Bitcoin's post-midterm pattern and hardware wallet security risks
A solo presenter on the Digital Asset News channel discusses Bitcoin's historical behavior around US midterm elections and warns about emerging hardware wallet vulnerabilities.
Summary
The host of Digital Asset News walks through historical data showing that Bitcoin has dropped to new lows after every midterm election cycle, though the time taken to reach those lows has been shrinking — from 71 days in 2014, to 39 days in 2018, to just 13 days in 2022 — suggesting any coming dip could be brief and sharp. He contrasts this with the strong post-midterm performance of the S&P 500, which has historically returned 10–17% in the six months following midterms regardless of which party wins.
The host then describes his dynamic dollar-cost averaging strategy, which scales purchase size inversely to a risk metric drawn from a set of analyst frameworks (including contributors he calls Ben, Ivan, and Wes), buying more aggressively as risk falls and reducing or stopping purchases as risk rises.
A significant portion of the episode is devoted to Block's new Desa AI voice agent. The host deliberately plays a clip of Desa without identifying it as AI, then reveals the deception to illustrate how convincing AI-generated voices have become — and how scammers will exploit this. He notes that CZ of Binance highlighted both the educational potential and the dangers of such technology.
The second half of the episode shifts to security warnings: a vulnerability in the Electrum hot wallet allowed malicious servers to intercept transactions before they reached the Bitcoin network and replace them with attacker-controlled ones using invalid signatures the wallet failed to detect, resulting in at least one confirmed theft. A data breach at Keystone hardware wallets exposed customer names, addresses, and order details — despite a stated 180-day data retention policy — with at least one user subsequently targeted by a scammer who referenced their personal order details. The host argues that AI is accelerating the scale and sophistication of these social engineering attacks by enabling cross-referencing of stolen data at scale, and recommends diversifying cold storage across multiple devices (he holds assets across a Trezor, a Ledger, and a Bitcoin ETF).
Key Takeaways
FULL TRANSCRIPT
Bitcoin's historical behavior around midterm elections
These are exciting times because we've got roughly one week until the midterm elections. The big thing we've talked about on this channel is just how well things go after the midterms — but there is a drawback specifically for Bitcoin before we get to some pretty big numbers, and that's what I'm going to talk about today.
Right now we're hovering around the $80K–$79K level, and we've got the midterm elections in roughly one week here in the United States. This is when everything goes pretty erratic volatility-wise, and it's because the traditional markets don't like ambiguity. They don't like the fact that they don't know where policy is coming from, because some people they give money to are potentially going to be voted out. Because of that, there is a lot of whipsawing in the markets. So what does that mean for Bitcoin in the long term — six months to a year after the midterm, and before that? That's the interesting part.
S&P 500 post-midterm returns
As far as returns go, let me show you something we've talked about for quite some time: the midterm election post time frame. Since 1942, buying the S&P 500 on midterm election day generated a return of plus 15% seven months forward. I always thought that was a bit odd the way they wrote it, and I can understand why — it's a lot easier to say "100% of the time everything works out" if you go seven months. But of course, everybody likes to play around with numbers.
If you just take a look at the normal six months to a year, there was one year — 1946 — when the S&P 500 did not perform six months later after the midterm elections. But as you can see over time, it really didn't matter as time went on.
The question then becomes: does it matter who wins? Is this going to be the president keeping both the House and the Senate, or is he going to lose the House, or maybe just the Senate? No, it really doesn't matter who wins or loses, who retains what. It's always the same thing — there's volatility, and then moving forward it plays itself out.
If the incumbent party loses the House — which happened in '46, '54, 1994, '06, 2010, 2018, and '22 — you got a six-month gain of almost 10%. If you lose the Senate only, which only happened in 1986 and 2014, you're up 11%. And if no chamber was lost, meaning they keep everything exactly the same, you're up almost 17%.
There have been a lot of pollsters saying it looks like the House is definitely going to go — so roughly 10%. The Senate probably will stay the same. So we'll probably see around 10%. But if nothing is lost, it could still be almost 17%.
Bitcoin's post-midterm pattern — and the drawback
Here's the big thing for Bitcoin. We've talked about this before. Six months after the midterm election, in 2014 Bitcoin was still down. In 2018, it was still down. In 2022, it was actually up almost 50%. But if you look a year later, everything was up.
