
Okay guys, so that's right. Today's video, highly anticipated and very timely given the market conditions today. Q4 earnings interview with Jason Les, the CEO of Riot Platforms. Jason, Anthony and I are in New York this week. We just finished listening up to the earnings and wanted to get into some of the details here under the hood about Riot's strategy, obviously the HPC, and really the plans for 2025 looking forward.
Awesome, well, thanks for having me on. Let's get into it.
Sounds good. So first things first, can you just give us a quick high-level summary of the Q4 numbers, the full year numbers, and maybe some of the highlights in your opinion?
So for Riot, I think the two key highlights, or a few key highlights for us in Q4 for the full year, we're hitting record levels of revenue, record net income, record EBITDA, adjusted EBITDA, and a lot goes into this. Of course, yes, we had the rising Bitcoin price that certainly helped us on the market for our Bitcoin balance, but what you also saw from Riot is our operations ramping up significantly over the year. Q4 was the first year that we saw a lot of Corsicana online, not complete, and then a lot of our Rockdale growth online, also not complete. But we really saw the benefit from that in mining over 1,500 Bitcoin in the quarter, of course post-halving. So we're really proud of the results that we were able to show here as a result of these efforts. Revenue for the quarter went from 67 million, 67 and a half million to 126 million, so really big jump quarter over quarter there. Our hash rate obviously increased a lot quarter over quarter as well. Net income was a 154 million loss for Q3 and then a 136 million gain for Q4, of course a lot of that driven by our Bitcoin treasury strategy increasing in value as well, but clearly succeeding in benefiting from the upturn in the Bitcoin market alongside our operational execution, which has continued to be very strong. So we're really happy about that. And then I think you saw a major theme of our earnings call was talking about what we're doing with respect to our AI HPC strategy as well, aggressively going after how we can maximize the value of our power. We have a huge portfolio of power, both used and unused, unused being a pretty significant amount here, and we think with how the market is rewarding growth in the AI HPC space, with our assets, we're very well positioned to execute there as well. So you saw a mix of a bunch of exciting topics, I should say, maybe a handful of exciting topics here: executing on the Bitcoin mining operational growth, Bitcoin yield, increasing value of our Bitcoin going up, and now us diving headfirst aggressively into the AI HPC opportunity.
Yeah, I mean, it's a good set of earnings there, and hopefully this blip that we're seeing the last, probably week of Bitcoin starts to recover to avoid having any sort of issue with the next earnings during the end of March, because obviously as you highlighted there, a big part of your earnings was in the revaluation of the Bitcoin value on your balance sheet there, and that was, like many of the other miners who've got HODLs, it's a real benefit this period. And seeing Cipher be the first to release their results had a similar position with theirs where they had a strong gain on that.
Anthony, just what's, I think, strong about our strategy is it's in addition to the value of our Bitcoin going up, it's increasing the amount of Bitcoin, increasing Bitcoin per share. And I think we are particularly proud that we were able to get a Bitcoin yield for 2025 of 39%. A lot of that came from the timely convertible debt offering we did at the end of the year, but a good portion of that also came organically from our Bitcoin mining operations adding Bitcoin to our balance sheet for monthly production and not selling it, drove that as well. And we had a good slide in our deck that highlighted the incremental value we've been able to appreciate from doing that strategy. So yeah, Bitcoin yield, key part of the Riot strategy.
Yeah, no, absolutely. I mean, operationally wise, we've been covering you for a number of years now. We've been highlighting this last, probably ever since we arrived at Corsicana last year, that there's been that improvement month by month. I mean, we were highlighting a hole before the Corsicana visit, and you kindly updated the retail investment community as to why that was occurring. But we've seen sort of month by month over the last sort of five, six months some strong gains in your operational hash rate, which have obviously then helped towards these results coming out. And looking at your power cost in the account, I mean, it's probably one of the lowest in the mining community, similar to what Iris have produced, because you're both operating in Texas, you're both benefiting from the market rates, certainly at Corsicana. Can you tell us a bit more about how that power differential between Corsicana and Rockdale came about for the year?
