Let me get the math out of the way first. I had a $15,000 gambling swing in about 30 days. Now, let me explain how an EM attending who writes a personal finance blog about retiring at 40 ended up there and why I think every physician with a high income and an iPhone is one bad month away from the same story. Let me tell you the cautionary tale I should have read six weeks ago.

First, here’s a quick word on who’s writing this, since $10,000 means different things depending on who’s losing it.

I’m an emergency medicine attending in Texas, two years into my career and married with one kid. My wife works from home and pulls in around $100,000 a year, so we’re a dual-income household. But the EM paycheck is the engine. As of this writing, our net worth is around $750,000 at age 32. I run a personal finance blog called RetireAt40MD where I document the path. The name is the bumper sticker version of the goal. The actual goal is financial independence by 40 but not necessarily retirement. The distinction matters to me. I want the option to walk away from a shift without needing to ask the question of how I’m going to pay the mortgage. Whether I actually walk away when I get there is a separate question I’ll answer in eight years.

The target FI number is $5 million. At a 4% safe withdrawal rate, that’s $200,000 a year in spending, which is what I figure a very comfortable life in Texas costs over a long retirement. Our current savings rate is around 50% of gross. My student loans will hit zero by January 2030 on the current paydown schedule.

I didn’t grow up with money. I’m a first-generation immigrant, and my parents never made more than $40,000 a year in any single year of their working lives. We were on food stamps. Every financial instinct I have was built by watching adults make do with not enough. The whole point of saving the way I do is to make sure my kid grows up with the resources I didn’t have and to have an easier life than I did.

So when I tell you I lost $10,000 in a month playing parlay games on my phone, understand that I am exactly the kind of person who should have known better.

A Quick Glossary If You Don’t Gamble

Skip this if you already speak the language of gambling. For everyone else, you have four things to know.

Odds

American odds come in two flavors. A negative number like -110 tells you how much you have to risk to win $100. So, -110 means you have to bet $110 to win $100. A positive number like +600 tells you how much you win on a $100 bet, so +600 means betting $100 to win $600. The more negative the number, the bigger the favorite. The more positive, the bigger the underdog. Even money is roughly -100, a coin flip in odds terms.

Player Props

You can bet on an individual player’s stat line instead of the game itself. For example, you could take LeBron James to score over 26.5 points. You could take Patrick Mahomes to throw over 1.5 touchdown passes. The book sets the line, and you take the over or the under.

Parlays

This is the one that got me. A parlay stacks multiple bets onto a single ticket, and every leg has to win for you to cash. If you miss one, the whole ticket is dead. Going 2 for 3 on a three-leg parlay pays exactly the same as going 0 for 3, which is nothing. Going 3 for 4 pays the same as 0 for 4. The payouts look enormous because the probability of stacking three or four independent outcomes is small, and the books that set the odds know retail bettors love chasing those numbers.

Self-Exclusion Tools

This is a feature most sportsbooks and prediction markets are required to offer. You go into the app’s settings and ask to be locked out for a set period (a month, a year, or permanently), and the platform blocks you from depositing or betting until that period ends. The reason it matters more than just deleting the app is that re-downloading restores access. Self-exclusion stays attached to the account no matter what you do with the app icon on your phone. Reversing it is intentionally hard, which is the point.

More information here:

How I Got to Plus $5,000

I’m not going to pretend I started gambling out of stupidity. I started for the same reason most docs do. I had fun, the apps looked legitimate, and I genuinely thought I was good at it. I participated in NFL Sundays at the bar with friends, a Kalshi market on the next Fed meeting, a Polymarket bet on something I’d been following in the news. It was $20 here, $50 there. It felt closer to fantasy football with stakes than to gambling.

To be fair, this wasn’t my first time near a table. Gambling for me started in high school on a game called RuneScape, in the duel arena, which is exactly as ridiculous in print as it sounds. After that, it was blackjack and roulette at casinos when I could get there. None of it was big money in dollar terms, because I didn’t have any. Twenty dollar bets felt like huge stakes on a resident salary.