Here's the thing I don't talk about enough, and I don't think anybody really says it clearly: in every midterm election — every single one, every four-year cycle — Bitcoin always goes down first. Now, is this time going to be different? We can look back and say that Bitcoin hit an all-time high before its last halving, which means some argue all four-year cycles are dead and defunct. But that's just one data point. Things could play out the same way.
In 2010 — the early years for Bitcoin — there was an early post-election drop of 71%. Don't worry, because that was when Bitcoin was a fraction of a penny to 8 cents. Back in those days it was difficult to keep that momentum alive when people thought it was internet nerd money and no one would be buying it. We were all proven wrong on that.
In 2014, Bitcoin bottomed 71 days after the election. In 2018, it hit a new low 39 days after the election. In 2022, it hit new bottoms after the election as well.
Now notice the time frames, because we're always looking for patterns. From 2014 to 2018 to 2022: 71 days to a bottom, 39 days to a bottom, 13 days to a bottom. The time is shrinking — 71, 39, 13. So if we do get a new low this cycle, I would hope and expect that we get there within roughly a week or so of the election. Now, that would be a significant drop, because the last low I can remember was somewhere around $57K–$58K — correct me in the comments section — somewhere around late June or early July. So if that's the case and we drop to that level, that is quite a big drop in a very short amount of time. But if it happens, it happens. And again, this has happened every single time in every four-year cycle. But if it doesn't, it doesn't.
Dynamic dollar-cost averaging strategy
That is why I dollar-cost average, but I do it dynamically, as we've talked about. It is Monday, which means I have to buy. I have a plan and I stick to the plan. The plan is every Monday, I buy using Cash App. Today the risk level was at 0.48, and because of that I am doubling what I usually do.
Here's how it works: as the risk levels go down, I buy more. As the risk levels go up and the price goes up, I buy less — or zero — and start to think about selling. In this framework, let's say you start with $100 between the 0.5 and 0.59 risk levels. If it goes to 0.40–0.49, like it is today at 0.48, you spend $200. If it goes lower, you spend $400, then $800, then $1,600. And if it ever goes to 0.00–0.09 — which has only happened in the early 2010–11 years, maybe 2012 — if that happens, you back up the truck. Unless AI decides to break Bitcoin's code and hack everything, but at that point everything else is going to hell in a handbasket, because that would essentially hack every bank account in the entire world.
As we've talked about before, I'm using what I call the big three for insurance: Ben for the risk layers, Ivan for the bull money line, and Wes for his bot. If I would have used those three plus a new aspect — which was suggested to me by Claude — and just looked at the 40-week simple moving average, instead of having $232,000 in this specific example, I would have had $559,000. The big thing it came down to is buying as things go lower, obviously, but that's not the issue. Everybody buys well. The problem that everybody has — including me — is taking profits. I'm going to rectify that this cycle.
Block's Desa AI and the implications for voice-based scams
Speaking of Cash App — Jack Dorsey owns Cash App, a subsidiary under Block — they just came out with something pretty interesting. It involves Block, Bitcoin, and AI. There were two people talking in a clip, and I want you to think about what they were saying about what's going on behind the scenes. Jack Dorsey is a huge proponent of Bitcoin, so when I first heard this I thought it was quite interesting.
Clip — Speaker 1: "I have been watching your journey of creating this. What did you think of it? Are you excited about it as I am?"
Clip — Speaker 2: "So blown away. I am so impressed."
Clip — Speaker 1: "It's so expressive, right? It's conversational — it can laugh, it can understand you and your emotions much better than anything I've seen out there."
Clip — Speaker 2: "I haven't really seen anything like that either, but it's pretty much human level."
Clip — Speaker 1: "Yeah, it's a bit scary to be honest, but I'm so excited to see where it would get us."
So I lied to you — sorry, but it's for your own good. That wasn't two heads of development talking about Bitcoin and AI. What that really was, was Desa. Desa was created by Block. The man talking was real. The woman talking was AI.
Usually when you hear AI, it sounds like AI. There are no pauses, no laughs, no corrections, no mistakes in the sentence — you can tell automatically. But when you were listening to that, you probably didn't think "this is AI." When I first heard it, I knew it was AI and I still couldn't believe it.
When you listen to something like that, imagine the things that are going to play out with AI — the good and the bad — moving forward. I believe AI is going to be great. It's going to push innovation, help eradicate certain diseases, help with socioeconomic inequality, and improve education. But like a knife or a gun, there are two ways it can cut. These are the things you have to be aware of, and it's coming faster than you think.