Yeah, so all of these facilities that you're mentioning, they're all in ERCOT. They are in slightly different load zones that have their own pricing. Corsicana is in load zone north, and then miners such as Iris, they're in load zone west. So the prices are often correlated, but not always. You will see differences driving those different power prices. In Riot's case, at Rockdale, we have a hedge, a number of hedges totaling 345 megawatts that go out to a number of years. We procured them at opportunistic times, so the prices in those hedges are low. Now what we're able to do is effectively trade that power, execute on our power strategy in that market. When demand for power is high, we are curtailing our operations and we're able to earn power credits, offsetting our energy bills by not using power, and we also participate in ancillary services and that sort of thing. Going into Corsicana, we did not procure the long-term hedge, and that was a question we got a lot a year ago: hey, are you going to hedge your price, are you going to fix your power price for Corsicana? And our message to the market was, we don't think that's a good idea right now. You know, the hedge prices, they reflect market sentiment, emotions in the market. We thought that the forward curve was elevated relative to what we could achieve procuring power in the market at Corsicana, so we opted not to do that. That's really a great advantage of Bitcoin mining and getting low power prices. If you have the flexibility to shut off during these high price hours, you're trading a small percentage points of operating uptime in order to get a significantly lower power cost. So we've been able to achieve a competitive power cost at Corsicana regardless. We are starting to do very small strategic hedges with Corsicana, not 24/7 for the next five years. We're picking different volumes of power at different periods of time, small relative to Rockdale, but because of that, we are still generally, the power price that we get at Corsicana is generally driven by the market price. That went up a little bit in Q4 versus Q3, but you can see all-in, we're driving a very competitive cost of power. For 2024, our all-in cost of power, net of all fees, taxes, everything, 3.4 cents per kilowatt hour across our whole portfolio of sites.
Jason, one topic we talked about in a prior podcast, maybe two, that I haven't heard a lot of other people talk about is the transmission benefits that you're going to receive at Corsicana this year. So you just mentioned the power price Q3 versus Q4 had gone up slightly. When could we expect to see some of those transmission benefits come in?
Yeah, I'm glad you brought that up, Bryce. So in ERCOT, you can basically earn a reduction on your transmission cost for the load that participates in what's called the 4CP program over the summer months in Texas, June, July, August, September. So we participated in that, our early months there at Corsicana as that facility ramped up, but you don't get the benefit of that program until the next year. So when we're participating in 4CP in 2024, we're earning the reduction in transmission costs for that portion of load for 2025. So as we get into 2025 here now, we will have reduced transmission cost for that load that participated. Now, we weren't all the way ramped up in Corsicana over the summer. We're using significantly more power there now than we were in the summer months, so we're not reducing transmission on all of that load, but we're going to start appreciating from participating in that program for a good amount over this year.
Are those transmission credits always a sort of like 12-month rolling, or is it the fact you have to be there a year before you start to receive them?
It's based on your participation in the previous year's four coincident peak months. The level you participate with load then saves you in the next year. So if you participate in 2024, looking back, it's not like you appreciate those reductions in transmission costs that year at all. It's all about the next year.
So the transmission benefits for '25 will appear next year then, will they, at Corsicana?
Right. So if we participate in 4CP in Corsicana this summer, you know, with now the incremental load that has come online since the last summer, then that would be earning additional reduction in transmission cost for 2026.
That's correct. Yeah, and I'm assuming, I mean, you know, you can't book any value of that into the accounts or earnings because...
And we don't report it in our monthly updates as if we have earned it until we are actually realizing it.
In terms of, I mean, your accounts are quite well set up, so you do split your accounts out into the various business units. You've got your engineering services and you've got these other services, which I think still got some of the hosting elements. Can you give us an update on sort of like the hosting position? And I believe you mentioned on the call about the litigation costs and all that. I'm assuming maybe some association with that, but how far down the line are we now getting this rectified?
So you're correct on our financials. The other bucket is primarily all, I believe 100%, it could be off by a small amount related to our legacy hosting business that exists at Rockdale with one customer. We are in litigation with that customer. The timeline of how that concludes is unpredictable. But as I noted on the earnings call, unfortunately we have elevated cash SG&A expense associated with legal fees with that right now. So unfortunately, can't give any more.
No, no, no, no, no, no, no, I don't. But in terms of those costs themselves, are you not, I take it you're not treating them as one-off costs? Because if they're one-off costs, would you not be applying them to just the adjusted EBITDA calculation?
You're referring to the one-off cost within SG&A?
If you have litigation costs there, which you could assume say these are one-off costs, would they fall into the adjusted EBITDA where you can take your one-off expenses off your EBITDA?
The way we report it, we do not adjust that out in adjusted EBITDA. I think a lot of that has to do with it's not its own line item, it's talking about a bucket of expenses within SG&A. So investors have to just kind of rely on our commentary of what is included in that bucket of cash SG&A expenses.
