NBA Betting Model
Most NBA models take a team and give you a number.
This one starts with the players who are actually on the floor.
Set the lineups, and the model works out expected minutes, possessions and points from those players before building the game spread and total.
So when a starter is ruled out an hour before tip, you do not have to guess what that news is worth.
Change the player in the dropdown.
The numbers move with the lineup.
I have used this model myself across multiple NBA seasons.
Across the periods I tracked, I recorded 604 bets at approximately 3.57% ROI.
That is my historical record.
It is not a promise of yours.
What you are buying is the tool and process that produced the numbers.
One-time payment · Lifetime access · 30-day money-back guarantee
A normal NBA night
Look at the right side of the sheet.
Most games say:
Not Qualified.
That is not a flaw.
That is the model doing its job.
The point is not to manufacture action. The point is to create an independent number, compare it with the market and leave the game alone when the difference is not large enough.
Most NBA bettors start with:
“What do I want to bet tonight?”
This starts with:
“Is there anything worth betting tonight?”
A look inside
Just the sheet and the process, sped up. This is what you are buying.
Why it is built this way
Five players are on the court.
That changes the problem.
In soccer, one missing player is one of eleven.
In baseball, one hitter is one of nine.
In basketball, a star sitting can change the entire shape of the game.
A team-based model still sees the Celtics as the Celtics.
But the Celtics with a key starter sitting are not the same team.
And when injury news breaks close to tip, you often have less than an hour to decide what the new number should be.
Most recreational bettors know the injury matters.
The problem is that they have no way to price how much it matters.
So they guess.
This model gives you another option.
You set the lineup.
It calculates the expected contribution of the players you entered.
Then the projected spread, total and player expectations are rebuilt from those lineups.
Where it works
My own tracked NBA betting record with this model is built around spreads.
Not player props.
Not trying to bet every market on the board.
Spreads.
That is not a weakness.
It is focus.
You do not need to have an opinion on every market.
You need a market where you can produce a number independently, compare it with the book and decide whether the difference is large enough to matter.
The model gives you a projected spread for every game.
It also gives you totals and expected points for players.
Those are useful outputs to inspect and develop.
But the tracked record I am showing you comes from the spread side.
So start there.
Inside the sheet
Projected spread and total
A projected spread and total for every game, based on the lineups you set.
Three calculation methods
The model calculates the matchup more than one way, so you can see whether the answer holds up or one method is driving the entire result.
Expected points for every player
Because the game projection is built from individual players, the sheet also produces expected possessions and expected points.
I treat this as a bonus output rather than the core product.
My own tracked record is on spreads.
So use the player numbers as a starting point to explore, not as a player-prop record I am claiming.
Qualifying criteria
Every matchup is compared with your thresholds automatically.
Instead of staring at ten games and trying to invent action, you can immediately see which differences are large enough to investigate.
The work
No coding.
No formulas to write.
Nothing to install.
Once you know the workflow, a full NBA slate takes roughly fifteen to twenty minutes.
If you know your way around Google Sheets, nothing important is locked away.
You can inspect formulas.
Change thresholds.
Experiment.
Build on top of it.
The routine
Before the season, write down two numbers.
Your goal.
And the point where you stop and reassess.
My own example was:
The exact numbers are yours.
The important part is deciding them before emotions are involved.
If the stop number is something you cannot genuinely tolerate, you will start making decisions to avoid the pain instead of decisions based on the price.
Then execute.
Run the slate.
Take what qualifies.
Pass on what does not.
The NBA season is long.
That gives an edge room to appear.
It also gives emotion months to interfere with the process.
Deciding your rules before the season is easier than inventing them in the middle of a losing week.
Real bets
I have used versions of this NBA model across multiple seasons.
I did not bet it every season and I did not formally track every season.
Across the periods I did track, the record contains:
That is historical performance.
Not a forecast.
Not a guarantee.
Different thresholds, sportsbooks, timing and lineup decisions can produce different bets from the same spreadsheet.
The reason I show the record is simpler:
this was built to bet with.
It was not a spreadsheet created afterward to make a sales page.
From a buyer
Brian bought the model and wrote back after using it.
What I like about his message is the second part.
He noticed how sharp NBA lines actually are.
That is an important lesson.
A model does not make the market easy.
It gives you a structured way to measure when your number is different enough to care.
The criteria
Spreads
5+ points
Default threshold: 5 points or more between the model spread and sportsbook spread.
Inside five points, I treat the difference as too small to care about.
The purpose of the threshold is not to find more bets.
It is to eliminate marginal ones.
Totals
15+ points
Default threshold: 15 points or more.
Totals qualify very rarely.
That is intentional.
I would rather the model tell you PASS nine times than manufacture a weak bet because you want action.
Both thresholds are visible and editable in the spreadsheet.
Bet size
The qualifying threshold answers the first question:
Is this difference large enough to consider?
The size of the difference answers another:
How strong is the discrepancy relative to the threshold?
The sheet can use a simple staking ladder so that a game barely clearing your threshold is not automatically treated the same as a game with a much larger gap.
This is not a promise that bigger model differences always win more often.
It is simply a structured way to avoid treating every qualified bet as identical.
You can use the default approach or change the staking rules yourself.
Read this before you buy
Not a picks service
The model gives you numbers and qualifying thresholds. You still decide whether to place the bet.
Spreads are the main product
My tracked record is on spreads. Totals and player expected points are useful additional outputs, but I do not present them as having the same tracked record.
Lineups matter
If the lineup is wrong, the output is answering the wrong question. You need to check who is actually playing.
Google Sheets
This is a Google Sheets model. Use it on a laptop or desktop. It is not designed as a phone-first betting app.