I had to set that up without telling you it was AI because when the scammers come, they're not going to say "hey, just so you know, this is AI." It's going to sound exactly like that, and they're going to try to take your money.
CZ of Binance made a pretty good point on this. He said people noted they'd never heard AI sounding so human, and he pointed to the potential use in education — specifically his own project, Giggle Academy, which aims to help underserved and underprivileged kids get a fantastic education. He said teaching kids with current AI voice was a challenge, but with this it's a much better outcome. He's right. But again, when you account for the negative potential, it is pretty jarring.
The "I don't know what I don't know" problem in Bitcoin security
The problem we have is we know what we know. I know I'm supposed to use cold storage devices. I know I'm not supposed to use a third party. I know I'm supposed to generate my seed phrase with sufficient randomness. I know what I don't know — I'm not a coder, I don't know exactly how everything works under the hood. But the bigger problem is I don't know what I don't know. Before the ColdCard wallet hack happened, you probably didn't know about that either. And I'm starting to think: what other stuff is out there that none of us know about yet?
Electrum wallet vulnerability
Here's one: "Bitcoin wallet bug lets rogue servers steal funds when users tried to speed up payments." A Bitcoin wallet security flaw allowed malicious servers to trick users into signing payments to attackers when they tried to speed up transactions, with developers confirming at least one case in which money was stolen.
Electrum, one of Bitcoin's oldest software wallets, disclosed the vulnerability, saying some servers were already exploiting it. Developers received two reports in October, including one from a user who lost funds after approving what appeared to be a routine transaction fee increase.
How does it work? An attacker controlling one of those servers could interrupt a payment before it reached the Bitcoin network and replace it in the wallet's history with a fake transaction. This replacement spent the user's coins to an attacker-controlled address using invalid signatures that Electrum failed to detect.
This is a hot wallet, so right there — you shouldn't be storing significant amounts in a hot wallet. That's something you know. But the scale of exposure here is significant. Electrum doesn't disclose its daily active user count, but on the Google Play Store it has over a million downloads, with trackers showing 1.3 to 1.4 million installs, and several sources listing it as having around 1 million users overall.
Imagine you've bought into Bitcoin as the future of finance, you try to spend some, and it's just gone. That's a terrible experience, and I can understand why people leave crypto because of it. There's a huge learning curve and a lot of pitfalls.
Keystone hardware wallet data breach
Here's another one: Keystone's website breach exposed hardware wallet customer data. Keystone — a cold storage device manufacturer — said it detected suspicious activity early in the week and confirmed after reviewing API activity with Shopify that an unauthorized third party had accessed customer shipping information, including names, contact details, shipping addresses, and order details such as items purchased.
There's also a question about Keystone's handling of older customer data. In a 2023 blog post, Keystone said it had reduced its customer data retention period to 180 days — which is still six months, a long time to hold data, though at least they purge it after that. But customers said they received breach notifications despite having purchased their devices two years ago. One user going by OG DGEN said they received a call from a scammer claiming to be law enforcement investigating a fraud case. The user suspected the caller obtained their information through a Keystone order placed in 2024, based on personal details the caller referenced.
And this goes back to what we just talked about. It takes time for a scammer to look up all your information — to find a name, cross-reference a zip code, see that the median income in that area is over $200,000, figure out that this person probably has money and owns a Keystone or a Ledger device. That takes time manually. But here's the problem I see now with AI: it's not just one hacker or ten hackers doing this work. AI does all of this and runs everything through it, and they're able to extract and cross-reference at scale. It doesn't have to be one person — it could be the equivalent of 10,000.
Diversifying cold storage
The takeaway is simple: diversify your cold storage. I don't know what's going to happen, so I have a Trezor, I have a Ledger, and I have some ETF exposure. Social engineering hacks are ramping up — you can see how much was extracted in 2025 and what's happened in 2026 so far. I think it will actually outpace previous years.
On the ETF side: if you have Bitcoin and want to move into an ETF, you can't do an in-kind transfer the way some large ETF holders can. You'd have to sell your Bitcoin, pay tax on the gain, and then buy the ETF. But if you go through certain providers, they do handle it in-kind — you just move your Bitcoin from wherever you hold it, they take custody of it, and you hold the ETF. Providers like iShares use Coinbase, Fidelity, and Fireblocks as custodians, the same as the other major players.
Q&A
Question from chat: Rob, do you buy via Cash App?
Yes, you can buy via Cash App, you can sell it, and you can offload it — which is what I do. I send it to the cold storage locations we just talked about.