How to actually make money with Iron Marker
The scanner finds the prices. It does not place the bets, size the bankroll, or sit through the losing weeks for you. This is the part nobody writes down: what to set, what to bet, how much, and how you know it is working before the money says so.
What you are actually doing
You are not predicting games. You are buying prices that are better than the sharpest market's price, over and over.
Iron Marker scans every major US sportsbook plus offshore books and exchanges every 2 minutes. It compares each price to Pinnacle's no-vig fair line and lists the bets where a book is paying more than fair by at least your minimum EV. That is the whole product. Nothing about matchups, injuries, or trends.
The edge on any one bet is small. 2 to 4% is typical on mainlines. On a $50 bet that is one or two dollars of expected profit. The money comes from volume and discipline, not from any one bet, and it takes hundreds of bets before the expected profit is bigger than the noise.
The whole playbook in one line: find the edge, size it small, bet a lot of them, judge yourself on closing line value, not on last night.
Before your first bet: the three things you need
1. A bankroll you can lose, kept separate
Not your checking account. A dedicated pot, spread across your sportsbook balances, that you would not miss if it went to zero. Under $500 is not worth the time. The edge is a percentage, and a percentage of a small number is a small number. $1,000 to $5,000 is where most people start.
2. At least 3 sportsbook accounts, ideally 5+
This is the one people skip, and it is the one that costs the most. Our Cost of Loyalty study measured what betting at a single book costs versus taking the best available price on the same bet: $20 to $47 per $1,000 staked. DraftKings loyalists paid the most, about $33 per $1,000. That leak is the same size as the edge you are trying to capture. One book cancels the product.
Which books? Our Closing Line Value by Book study scored 6,920 flagged bets against Pinnacle's no-vig close. The flagged prices that held up best were at BetRivers (median +2.70% CLV, 81% of flags beat the close), Bally Bet (+2.6%), and DraftKings (about +2%). Have those funded if they are legal where you are. FanDuel sat at the bottom (+0.72%, about 56% beat the close), which tells you its mistakes get corrected fastest. It is still worth an account. Just bet it from an alert, not from a nightly review.
3. 20 minutes a day, or Telegram alerts on your phone
Edges appear and disappear on a clock measured in minutes. Either you check the scanner a few times a day at the times lines get posted, or you let the Telegram bot tell you when something clears your threshold. The second is better. The first still works.
Set up Iron Marker in 5 minutes
All of this lives in Settings. Do it once, then leave it alone for two weeks.
- State. Pick yours. The app filters to the books that are legal there so you are never chasing a price you cannot bet.
- Bankroll. Enter the total across all your book balances. Every stake the app shows is sized from this number.
- Kelly fraction: 0.25. Quarter Kelly. Full Kelly is the stake that maximizes growth if your edge estimate is exactly right, and it is never exactly right, so full Kelly turns a slightly overestimated edge into brutal drawdowns. Quarter Kelly gives up a little growth for a lot less pain. The Kelly guide has the full argument.
- Minimum EV: 2%. Our by-book study found flagged mainline prices beat the close by a median of about 1.4%, so a 2% floor leaves room for the number to be a little wrong and still be worth betting. Lower and you are mostly betting noise. Higher and you cut your volume in half.
- Max odds: +300 to start. Long shots inflate variance, and a flagged edge on a +900 price is more likely to be a phantom created by a stale number than a real mistake.
- Telegram alerts on, alert threshold 3%. Paste your chat ID in Settings and press Start on the bot. The scanner keeps flagging at 2%; your phone only buzzes for the 3%+ ones so you actually look when it does.
- Mainline markets only for the first two weeks. Moneyline, spread, total. Props have bigger edges and bigger vig and faster limits. Earn a clean CLV read on mainlines first, then add props.
Your first week, day by day
| Day | What to do | Where you should be |
|---|---|---|
| 1 | Settings above, fund the books, place 3 to 5 flagged bets at the stake shown, press Track on each. | Familiar with the row: book, price, fair probability, EV %, stake. |
| 2 to 4 | Bet every flagged edge above your threshold at the sized stake. Track every one. Skip anything that has already moved. | 10 to 20 tracked bets. Do not look at profit. |
| 5 | Open the Analysis tab. Ignore win/loss. Look at average CLV and what share of bets beat the close. | A first read, probably noisy. Positive is good. Negative on 15 bets means nothing yet. |
| 6 to 7 | The weekend slate is where most volume is. Bet from alerts. Take the price shown or better. | 25 to 40 tracked bets and a real first CLV read. |
The only number that matters at the end of week one is the CLV column. The profit column is a coin flip at this sample size, and the next section is why.
The math of why your first week can be red and you are still winning
Take a 3% edge, 1-unit flat stakes, and bets near even money. Each bet is worth +0.03 units in expectation, and each bet swings about 1 unit either way. Expected profit grows with the number of bets. The noise grows with the square root of the number of bets. That gap is the whole story.
| Bets | Expected profit | 1 SD of noise | Typical range (1 SD) | Chance you are down |
|---|---|---|---|---|
| 30 | +0.9 u | ±5.5 u | −4.6 to +6.4 | ~44% |
| 100 | +3 u | ±10 u | −7 to +13 | ~38% |
| 300 | +9 u | ±17 u | −8 to +26 | ~30% |
| 1,000 | +30 u | ±32 u | −2 to +62 | ~17% |
Read the 100-bet row again. A real 3% edge, executed perfectly, leaves you with roughly a 1 in 3 chance of being down after 100 bets. Mixed odds push the noise a little higher still, about 1.0 to 1.1 units per bet. Nobody's first week proves anything. Nobody's first month proves much.
