Excel LADZ
patreon
How to Use the ‘Excel LADZ - NBA Model’
🕑 Added 2024-11-03 00:44:59 +0000 UTCG’day, lads! In this article, I’ll run through the steps required to make daily predictions using the ‘Excel LADZ - NBA Model’. If you haven’t already become a member, and you are interested in being able to back your tips with statistics, then download the model below!
Excel LADZ - 2024-25 NBA Model: https://www.patreon.com/posts/114792748
1. Updating the Model
To ensure you’re working with the latest stats daily, the model is linked to Basketball Reference via Power Query for automatic data updates. Instead of manually copying and pasting new data each day, simply go to the ‘Data’ tab and click the “Refresh All” icon. This action refreshes the model’s connection to Basketball Reference, bringing in the day’s stats seamlessly.
This will update two worksheets: “2023-24 Raw Player Data” and “2024-25 Raw Player Data”. The “Player Data” worksheet will then automatically pull from both sheets to calculate each player’s 30-game stats.
2. Selecting a Matchup & Forecasting Score
To choose a matchup, select an Away and Home team from the dropdown lists in the “Results” worksheet. After selecting each team, allow 10-30 seconds for the data to update, depending on your Excel version and computer speed.
After selecting each team, the Win % and Expected Score will appear. The Win % is calculated by dividing the number of trials won by each team in the “Simulations” worksheet by 1,000 (the number of trials). On the right, you can input the current score, quarter, and time remaining. For instance, with 5:00 left in the third quarter and a current score of 70-60 in favour of the Hornets, the Win % and Expected Score will adjust accordingly.
3. Adjusting Lineups to Account for Player Injuries
Sometimes, a player included in the simulations may not be playing in a particular game. For example, Miles Bridges might be rested against the Celtics. Below is the current roster, assuming the strongest lineup is selected and no players are injured. Don’t be concerned if an error appears - this simply shows that, six games into the season, Charlotte has only used 12 players, so there’s no 13th player available in the Basketball Reference dataset to fill that slot.
To exclude Miles Bridges, set his xMP (expected minutes played) to zero. This change is best made in the “Player Working” worksheet, where I’ll overwrite the formula that pulls in his minutes with a “0”. I’ll highlight the cell in yellow as a reminder to restore the original formula for the next matchup.
With Miles Bridges expected to miss the game against the Celtics, the simulated lineup in the “Results” worksheet has been adjusted accordingly. Bridges’ xMP is set to zero, fully excluding him from the Hornets’ lineup. As a result, the Hornets are approximately 2 points weaker, and their win probability has slightly decreased to 16.7% across the 1,000 simulations performed.
Last season, I published a video tutorial on this topic: https://www.youtube.com/watch?v=WBqpxYZeSU4&list=PLIuPNiYeH9ytCvP9KJAZWwdef17_kVXBk&index=2
4. Analysing the Moneyline, Spread and Total Points Markets
Once the final lineups are confirmed for both teams, the “Betting Markets” worksheet analyses the simulations to calculate probabilities for the moneyline, spread and total points markets. After selecting your preferred odds format (American or Decimal), you can set the Total Points threshold in cell B7 and the Spread in cells B14 and B15. The resulting odds represent the fair value of each market based on the simulations. If a bookmaker offers higher odds, the model indicates that the market may be undervalued.
For a more detailed explanation behind the maths behind all three markets, check out this article I’ve written: https://www.patreon.com/posts/nba-model-93427951
5. Rinse and Repeat!
If you found any value in this article, make sure to join hundreds of others and become an Official Excel LADZ! Not only do you gain access to the NBA Model, but also the NFL model, Soccer Model, NHL Model, and more! You’ll also be able to join the Discord community we have and ask any question on sports modelling, Excel or statistics you may have.
I’m also planning on adding a few new features to the NBA Model, including:
Live Odds
Confirmed/Expected Lineups
Thanks for reading, lads!
Comments
Excel LADZ
Hello, Bastian 👋 Welcome! That's unusual... Send me a screenshot through Discord or my email so we can get this fixed up for you. Here's my email: ladz.excel@gmail.com
Bastien Lecas
hi, I've just subscribed and the world documents launch well but as soon as I update there are value errors.
Excel LADZ
Hi Jacob! Yes, that's definitely wrong. The model should not show that, 60 expected points is absurdly low. Based on your other comments, I suspect it's an issue with how the power query is being loaded into your Excel. Tomorrow I'm releasing a second NBA Model that relies entirely on team stats rather than being a player-based model like this one. In the meantime, send me a screenshot of your issue via Discord or email and we can work it out mate 😉
Jacob Labonté
Also “refresh all” doesn’t do anything, whenever you enter the 76ers Joel Embiid doesn’t show up. Can you please be more clear on how you run this?
Jacob Labonté
And yeah sure I follow your directions, but according to how you work your models the pelicans are only scoring 60 points a night
Jacob Labonté
Just wondering, do you have expect to be profitable on this…. Doing this math and your models are hitting at less than a 55% ratw
J
Hey are we still getting this feature? The confirmed and expected lineups would 100% be amazing!
