Learn
How many sportsbooks should you have? The Moving Bar answer for one pick
How many sportsbooks should you have? Enough that one pick gets more than one price, because every price sets its own bar your pick has to clear. In a hypothetical where you rate an over at 51.5 percent, the bar is 52.38 percent at -110 and 51.92 percent at -108, so you are under both, but at -105 it drops to 51.22 percent and the same opinion clears it. Nothing is guaranteed.
The Moving Bar is three steps you run on one pick before you place it. It is the table from our video on line shopping sports betting, written out so you can copy it, and it answers the question most bettors ask the wrong way: not which book is best, but what bar each book sets for the bet you already want.
- Write down your number. Before you open any app, decide how often you think your side wins, as a percent.
- Turn every book's price into its bar. Each price has a win rate it needs to break even. That is the bar your pick has to clear at that book.
- Bet only where your number clears the bar. Compare your number with each book's bar, and take the book with the lowest one. If none is below your number, skip the bet.
The rest of this page goes through each step with the formula, the worked example from the video, the full bar table from -120 to even money, and the mistakes that make the comparison lie to you.
Who this is for
It is for a bettor who forms an opinion first and looks at prices second, which is most of us. You like an over, you open your usual app, you bet it. The Moving Bar adds one check in the middle. You need accounts at two or more legal books where you live, a way to write down a number before you look, and a minute per bet. It does not pick sides for you, and it does not make a weak opinion strong. It only stops you from paying a price your own opinion says is too high.
How many sportsbooks should you have
Enough that the bet you want usually comes with more than one price. One book gives you one bar, and you either clear it or you don't. Two books give you two bars and a choice. Each book after that is one more chance that somebody is a few cents cheaper on your side at that moment.
The useful test is simple. Keep a note for a month of how often each book had the best price on the bets you actually placed. A book that is never the best on your markets is an account you can stop checking. A book that is often the best is the one that lowers your bar. The goal is the lowest bar on your bets, not the most apps on your phone.
1. Write down your number
Your number is your own estimate of how often the side wins, said as a percent. It might come from a model, a rating, or your read of the matchup. The rule is that you write it down before you see the prices, because once you see -105 it is very easy to decide you always thought it was a 52 percent shot.
Be honest about how rough it is. A number like 51.5 percent is a claim that the side wins a little more than half the time. If you can't tell 51 from 53, say so in your notes, because the next two steps only work as well as this one.
2. Turn every book's price into its bar
The bar is the price's implied probability, the win rate it needs to break even. For a negative American price, divide the price without its minus sign by the price plus 100. For a positive price, divide 100 by the price plus 100. This is the standard moneyline conversion: at -110 you risk 110 to win 100, so the bar is 110 divided by 210, about 52.38 percent.
Here is the bar for the prices you see most on sides and totals, worked with that same conversion. The last column is a hypothetical: the expected result per 100 dollars staked if your number is right at 51.5 percent, using the expected value formula (chance of a win times the profit, minus chance of a loss times the stake).
| Price | Profit on a 100 win | The bar | EV per 100 at 51.5% (hypothetical) |
|---|---|---|---|
| -120 | 83.33 | 54.55% | -5.58 |
| -115 | 86.96 | 53.49% | -3.72 |
| -112 | 89.29 | 52.83% | -2.52 |
| -110 | 90.91 | 52.38% | -1.68 |
| -108 | 92.59 | 51.92% | -0.81 |
| -105 | 95.24 | 51.22% | +0.55 |
| -102 | 98.04 | 50.50% | +1.99 |
| +100 (even money) | 100.00 | 50.00% | +3.00 |
Read it top to bottom and the bar slides down about half a point for every two or three cents. That is the whole idea behind the name. Your pick sits still while the bar moves under it.
3. Bet only where your number clears the bar
Now put your number next to every bar you found. This is the example from the video, and it is a hypothetical, not any book's real line. You rate an over at 51.5 percent and check three books. One has it at -110, so the bar is 52.38 percent and you are under it. One has -108, a bar of 51.92 percent, and you are still under. The third has -105, a bar of 51.22 percent, and the same opinion clears it by about a quarter of a point. All three bars come from the moneyline conversion.
