Hi, I am not the best at statistics so this may not be a very good question, but if you want projections for bets like over, under, and runline then why use an interquartile range to create a high and low? I understand that it is to keep the extremes from skewing data, but if you're eventually running 10,000 different outcomes to base the percent chance for different scores then the extremes are already accounted for as they are rare occurrences. I imagine you may have tried that or knew better, but I wanted to know if the specific reason. I suppose my follow up question would be what happens if you don't create a high and low but use another factor like starting pitcher and bullpen to keep extreme values from skewing results? In any case the model is very interesting, and I enjoyed the video!
Great content, and well explained, thanks Benjamin . I'm not sure, but to update the sheet, is it done when we open it or do we have to do something else to activate the update on scraping the data? Update: sorry, I just understood (hopefully), it's when we select the game, that the data is collected, is it?
Hello, I am following your work and I find it very interesting. I have a copy of your previous work and if it is not a bother, could you share the link to the finished sheet?
@@TheBenjaminWagner Great work. I'd be interested in looking at one that accounts for starting pitchers and lineups since that affects the outcome. I'd love to follow along because I'm sure you can figure it out!
First of all, love the model. I’ve been messing around with it for a couple days now. I was wondering if there is a way I could add statcast ballpark factor into the model to make the runs scored even more accurate. Im not very savvy with sheets but would love to learn more!
@@TheBenjaminWagner Thank you for everything, I also would love this! Your videos are great! Would you possibly be open to discussing custom models with particular advanced metrics as a service? I have a few I really like to incorporate with some ideas, however, I'm not great with the computer portion of the model
@@TheBenjaminWagner There are some that are considered more “important” vs others such as DVOA (defense adjusted-valued over average), EPA (expected point average), success rate (success of QB passing vs def, success of WR vs defense, etc). Just to name a few… Most oddsmakers and professionals use these metrics, but My thought is to build a model similar to what you have while taking these advanced metrics, perhaps create a correlation (maybe R script?) through regression analysis to determine some type of weighted benefit of each metric for its impact on the game and final score. That’s where I think would differentiate things…. I would be happy and very grateful to connect with you to discuss more if interested.
@@JorgeDiaz-xs5ou my prior video had pitchers as well. You could combine the two sheets and make a real nice model. Here’s the link to that one: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-_9SVNmc1nHo.htmlsi=aaQWC25AQo2hNv-T