Excel LADZ
Just posted something like that 💪 Check it out, and let me know what you think 😃 Pretty basic ELO model, but the "Warren Nolan" website is usually very reliable 🔥
Didier Cobos
thank you very much for your attention
J
Hey will there be a college basketball one this year?
Excel LADZ
That’s great! To update every player’s stats, you must select the “Refresh All” icon in the ‘Data’ tab. This updates everything in the spreadsheet. I do this once a day. However, updating a cell (for example, selecting a team in the “Results” worksheet), this updates everything in the spreadsheet except the player stats. That is, every time you change a cell the simulations will update and take about 4-5 seconds. This may be a bit annoying unfortunately. There is a way to temporarily disable the simulations and update the spreadsheet much easier (before running the simulations again to see each team’s win probability). I might do a Patreon tutorial video on this shortly 👍 Does this answer you question, mate?
Didier Cobos
Thank you very much I was able to use it on another computer, just one doubt when I put the equipment in I have to update it again
Excel LADZ
Hello, Didier 😃 The NBA Model is the most complex model on the Patreon due to it being entirely player-based. That, is the simulations take a long time to load. With the normal file, this is about 5-10 seconds and for the ‘Legacy’ file, about 20-30 seconds. If after this period of time the file still hasn’t loaded, then your computer may be a bit slow. If the issue persists, don’t hesitate to send me a screenshot of your issue through Discord 🔥 Or, you can send it straight to my email: ladz.excel@gmail.com If we still can’t solve it, let me know if you’d like a refund 👍
Didier Cobos
I already purchased the templates, but it doesn't load the templates, it comes out blank, is it because my Excel is in Spanish or why is it?
Excel LADZ
Hello, Didier 👋 Most of the models are self-updating. That is, once you download the spreadsheet, you need to select the "Refresh All" button, and that will automatically update all the stats necessary 👍 This includes the NBA, MLB, NFL, European Football , ELO and NHL Models 😉
Didier Cobos
After having obtained the templates, it is only updated or what should we do about it to predict. Thank you
Excel LADZ
Thanks Mark 🙏 I've just edited the files in this post to patch a few slight errors. Some spots in a team may of come up with a "#VALUE" error previously. This meant that a spot (out of 13) couldn't be filled by the model. For example, if a team hasn't used a "PG" as defined by the Basketball Reference dataset so far this season, the model couldn't match the PG position to a player for that team and would show the error. This doesn't affect the model though, as the other players are given an Expected Minutes Played figure (xMP) to sum to 240 all together (5*48). In the latest release, I've edited a few formulas so that it shows a blank row instead of a messy "#VALUE" error 👍 Try re-downloading the model and seeing if that helps, lad 😃 Let me know if you have any more questions.
Mark D
I should point out. Its the legacy versions I'm referring to
Excel LADZ
G’day, Mark! Welcome 🔥 I’ll check that out and get back to you soon mate 👍
Mark D
Hey. When using the 'live win probability, some players are coming up with #value. Any reason why? I've refreshed the data
Excel LADZ
Hi, Cody! I was planning to use the 2023-24 defensive stats for the first 30 games of the season, then switch it to 24-25 afterwards. This is mostly so that luck factored into the ratings in the first few games of the season can be reduced 👍
Excel LADZ
G'day, J! Very soon, within a week 😃 From now until Christmas, a LOT of content will be released 🔥
Excel LADZ
G'day, Nate! There's always a way with Excel... 🔥 I'm planning on adding in a power query to bring in lineups to the NBA Model. However, I don't know if I'll make it automatically adjust the lineups. Rather, it might be information for the user to make the changes themselves. This would be to keep the model simple. If you really wanted to make it automatic, you could 👍 I did this with the Player-Based MLB Model. First, you would have to bring in the Power Query. Then, you need to come up with a custom formula to match the players in the starting lineup with the players in the team's dataset. This part is usually more difficult than it sounds, as different datasets/websites name players differently. I know it sounds ridiculous, but one dataset might have "Nikola Jokić" and another "Nikola Jokic", which wouldn't initially match. Once you've overcome the logistical issues, then the starting lineup would be produced. If you give it a go, don't hesitate to let me know if you have any questions. Apologies for the late response this time, for the rest of the year I'm able to respond pretty much within a day.
Excel LADZ
Hello, Didier! Yes you can 👍 🔥 By becoming a member, you can download every model on the Patreon.
Didier Cobos
I am from Mexico, can I buy the Excel templates?
cody muncy
One question. Why in the team stats it has all of last years team stats instead of this year?
J
Hey Lad! Any idea when the new features will be added? The Live Odds and the Confirmed/Expected Lineups? Thanks in advance….
Nate Contreras
Hey Lad! Is there a way to automate adjusting lineups for injuries? Just curious
Jacob Labonté
Thanks, cheers
Excel LADZ
G'day, Jacob! Apologies, 15:00 was a typo in the model I uploaded. That's correct, 1 and 12:00 will predict the game 👍
Jacob Labonté
Should I keep the time left at 1 and 12:00 if I'm wanting to simply predict the game? Or do I need to enter a different number, as example 15 was saved for me.