The sharpest markets are hard
This is not a promise that you will outprice the sharpest NBA market every night. It gives you a repeatable method for creating an independent number and deciding when the difference is large enough to investigate.
Variance exists
The model can lose over any future sample. Historical results do not guarantee a profitable season.
The actual change
Right now you can look at an NBA game, know who is playing well, read the injury news and form an opinion.
Then you look at the sportsbook.
But without a number of your own, you cannot really answer whether:
−5.5
is cheap, fair or expensive.
You have nothing independent to compare it with.
That is a tools problem.
After this, the workflow changes.
Open the sheet.
Set the lineups.
Create your projected spread.
Then look at the market.
Most nights that process should eliminate bets.
Sometimes it leaves one or two.
Either way, you now have a reason.
That shift happens the first time you run a slate.
It does not require copying my picks because there are no picks to copy.
Fit
This is for you if
Skip this if
What you get
NBA Model
Google Sheets model with:
Instruction video
A walkthrough showing how to load the data, set the lineups, run the slate and read the output.
Bonus
American Odds Bet Tracker
Bonus
Decimal Odds Bet Tracker
Both trackers let you record your bets and compare the price you took with the market close.
Lifetime access
One payment.
No annual model subscription.
You keep access to the model and walkthrough.
You keep the spreadsheet and training permanently.
The workflow uses current-season data, and the walkthrough shows you how to carry the process into future seasons.
Small bonus
50 complimentary AI Assistant credits.
The Underdog Chance AI Assistant is a separate product. The 50 credits let you try it on real games. Buying the NBA Model does not start an AI subscription. Nothing renews automatically. Credits are added manually after purchase.
Price
$299
One-time payment.
Lifetime access.
No recurring model fee.
30-day money-back guarantee
Buying several models? All Models is $997 and includes the complete model library plus A Journey training. $1,920 if purchased separately.
See all models →Delivery
Your access is connected to the email address you use at checkout.
Go to My Library.
Enter the same email.
We send you a magic sign-in link.
Click it and your NBA Model content unlocks.
Inside you will find the model spreadsheet, walkthrough and included resources.
Make your own copy of the Google Sheet into your Google Drive.
No software to install.
No password to remember.
Support: mb@underdogchance.com
Guarantee
Open the model.
Run some slates.
Change lineups.
Compare your numbers with the market.
If you do not see the value, email me within 30 days and I will refund the purchase.
No questions.
No call.
No proof required.
In their words
Brian
On the NBA model after buying it
Member feedback
On getting set up and learning the models
Terry
On the videos and the numbers behind the odds
Authority
Comments from people with real experience in sports betting and statistics.
Dave Cokin
Dave Cokin is a longtime sports handicapper and respected voice in sports betting.
“Definitely worth reading for anyone trying to make their own numbers.”
“Worth the wait. Good job with your models once again!”
David Beaudoin · Professor MJ
David Beaudoin, known as Professor MJ, is a PhD in statistics and professor at Laval University in Quebec City.
“His vision of sports betting is exactly like mine: you absolutely need to mix statistical models with your own judgement in order to be successful in this business.”
“If you only watch games and don’t run any numbers, you are very likely to lose in the long run. If you only focus on numbers and you don’t take into account important factors like injuries and scheduling, you are also likely to run into some trouble.”
Professor MJ originally wrote this about my betting work and the earlier A Journey Betting Masterclass. This is historical authority and methodology proof, not a review of the NBA model.
Questions
Bring in the required data, set the lineups and enter the night's matchups. The sheet calculates projected spreads and totals from those players and also produces expected player numbers. It then compares the game against your qualifying criteria.
No. The calculations are already built. You need to be comfortable using Google Sheets, copying data and checking lineups.
Usually around fifteen to twenty minutes once you know the workflow.
No. The model is designed around freely available statistics.
It produces expected points for every player in the lineups you set. I consider that a useful bonus output and a place to explore. My own tracked NBA record is on spreads, so I would not buy the model for player props alone.
No. It creates a number and applies your thresholds. You decide whether the sportsbook price is worth taking.
5 points or more on spreads. 15 points or more on totals. Both are visible and editable in the spreadsheet.
Because NBA player availability can materially change the game. When someone is out, change the player in the lineup and the model recalculates from the new combination.
Yes. It is a Google Sheet and the important assumptions and thresholds are visible. You can inspect it and develop it further.
Any sportsbook can be compared against the model. Having access to more than one book can help you shop for a better price.
Yes. The model lives in Google Sheets and the underlying workflow is not limited to one country.
You keep the model and walkthrough permanently. The workflow uses current-season data, and the training shows you how to carry it into future seasons. There is no annual model renewal.
No. You receive the spreadsheet and training so you can use the workflow yourself. I am not promising to personally maintain every customer's copy each season.
Closing line value compares the price you took with where the market closed. Individual bets can win or lose for many reasons. Tracking the price you took against the closing market gives you another way to evaluate whether your process was consistently finding better numbers.
No. The 604 tracked bets and approximately 3.57% ROI describe my historical record across the periods I tracked. They are not a guarantee of future performance.
Yes. All Models is $997 and includes the complete model library plus A Journey training.
Email me within 30 days of purchase and I will refund it.
mb@underdogchance.com
Last thing
That full-slate screenshot near the top is the point of this model.
Ten NBA games.
Most of them:
Not Qualified.
A couple worth looking at.
You do not need more action.
You need a number of your own.
Set the lineups.
Run the slate.
Compare your number with theirs.
Then decide.
One-time payment · Lifetime access · 30-day money-back guarantee
Stop following picks.
Start pricing bets.