But the pattern was there the whole time. In RuneScape, I had stretches where I’d lose a duel, decide I needed to win it back, and gamble bigger and bigger until I was out of in-game gold. The only thing that kept it from ever costing me real money was that I never had real money to lose. I’d tell myself afterward that I’d learned the lesson. But it never stuck.

The sports betting run was the first time the pattern met an attending paycheck. It went exactly the way you’d expect.

For the first stretch, I was up. I hit a couple of parlays, called a couple of macro events on the prediction markets, and pulled $5,000 of withdrawals back into my checking account. I told my wife about it. I told the group chat about it. I started tracking it in a spreadsheet, which is the move of someone who already thought he was running a business.

You can see where this is going. Every single thing I did to “build on my edge” was actually the apps slowly teaching me to bet bigger.

How I Got to Negative $10,000

I knew, on some level, that I didn’t really have an edge. I’d hit a hot streak, and hot streaks end. I knew that the same way I know any other math I’ve learned. But knowing it and feeling it are different things.

The crash didn’t happen in one bad night. It was 30 bad days strung together.

It started with bigger units. When you’re up $5,000, a $200 bet feels small. Then, a $200 bet that loses feels like nothing because you’re still up $4,800. Then, you chase. Then, the chase loses. Then, you tell yourself you’ll get back to even and come out ahead, which is the exact thought every losing gambler in history has ever had.

Single bets would have been the smarter play, and I did stick to them at the start. The first stretch of being up was mostly straight bets with a few small parlays mixed in. The shift happened the way doses escalate on anything. A $50 single bet pays you $45, so after you get your initial wager back following a win, you end up with $95. A $50 three-leg parlay pays $300. When you’re already up and feeling smart, the size of the payout works on you faster than the math of the probability. I told myself the parlays were where my “research edge” actually mattered, because the books supposedly hang softer lines on player props than on game spreads.

Maybe that’s true at the margins for people who actually know what they’re doing. It wasn’t true for me.

Here’s a rough breakdown of where the money actually went:

  • ~85% NBA parlays. Player props, same-game parlays, late-night West Coast tip-offs I had no business watching, let alone betting on.
  • ~10% UFC parlays, mostly fight night cards where I’d convinced myself I had a read on the matchups.
  • ~5% NFL parlays, which is funny because NFL Sundays were how I started gambling in the first place.

Sometimes, I had a “feel” for how the night would go. The feel was wrong. The feel is always wrong eventually.

With the $15,000 swing, I was now down $10,000. That’s one month of post-tax savings as an attending that was gone. To watch it leave the account in the same app where I’d watched it arrive a few weeks earlier was an experience I will not forget.

More information here:

Why Doctors Are Uniquely Susceptible to These Apps

I think physicians get hit harder by sports betting than the general population. It’s less about the income we make and more about the wiring inside us that got us here.

We’re trained to believe we can analyze our way through anything. We had years of standardized tests where the right answer existed and we had to figure out where to find it, years of clinical reasoning where pattern recognition paid off. The apps are designed to feel like that. You have stats, and you have injury reports. You have, in your head, a hypothesis. The platform reinforces this constantly. The interface looks like a research tool, even if it’s more of a slot machine.

There’s a second piece to this. Doctors spend their entire careers being the person in the room who knows more than the other person. The patient doesn’t know what’s wrong, and you tell them. The student doesn’t know the answer, and you correct them. The sports betting app is the rare interaction where the other side, the oddsmakers, will always know more than you do. They have models, they have data, and they have real-time information about how the serious bettors are moving money. You have an injury report and a hunch. Most physicians have no muscle memory for being the dumber party in a transaction, and the brain refuses to register what’s actually happening.