This is why the app grades every tracked bet on closing line value instead of on the result. CLV measures whether you got a better price than the market's final, sharpest number. It is available on every bet, not just the ones that cashed, so it converges hundreds of bets sooner than profit does. The CLV guide covers the mechanics.
The rule. After 50 tracked bets, if your average CLV is positive, you are on the right side of the math regardless of the profit column. After 200, if CLV is still positive and profit is still negative, that is variance. Keep going. If CLV is negative, stop. Your process is broken: betting too late, wrong books, or chasing long shots.
Speed is the whole edge
A flagged bet exists because one book has not caught up to the sharp market yet. That condition lasts minutes, sometimes seconds on big games. The scanner refreshes every 2 minutes. The window between a flag and the book fixing it is where all the value lives.
- Bet from the Telegram alert, not from a nightly review. A list of edges from three hours ago is a list of prices that no longer exist.
- Take the price shown or better. The EV % on the row is computed at that price. If the book now shows a worse number, the edge is smaller or gone.
- If the line has already moved, skip it. Do not bet the worse price because you already opened the app. There will be another one.
How to not get limited (and what to do when you are)
Books limit winners. Not maybe. If you consistently take their mistakes, the risk desk will notice and cut your max stake, sometimes to a few dollars. This is the tax on doing it right. Some habits slow it down.
- Round your stakes to normal amounts. $25, $50, $100. A stake of $47.13 on a stale line reads like a bot.
- Do not max out every bet. Betting the book's exact limit on every mispriced line is the fastest way to a limited account.
- Mix in some normal-looking bets. A few small, ordinary wagers on the game everyone is watching make the account look like a fan's.
- Withdraw sensibly. Frequent withdrawals right after wins are a flag. Let balances sit.
- Do not only hit the same book's mistakes. Spread the volume. Five accounts limited slowly beats one account limited fast.
When a book limits you, that is not a failure. It is the sign it worked. Move your volume to the next book and keep the limited account for small bets and whatever ordinary action you want. The arbitrage guide covers how quickly limits arrive when you are also hitting the other side.
What "profitable" looks like at your bankroll
Here is the honest expectation. Assume a 3% average edge, quarter-Kelly stakes that work out to about 1% of bankroll per bet, and 150 bets a month, which is a normal football-season pace on mainlines with a 2% threshold.
| Bankroll | Typical unit | Bets / month | Expected profit / month | After the $39 sub |
|---|---|---|---|---|
| $1,000 | $10 | 150 | about +$45 | about +$6 |
| $5,000 | $50 | 150 | about +$225 | about +$186 |
| $10,000 | $100 | 150 | about +$450 | about +$411 |
Those are expectations, not promises. Any single month will land well above or below them, per the variance table above. And notice the first row. At a $1,000 bankroll the subscription eats almost all of the expected profit. Pro at $39 a month only makes sense at roughly a $1,500+ bankroll with real volume. Below that, do not pay us. Use the free calculators, build the bankroll, and come back. We would rather say that here than have you find it out on your statement.
Who this does not work for
- One-book bettors. The loyalty leak is the same size as the edge. There is nothing left.
- Parlay players. Every leg compounds the vig. The scanner cannot fix a parlay.
- Anyone under $500. The math works, but the dollars are not worth your time or the subscription.
- Anyone who will not track bets. Without CLV you cannot tell a broken process from a bad week, and you will quit at the wrong moment.
- Anyone who needs to be up every week. Reread the variance table. A third of your first hundred bets will end in the red even when everything is right.
Frequently asked questions
How many bets a week do I need to place?
Enough that variance can't fool you. A realistic floor is 25 to 40 tracked bets a week, which is roughly what a 2% minimum EV on mainline markets produces during football and basketball season. Fewer than that and 200 bets takes months, so you'll be judging your edge on noise. More is fine as long as every bet clears your threshold and is sized by the app.
Do I have to bet every flagged edge?
No, but you should bet every one that clears your threshold, is still available at the price shown, and is at a book where you can get down. Skipping bets on gut feel reintroduces the handicapping you're trying to remove. The one filter worth applying is the max odds cap: long shots inflate both the variance and the chance the flagged edge is a phantom.
Which sportsbooks should I open first?
Open whatever is legal in your state, then fund the books whose flagged prices held up best against Pinnacle's close in our data: BetRivers (median +2.70% CLV, 81% of flags beat the close), Bally Bet (+2.6%), and DraftKings (about +2%). FanDuel's edges get corrected fastest, so treat it as a book you bet from an alert, not from a nightly review. Three accounts is the minimum. Five or more is where line shopping really pays.
What EV threshold should I use?
Start at 2% minimum EV on mainline markets (moneyline, spread, total) with a max odds cap of +300, and set your Telegram alert threshold at 3% so your phone only buzzes for the better ones. After two weeks of tracked bets, look at average CLV in the Analysis tab. If your 2 to 3% bets are not beating the close, raise the floor to 3%. If they are, leave it, because volume is where the money comes from.
How do I know if Iron Marker is working for me?
Look at closing line value, not profit. After 50 tracked bets, positive average CLV means you are on the right side of the math whatever your profit and loss says. After 200, positive CLV with a negative profit and loss is variance, so keep going. Negative CLV over 200 bets means the process is broken: you're betting too late, betting the wrong books, or chasing long shots. Fix that before adding volume.
Start the 30-day playbook.
7-day free trial, then $39/mo. Card required at signup, cancel any time before day 7 and you're never charged.
Start your 7-day free trialIron Marker is an analytics tool, not a sportsbook, and this guide is educational — not betting or financial advice. Odds shown are illustrative. Must be 21+. Problem gambling? Call 1-800-GAMBLER.