Nothing about your pick changed between the three books. If you had only one app and it was the -110 book, the honest answer would be to pass. With three, you have a bet, and by the expected value math the hypothetical result moves from about minus 1.68 dollars per 100 to about plus 55 cents per 100.
Two rules keep this step honest:
- Clear it by a margin you believe. A quarter of a point is a thin edge on a rough number. If your estimate could easily be off by a point, a bar that is a quarter point below it is a coin you are still flipping.
- No clear, no bet. If every bar is above your number, the right move is to wait or skip. Shopping found you the cheapest price, and the cheapest price was still too high.
Why the bars differ at all
Each book sets its own prices and adds its own margin. The mathematics of bookmaking calls the amount the two sides' implied chances add up to above 100 percent the overround, which is the book's cut. When books disagree on a side, the cheapest one is the one whose margin on your side happens to be smallest at that moment.
That margin is not small. A measured panel of bookmaker margins covering 52 bookmakers from 29 June to 2 August 2026 found averages of 4.77 percent on head-to-head markets, 4.90 percent on spreads, 5.45 percent on totals and 7.93 percent on player props. The higher the margin on a market, the higher the bars tend to sit, and the more a cheaper book can matter.
Mistakes that make the bars lie
- Different lines. An over 220.5 at one book and over 221.5 at another are different bets. Compare bars only when the line matches, or treat the half point as its own question.
- Stale prices. A price from an hour ago may be gone. Check all your books for one bet within a minute or two of each other.
- Moving your number after you see the board. If -105 makes you suddenly sure it is a 53 percent shot, step 1 didn't happen. Write the number first.
- Props treated like sides. Props carry more margin on average, 7.93 percent in the margin panel, so even the lowest prop bar can sit well above a rough number.
- Forgetting the book can say no. Researchers who bet real money on a strategy built from bookmakers' own odds reported that the books began to restrict their accounts (Kaunitz, Zhong and Kreiner, arXiv). The lowest bar is worth less if that book will only take a small bet.
How to tell it's working
Add three columns to whatever you use to track bets: your number, the bar you took, and the highest bar on offer for that bet. If you are running the Moving Bar, your number is above the bar you took in every row, and the bar you took is the lowest one you found. The gap between the bar you took and the highest bar is what checking saved you on that bet. It says nothing yet about whether your numbers are any good; that takes a much larger sample.
If you want every book's price for a bet on one screen instead of three apps, the odds comparison board lines them up with the best price per side, and the line shopping page explains why the same bet can cost less at a different book.
Quant+ is analytics for informational purposes. Nothing is guaranteed. 21+. If you or someone you know has a gambling problem, call 1-800-GAMBLER.
Questions people ask
How many sportsbooks should you have?
Enough that the bet you want usually has two or more prices to choose from. Each extra legal book adds one more bar to compare, and the lowest bar is the one your pick has to clear. Past the point where new books rarely beat your best one, more accounts add work without lowering your bar.
What is the bar for a price?
It is the win rate that price needs to break even over time, the implied probability of the odds. For a negative price, divide the price without its minus sign by the price plus 100. At -110 that is 110 divided by 210, about 52.38 percent.
Does line shopping make my picks better?
No. Shopping never changes which side you like or how often it wins. It changes the price you pay, which moves the bar your pick has to clear.
Is a pick that clears the bar a sure winner?
No. Clearing the bar means your own estimate is above the price's break-even rate. Your estimate can be wrong, and even a correct one loses often in the short run. Nothing is guaranteed.
Why can a two-cent difference matter?
Because the bar moves with every cent. In a hypothetical with your number at 51.5 percent, -108 has a bar of 51.92 percent and -105 has a bar of 51.22 percent, so the same pick is under one and over the other.
Sources
- Moneyline odds, Wikipedia
- Expected value, Wikipedia
- Mathematics of bookmaking, Wikipedia
- Bookmaker Margin Panel: Daily Measured Betting Margins Across 22 Leagues and Player-Prop Markets, Mendeley Data (E. Glush)
- Beating the bookies with their own numbers, and how the online sports betting market is rigged, arXiv (Kaunitz, Zhong, Kreiner, 2017)
