Sports Modeling Excel Logistic Regression
For this video, we are going to cover logistical regression. Now whether you are betting on the NBA, NHL, NFL, or even soccer, you can use this kind of modeling.
I personally will use logistical regression as a primary model, or to feed into some of my prop models. For sports betting, it should be a staple in your tool box.
When you are doing logistical regression, it is very similar to linear regression, however rather than looking at a continuous outcome, we are looking for a discrete outcome. That being it is a binary result. It is pass, fail. It is true, false.
In the video today, we are going to demonstrate a scenario where we believe one basketball player causes a team to win or lose. By using the basic data of points, rebounds, and assists, we will model out the probability for the team to win give a player gets a specific number of points.
I will be building this in Microsoft excel, however you will want to also install solver, as that is going to be a critical piece to this model. The other component is setting up what your minimum value will be for qualifying probabilities. What this means is that after you build the model and complete some testing, you should evaluate how well it is working, then set a line for what results you will keep. For some models you may want to keep values if it returns a .5 probability. Others, you may want to keep if it is .6
Пікірлер: 14
Great video as always. Very informative.
@wageredontilt1649
5 ай бұрын
Hopefully it helps with your predictions!
Would you create a model that predicts wins or losses in the NHL using Logistic regression? I would love to see it.
@wageredontilt1649
5 ай бұрын
I would be able to do a basic one, however it would be more conceptual to get you started on ideas.
When I add the IF statement, it gives me a #REF error for the 0 and does just fine for the 1...help
@wageredontilt1649
3 ай бұрын
Is the IF for the p()? If so, it sounds like there is a reference key mistaken. Can you copy and paste the formula you have into a comment? I can take a look from there.
haha. Why does my data analysis go in reverse? Better states = lower probability?
@wageredontilt1649
3 ай бұрын
It would depend upon what you set your pass fail probability too, as well what you mean better stats = lower probability. Do you have an example of data you see that happening with?
Dam man good stuff I mainly learned ill never be smart enough to beat the books.
@wageredontilt1649
4 ай бұрын
Well, you can if you can attack weaker points. Don’t go after main lines like sides and totals. Try out player props, 1H or periods. Many books have too much surface area to cover all it well
Very useful, thanks. I'm mostly interested in soccer but unfortunately Logistic Regression works with binary outcome only, as you explained. There are other variants of LR (multinomial LR) to deal with more than two possible outcomes (e.g. win/tie/lose) but I've never seen it done in excel. Cheers
@wageredontilt1649
5 ай бұрын
Correct, for for Multi you can look at using Ridge. For logistical, this can be used as will player X make a goal. And then use that into another model to see scores. Just depends on what angle you want to attack the solve.
Will this work on google sheets?
@wageredontilt1649
5 ай бұрын
Yes, this will work in google sheet, you would need to just install a solver. If you are in your google sheet, you would be able to get a plug in like Open Solver.