The one number that tells you if you're any good

You went 7 and 3 last week. Felt like you'd figured something out.

Then you went 3 and 7 the week after, and now you're not so sure.

Here's the thing nobody tells you when you start betting: neither week meant much. Not the good one, not the bad one. And that's not me being negative. It's math, and once you see it, you'll never look at your bets the same way again.

Because there's a better number to watch. Sharp bettors have watched it for years. It's free, it's on every bet you've ever made, and almost nobody betting from an app on Sunday knows it exists.

Your record is mostly luck. For a long time.

At standard juice, -110, you need to win 52.4% of your bets just to break even. That's the bar.

Now picture someone with no edge at all. Every bet is a coin flip. Over 100 bets, how often does that person win?

Anywhere from about 40% to 60%. That's the normal range for pure luck over 100 bets. A coin-flipper can look like a genius or a disaster, and nothing about how they picked games changed.

So it takes something like a thousand bets before your record starts telling you the truth. If you bet a few games a week, that's years.

Nobody waits years. You decide you're good after a heater, or you quit after a cold month. Both are reacting to noise.

You need a number that answers faster.

The closing line

The closing line is the last number before the game starts.

Think about everything that's happened by then. The injury report is out. The weather's in. The starting lineups are set. And every serious bettor in the country has had days to bet into that number and push it where they think it belongs.

By kickoff, the line has soaked all of that up. It's the most informed guess anyone has about that game. Not perfect. But better than any one person, including me, and including you.

That's why it makes such a good measuring stick.

Beating the close

Say you bet the Bills -3 on Tuesday. By kickoff Sunday the line is Bills -4.5.

You got the Bills a point and a half cheaper than the market's final answer. You beat the close.

Flip it. You bet Bills -3, it closes at -1.5. Now you paid a point and a half more than the market ended up thinking the game was worth. The close beat you.

Sharps have a name for this: closing line value. I'm just going to keep calling it beating the close, because that's what it is.

And here's the part that matters. Whether that bet wins or loses doesn't change the answer. The Bills can lose by 10 and you still made a good bet at -3 if the market agreed, by kickoff, that -3 was too cheap.

Why it tells you more than winning does

A win or a loss is one coin landing. Heads or tails, mostly luck.

Beating the close is different. Every single bet gives you a reading on it, a little plus or a little minus, and those readings aren't mostly luck. They're a measure of one thing: did you get a better price than the sharpest number in the market?

If you keep doing that, you're consistently buying games for less than they're worth. Over time that turns into money, because the closing line is right on average. You don't need a thousand bets to see the pattern. It shows up much sooner, because you're not waiting on the coin.

And if you keep getting worse prices than the close? Your record might look fine for a while. It usually doesn't stay that way.

Stop asking "did I win?" Start asking "did I get a better number?"

Which closing line?

Not all books are equal here.

The book you bet at moves its line partly based on what its own customers are doing. That's not the market's best guess, that's one store's guess.

The number to measure against is a sharp book. Pinnacle is the standard. It takes big bets from serious players and moves fast when they hit it, so its closing line is about as close to the market's real opinion as you'll find.

One more step, and it's the one people skip. Every price has the book's cut built in. You have to take it out to see the real number.

My own bets, the honest version

Those Pinnacle numbers above are real. That was Celtics at Pistons on opening night, and I had the Celtics.

I took them at -118. That's a 54.1% win chance baked into my price.

By tip-off, FanDuel had moved to -122. So against my own book, I beat the close. Felt good.

But Pinnacle's fair number at the close was 52.9%. I paid 54.1%. Against the sharp line, I paid about a point too much. And the Celtics lost.

Next night I had the Warriors +7.5. The line closed at +6.5. Looks like I got a full point better than where the market finished.

I didn't, really. That +7.5 was an alternate line. I bought the extra point, and I paid for it: -124 is a 55.4% win chance, against 52.4% at the standard -110. So the point cost me 3 points of win chance.

Why buy it at all? Because the point I bought was 7. Going from +6.5 to +7.5 only changes anything if the Warriors lose by exactly 7. That turns a loss into a win.

In the NFL, 7 is a famous key number, and you should almost always pay to get on the right side of it. Of the last 2,826 NFL games, 8.5% ended on exactly 7 (and 14.7% on 3). In the NBA it's flatter. I checked the last four seasons, 5,108 games: 7 came up 5.2% of the time. Every margin from 2 to 9 lands between 5.2% and 5.7%. Basketball scores move in too many small steps for any one number to stand out.

But when a favourite is laying about 7, the single most likely margin is right around 7. The game lands on it about 3 times in 100. So the point was worth roughly 3 points of win chance, and I paid 3. Fair, not a steal. Call it even with the close.

The Warriors lost by 22.

So my record from those two says 0 and 2. The close says one slightly bad price and one fair one. Neither was the great bet it felt like when I made it. That's exactly what the close tells you that the record never will, and yes, even my own bets get an honest read.

How to check your own bets

1. Write down your price the moment you bet. The line and the odds.
2. Check the closing number right before the game starts. Pinnacle if you can see it, a sharp book or an odds screen if you can't.
3. For spreads and totals, compare the points. You took +7.5 at -110, it closed +6.5 at -110: you beat it by a point. If you bought that point on an alternate line, it isn't free. Count the extra juice you paid for it, and ask how often the game actually lands on the number you bought.
4. For moneylines, turn both sides of the closing price into win chances, take the book's cut out, and compare that to the win chance in your price.
5. Keep the running average. Not the last bet. The average over everything.

Do that for a couple of months and you'll know more about whether you're good than a full season of wins and losses would tell you.

What it doesn't tell you

I'm not going to oversell this, because that's the whole point of the page.

One bet tells you nothing. You can beat the close and lose for a month straight, that's still the coin.

Lines sometimes move for bad reasons. A wave of public money on a big name can drag a number that the sharp money hasn't touched. Measuring against Pinnacle, not your own book, takes most of that out, but not all of it.

Some markets barely have a real closing line. Player props and small markets don't get bet hard enough for the close to mean much. This works best on sides and totals.

And if you get really good at it, some books will notice and cap how much you can bet. That's a good problem, but it's real.

Why it's built into everything here

This is why the site works the way it does.

Every pick on the record is logged before kickoff and graded against where the line closed. Every bet I make gets checked against Pinnacle's fair closing number, wins and losses both. When I'm getting worse numbers than the close, you'll see that too.

Our number isn't a pick. It's a reference. The point is to know what a game is worth before you bet it, so you can tell when the price is good.

Win or lose, beat the number. The wins follow the number, not the other way around.