Now, add shift work. Add the post-shift adrenaline crash where you can’t sleep and you’re scrolling at 2am. Add an income that lets you absorb a $500 loss without immediately feeling it in your bank balance, removing the natural circuit breaker that protects people with less margin. Add the fact that almost nobody talks about losing, because the only people who post about gambling on physician forums are the ones with a good week.

You have a population that’s overconfident, sleep-deprived, financially insulated from short-term pain, and surrounded by silence. Doctors have that profile.

The Math You Ignore

Every sportsbook and prediction market builds in a vig, which is the cut the house takes on every bet. It’s why a 50/50 bet doesn’t pay even money. On a typical -110/-110 NFL spread, you have to risk $110 to win $100, even though the spread is set to make the two sides roughly a coin flip. That extra $10 is the book’s edge, and across millions of bets, it guarantees the house comes out ahead regardless of who wins any single game. The implied break-even rate at -110 is 52.4%, meaning you have to be right more than 52.4% of the time just to not lose money.

A “sharp” is someone who bets on sports for a living. They run full-time models and follow strict bankroll discipline, and the best of them hit something like 53%-55% on their best lines. We are not sharps.

Parlays are worse. A three-leg parlay paying +600 has a true fair value closer to +700 or +800. Every leg you add multiplies your edge problem and the book’s edge advantage. Player props on FanDuel and DraftKings carry some of the highest holds in the industry, often 6%-10%, because the books know retail bettors love them and don’t shop lines (and, therefore, aren’t looking for the best deals from different sportsbooks).

Prediction markets are slightly different. Kalshi and Polymarket have tighter spreads on liquid markets. But the moment you bet on something niche, the spread widens, and you’re paying it both ways. The interface makes you feel like a trader. You are not a trader. You’re a retail customer paying a market maker for the privilege of being wrong.

I knew all of this six weeks ago. I have a finance blog. I wrote about expected value at one point. But the apps don’t care how much math you know. The only way you can defeat them is not opening them in the first place.

What Quitting Actually Looked Like

I deleted DraftKings, FanDuel, Underdog, Kalshi, and Polymarket in one sitting. I went into each one first and used the self-exclusion tools, which most of the apps have buried two or three menus deep.

I told my wife the full number. That was harder than deleting the apps. She didn’t yell. She got quiet, which is the thing she does when she’s actually upset rather than just frustrated. After a beat, she said something like, “That’s a business class trip somewhere. That’s a week in a villa. That’s a memory.” She wasn’t trying to twist the knife. She was just saying it out loud, but it landed harder than if she’d been yelling. The disappointment, not the loss of our money, was the actual punishment.

I told two friends in the group chat. That was harder than telling my wife, because the group chat is where a lot of the bets started. They were good about it. One of them is, I suspect, going to do the same thing in a few months. I hope it’s even sooner.

More information here:

What I Want Every Physician Reading This to Know

You are not smarter than the book or the people who are setting the odds. Your training does not transfer. Your income does not protect you; it actually accelerates the damage. The apps are not entertainment platforms with a gambling feature. They’re gambling platforms with an entertainment skin.

If you bet recreationally and you’re honestly net positive over a year of tracked activity, fine. Most people who say this aren’t tracking honestly. I wasn’t. If you’ve been thinking about quitting and you’ve been telling yourself you’ll quit after one more good week, that’s the spreadsheet talking. There is no good week coming. That good week was what hooked you.

I’m out. The retirement target at 40 just got $15,000 further away, and I’m writing this so the next physician with a hot start and a polished app reads it before they write the same post in six weeks.

[EDITOR’S NOTE: If you or someone you know is struggling with gambling, the National Council on Problem Gambling helpline is available 24/7: call or text 1-800-GAMBLER, or visit ncpgambling.org]

Have you used gambling apps or prediction markets? How have you done with them? Has it ever been a problem for you or anybody you know? What did you do?

Did you find this article helpful? Share it!