Hi! This looks really good! Watched many of your tutorials and they are all amazing! I’m just wondering, do you not factor in goalie save %? Or is it already calculated in the “Def Rating”, if you understand my question..
Thanks for watching, lad 🔥 The average goalie save % is factored in as an average. If a team’s goalie has a higher save %, then the team’s DEF Rating will inevitably be lower (a good thing) as the team concedes relatively fewer goals. It’s not directly factored in, but the DEF Rating is a result of a team’s average save % among other factors 👍
Hi lad, excellent idea 🔥 I've recently been thinking about how to do something like this. Ordinal logistic regression may be the solution. In the short term I may make a video/model on this 💪
Although I'm not into hockey, I always find your models very interesting and inspiring. Question: wouldn't the Poisson distribution be more suitable for predicting Hockey scoring? Thanks. Cheers
G’day lad, thanks for watching 🙏 Yes, you’re right. However, Hockey is split into three different parts, with the last two (Extra Time and Shootout), ending when a particular team takes the lead. That’s why I had to use the BINOM.INV function to simulate the time of the goals scored, which POISSON.DIST doesn’t have the power to do by itself.
Hey man looking to join the patreon. Curious as to when I join and download your models does the finished model load up from the download or do you need to import the data yourself?
G'day lad 😃 Yes, the model is completely automatic. At the start of the video, I brought in the stats using a Power Query. This grabs the most up to date stats from Hockey Reference, and brings them in live to Excel. All you have to do is go into the 'Data' tab, and select the "Refresh All" button to update the file. I highly recommend you join lad, it's worth it 💪 And, once you're a member, you can ask me any questions you like about sports modelling and I'll get back to you ASAP 🔥
Hi lad, the Hockey data is from this link: www.hockey-reference.com/leagues/NHL_2024_standings.html I brought it into Excel using a Power Query, so that all I have to do is refresh the file and the stats update. This is all included as part of the Patreon Membership, check it out here: www.patreon.com/posts/99755288
awesome work on this, i did find one issue that i am trying to figure out how to write the code in for. in games that go to a shootout, say it is tied 2-2 in the sim. ot is still tied 2-2. in a shootout where multiple goals are scored, say 3 for the home team and 2 for the away team, the final score then presents itself as a 5-4 win as opposed to a 3-2 win. no matter how many goals are scored in a shootout the winning team is still only credited for one goal. I'm trying to figure out how to code this but have yet to succeed. any advice? thanks!
G’day, Terrell! Sorry, I don’t have the answer right now. I barely ever use Google Sheets, so I’m unsure if there is a What If Analysis substitute. I might look into this 👍
You make very good tutorials. thanks for that. But personally i think the simulations do not offer any benefit. Let's take a real match. Senators-Flames. The bookie offers me an odd for the senators. 2.06. When i run the simulation that gives me 1.92 so there is a little value. i run the simulation several times and it gives me : 1.96 - 2.22 - 2.50 - 1.67. so with 2.50 i have no value at all where on the other hand with 1.67 there is great value. So it just depends on the rand function when the program runs and you can not know if there is any value in the bet.
Thank you, lad 😃 I agree with your point, but for that result to come up means that there is either an error in your model or you're not running enough simulations. For a single simulation, the RAND() function randomises it according to the certain probability distribution you choose. However, after running thousands of these trials, the average of the results converge to the true probabilities. In most of my models I do at least 1,000 simulations. If I could, I would do 100,000! With 1,000 simulations, the variance is usually around 1-1.5% every "batch" of simulations". Meaning a team could go from a 50% chance of winning to a 51.1% chance of winning with a refresh. Anything more than 2% is too volatile, and you should run more simulations or look for an error in your model 👍 I hope that helps mate 🔥
Hello buddy! I bought a Patreon subscription yesterday. But for some reason, when updating data, “#value!” pops up almost everywhere. What could be the problem and how to fix this problem?
Hi lad, welcome 🔥 Great to have you! Usually these errors happen due to an accident in deleting a cell/formula while using the model. My first recommendation would be to re-download the file from the Patreon post. If the error is still there, please send a screenshot to me through the Discord. Of course, if you’re having trouble with this, or anything else, please don’t hesitate to send me a message through Patreon 💪 Let’s get your error fixed as fast as possible 🥇
G'day lad 🔥 Yep! All you have to do to update the stats is go into the 'Data' tab and select "Refresh All". This will update the model's connection with the internet, and bring in the most up to date stats!
G'day lads, if you have any questions please let me know below!
Pls I need the Excel
Hey im having troubles because of my teams now being in a different order then when you did it how can i get around this issues
hello, is it possible to do the same thing with shots and goals for the skaters or goalies and saves?
Your model cleaned up in College Basketball this season. Killed it
Finally figured this out and got it all working! The tables were importing with the incorrect data types - great video, thanks a lot!
Amazing stuff!
Hi! This looks really good! Watched many of your tutorials and they are all amazing!
I’m just wondering, do you not factor in goalie save %? Or is it already calculated in the “Def Rating”, if you understand my question..