Yeah, I mean, the challenge we had this morning, because last night we had a meeting to go to, but this morning it's going through the... it's your calendar year earnings, it's your full earnings. For other miners, it's like Q1 or Q2, and I'm sure there'll be a Q3 coming out as well. Everyone has different calendar years. You know, you're reporting on the whole year and showing your margins and everything like that. It's sort of a challenge then when you sort of say, well, these are Riot results, how do we compare them to the other mining results there? So I was trying to strip out this morning the effect of Q4 by taking your previous Q3 numbers and sort of saying, well, the difference between that and this is the Q4. And it's not an exact science because Q3 is not audited either, so there could be some sort of like minor changes between what you released in Q3 and what you've put in your final Q4 there.
So we did break it down on our earnings deck. We broke out Q4.
You did break out the margin. I saw that for the margin. Yeah, and you show the SAR thing that I've been producing in my sort of comparison table is the breakdown of the Bitcoin as well. So that was all useful. In terms of obviously Q4, when you look at that, because when you're looking at the whole year, you had a pre-halving part of that year where you're obviously benefiting from a higher hash price versus the fact that you've grown your hash rate significantly through the year and you've lowered your efficiency and you've been achieving really good energy costs. So Q4, in terms of how do you see Q4 as a standalone quarter compared to the other quarters?
Well, I would look at Q4 compared to Q1 in light of the halving and everything that took place there. And I think what's important to note for Riot is we ended the year mining just about the same Bitcoin per day as we did at the start of the year, despite the halving and despite network difficulty going up, I think 65, 70%, maybe more than that over the course of the year. I don't have the number off the top of my head right now. So the increase in network difficulty, the halving, those are two major headwinds that all miners were facing. And because of the growth that we achieved, we're able to mine a consistent amount of Bitcoin at the year-end position compared to the beginning. Also because we're good on an energy cost, because we've been scaling up and starting to achieve economies of scale, and because the new miners that we're deploying are latest-generation equipment in terms of power efficiency, our margin for Q4, gross margin for the Bitcoin mining segment at least, was not that much less than Q1. So when hash price was higher in Q1 and obviously pre-halving, we had a 56% gross margin, in Q4, 50%. This is the result of us scaling up with good equipment and focusing on operational efficiencies.
With the Corsicana site being like fully immersion cooling and you have your 200 megawatts at the Rockdale site, so you're the largest miner with any immersion cooling within their business model. Are you seeing the real benefits during the winter period? Are you having less effect than maybe some of the miners who haven't got that type of technology?
Well, I would say the winter periods are not really when the benefits of immersion specifically apply. When we talk about immersion more generally, we think about the lifespan of the equipment. If we can keep these machines operating at more stable, cooler temperatures for years, what does that do to them? And from what we've seen, it helps reduce the amount of maintenance, repairs required, so we can get as long of an economic life out of them as possible, even past our depreciation. That's one of the main benefits that we see with immersion. And then there's the opportunity to overclock as well. You know, we're not doing too much of that at scale yet, but we're thinking about that. Another real benefit of immersion is the density that you're able to achieve there. We're able to put so much, and you guys saw the facility, we're able to put so much more power in one location by using immersion cooling and not having the same type of challenges you would have putting that much air cooling in a single location. But in general, Corsicana is running very well, definitely benefiting from the improved immersion design. You know, there's different types of maintenance that go on with doing immersion as compared to air cooling, but you're seeing in the monthly results we're getting pretty solid average operating hash rates as a result of doing that. And obviously the summer months is when the real challenge comes in for all miners, and we'll see how everything hangs up there. But we're very optimistic about this new design.
Just final bit on where you mentioned repairs and the benefits of the immersion cooling. We've been to a site not too far from Corsicana, the Denton facility, Core Scientific, and they highlighted how many repairs have been completed in that first six months at that facility, and it was like 11,000. Are you seeing them, having run these businesses in both air-cooled and immersion cooling, are you seeing this massive drop down in sort of like requirements for the immersion cooling machines?
Well, I think we need to gather some more data on that because there are other variables at play here as well. We're also using these MicroBT machines now at a wider scale. That is now the vast majority of our fleet. What's great about their machines, they're very robust. Not only are they very competitive with what's on the market, but they're engineered to really high quality. So these machines in both air-cooled environments and immersion cooling environments are operating a lot stronger. So I don't have good data for you yet, and you know, it depends on what you qualify as a repair. Are you doing chip-level repairs, PSU, maybe just swapping out some fans? But you know, on that specific example alone, in immersion, you don't have fans. There's no fans on the miners to repair. That's a common thing you deal with with air-cooled miners. But to be fair, you know, then we have immersion equipment and there's repairs and maintenance on that that you maybe wouldn't have to deal with in an air-cooled facility. So there's pros and cons to each. There's not a right or wrong way, right? We operate both at pretty big scale. It depends on the specific application, the specific problem you're trying to address at that time, and kind of what the market looks like.