Thanks for watching, lad 🔥 The average goalie save % is factored in as an average. If a team’s goalie has a higher save %, then the team’s DEF Rating will inevitably be lower (a good thing) as the team concedes relatively fewer goals. It’s not directly factored in, but the DEF Rating is a result of a team’s average save % among other factors 👍
Can the simulation portion be done in Google Sheets and if so how? Great video!
Great video as always! Would there be a way to simulate a motorsports seaspn? Like F1,Nascar,Indycar? Thank you!
Hi lad, excellent idea 🔥 I've recently been thinking about how to do something like this. Ordinal logistic regression may be the solution. In the short term I may make a video/model on this 💪
Although I'm not into hockey, I always find your models very interesting and inspiring. Question: wouldn't the Poisson distribution be more suitable for predicting Hockey scoring? Thanks. Cheers
G’day lad, thanks for watching 🙏 Yes, you’re right. However, Hockey is split into three different parts, with the last two (Extra Time and Shootout), ending when a particular team takes the lead. That’s why I had to use the BINOM.INV function to simulate the time of the goals scored, which POISSON.DIST doesn’t have the power to do by itself.
Hey man looking to join the patreon. Curious as to when I join and download your models does the finished model load up from the download or do you need to import the data yourself?
G'day lad 😃 Yes, the model is completely automatic. At the start of the video, I brought in the stats using a Power Query. This grabs the most up to date stats from Hockey Reference, and brings them in live to Excel. All you have to do is go into the 'Data' tab, and select the "Refresh All" button to update the file. I highly recommend you join lad, it's worth it 💪 And, once you're a member, you can ask me any questions you like about sports modelling and I'll get back to you ASAP 🔥
where did you get data from what website and do you have a templet? for this spreadsheet
Hi lad, the Hockey data is from this link: www.hockey-reference.com/leagues/NHL_2024_standings.html
I brought it into Excel using a Power Query, so that all I have to do is refresh the file and the stats update. This is all included as part of the Patreon Membership, check it out here: www.patreon.com/posts/99755288
Thank you
I always really enjoy your videos, but my sport of choice is soccer. Do you have much soccer content on your Patreon and do you have plans for more?
awesome work on this, i did find one issue that i am trying to figure out how to write the code in for. in games that go to a shootout, say it is tied 2-2 in the sim. ot is still tied 2-2. in a shootout where multiple goals are scored, say 3 for the home team and 2 for the away team, the final score then presents itself as a 5-4 win as opposed to a 3-2 win. no matter how many goals are scored in a shootout the winning team is still only credited for one goal. I'm trying to figure out how to code this but have yet to succeed. any advice?
thanks!
Any tips on how to do the simulation part on google sheets?
To clarify, google sheets does not have a 'what if analysis' button
G’day, Terrell! Sorry, I don’t have the answer right now. I barely ever use Google Sheets, so I’m unsure if there is a What If Analysis substitute. I might look into this 👍
@ thanks anyways👍🏽
Mad idea but have you thought about any tennis related models? I’ve always wondered if Poisson can be applied to total games or sets in tennis?
Hey where is the link for the patreon i want to download the model
Welcome lad! Don’t hesitate to ask any questions you may have through Patreon or Discord 💪
You make very good tutorials. thanks for that. But personally i think the simulations do not offer any benefit. Let's take a real match. Senators-Flames. The bookie offers me an odd for the senators. 2.06. When i run the simulation that gives me 1.92 so there is a little value. i run the simulation several times and it gives me : 1.96 - 2.22 - 2.50 - 1.67. so with 2.50 i have no value at all where on the other hand with 1.67 there is great value. So it just depends on the rand function when the program runs and you can not know if there is any value in the bet.
Thank you, lad 😃 I agree with your point, but for that result to come up means that there is either an error in your model or you're not running enough simulations. For a single simulation, the RAND() function randomises it according to the certain probability distribution you choose. However, after running thousands of these trials, the average of the results converge to the true probabilities. In most of my models I do at least 1,000 simulations. If I could, I would do 100,000! With 1,000 simulations, the variance is usually around 1-1.5% every "batch" of simulations". Meaning a team could go from a 50% chance of winning to a 51.1% chance of winning with a refresh. Anything more than 2% is too volatile, and you should run more simulations or look for an error in your model 👍 I hope that helps mate 🔥
Hello buddy! I bought a Patreon subscription yesterday. But for some reason, when updating data, “#value!” pops up almost everywhere. What could be the problem and how to fix this problem?
Hi lad, welcome 🔥 Great to have you! Usually these errors happen due to an accident in deleting a cell/formula while using the model. My first recommendation would be to re-download the file from the Patreon post. If the error is still there, please send a screenshot to me through the Discord. Of course, if you’re having trouble with this, or anything else, please don’t hesitate to send me a message through Patreon 💪 Let’s get your error fixed as fast as possible 🥇
Once I join and download NHL model do you keep it the model update always? Also do u have a college football and NFL?
Curious about this as well. Particularly if the data for the model is automatically in the file when we download it.
G'day lad 🔥 Yep! All you have to do to update the stats is go into the 'Data' tab and select "Refresh All". This will update the model's connection with the internet, and bring in the most up to date stats!
@@excel_ladz Are you building a MLB model yet?
Did I miss the Patreon Link?
ошибка впр