Jimmy Garoppolo, Football, Kevin Cole discussed on Niners Nation
Week we bring you a special conversation with a data scientists from pro football focus. It's Kevin Cole. You wrote an interesting article about Jimmy. Garoppolo using some baynes statistical things and I thought you know what? Let's talk a little bit about that because Jimmy Garoppolo's obviously been on my mind. I'm sure he's been on the mind of lots of forty niners fans after that Super Bowl and the question I kept coming back to over and over and over again was is. Is Jimmy. Garoppolo someone that you can win with is. He's someone that you can rely on. And is this someone that the niners may have to think about in some way shape or form replacing. Is he going to be good? What's the sample size? So many questions. Of course that that jumped into my mind and as I was thinking about that I came across an article on pro football focus. It took a look at what Jimmy Garoppolo could be and his upside and how he's got perhaps a bit more upside than you would think. And so I reached out to the author of article. Kevin Cole. He's a data scientist four. Pf and I thought you know what let's kick off the off season pods which with a bit of a look at the most important position and what that position could be for the forty niners so we are gonNA kick off our offseason content here in a couple of weeks David is taking some much deserved rest. I will also be taking some risks here in a little bit. But we're going to kick off with a talk about Jimmy. Garoppolo WE'RE GONNA have our roster model which will return again this season. We're going to have a free agency preview in just a couple of weeks. Those are going to be coming soon. But until then here's my conversation with Kevin Cole about Jimmy Garoppolo. And what statistics tell us about his future for the forty niners. Kevin Cole. Data scientists at pro football focus. It is a pleasure having you on especially on your vacation. How're you doing Kevin? I'm doing good. Yeah I was telling you earlier that I am in the bay area right here so I can really harness the local energy when talking about the forty niners. Yeah let's do that. Let's jump right into it because this is going to be really Jimmy versus Nick Mullins Pot. That's what we're going to get into No I'm just kidding. We have pleaded. You never have enough. Let's actually start at the top before we get into the all the talk about statistics and a little bit about what we can really glean about Jimmy Garoppolo about his performance from performance this season. Let's get a little bit more about you. I've got four questions for you. So listeners can figure out whether to immediately dismiss what you're saying or really leaning so question about the top Kevin Cole Witch Ninja Turtle. Are you guys spit? Spit Awhile since the since I've made I I don't remember. I believe Leonardo is kind of like leader. And maybe there's nothing you can really pick out is being that extreme so I think I will pick him if I even remember my turtles correctly. Which is a fifty fifty better best? Yes you are correct. Leonardo the Blue Ninja Turtle. Wasn't he leader and absolutely acceptable answer? Now this is going to be more one of my personal favorites. Best cheeseburger you've ever had go ever had come at me. There are quite a few places. I live in New York City so there are quite a few places here. I mean I think bang for your buck is someone who grew up in California in southern California. You know I love in and out Burger and you really can't can't beat that no. It's it's kind of a divisive topic during the election season visit by even more divisive documented number. So that as number one in New York there are quite a few places in. There's a place in Brooklyn. I believe it's called the manor where they had god that one of the best cheeseburgers ad that I've ever had their burgers in general there. That sounds amazing. I'm glad we're not talking about French fries because I would tell you to leave the podcast immediately back Bell French fries to which I know Israel. Okay you can get at it that I actually worked at in and out Burger of years. Yeah ad where the holy uniform hat everything it was. It was pretty. It's actually kind of funny story behind that. Because I interviewed once when at the beginning of a high school or mid mid to high school and I didn't get the job in that I learned that all you gotTa do is just be as friendly as possible like over the talk ridiculously friendly and they'll hire whether they think you're like dumb as a rock or anything so then. I just pretended to be really friendly. Once I got on the job I was kind of more of a grumpy guy in bringing everyone down there but I think the uniform itself is is a sight to behold. Especially when you have to wear that look at yourself in the mirror every day all right. What are you watching on TV right now? Television show and or movie not right the second obviously things that you're watching right now. I wish I watched more stuff. We I have a couple of young kids was staying up way too late so I'm watching that much recently. By the most recent thing we finish was the the documentary. Cheer on that Flex. A decent amount of hyper. It's following a cheerleading team to junior college in Texas. That's one of the best teams but I thought it was. It's actually very interesting. Because the the coach for the cheerleading team. She is you know. A Very well-known person in that industry does a lot of similarities between. I felt like football and cheerleading. Because they really get injured a lot and playing through pain and doing stuff like that or cheering fan. I guess in that circumstance. I thought it was a very interesting leadership style that she had because it wasn't very raw raw but it was really able to bring together a lot of these kids who came from diverse in very seldom very troubled backgrounds in kind of brought them all together there. So it's really interesting documentary. If everyone wants to check it out on Netflix in his I really enjoy watching to a few weeks back in. Yeah it was. It was definitely phenomenal. Would recommend last question. Do you love to win or do you hate to lose? I would say love to win more than hate to lose. I mean I basically just like to talk a lot of trash so and then if I lose who cares so So I think it's one of those things you gotta be able to talk trash in both circumstances so I still talk trash somehow. Even after I lose I'll find a way to do that so in that way. Winning is is that much better because you can really just pile it on at that point all right so that's a good. That's a good base for the soup. Now we know where you're coming from kind of no to basically disregard all of your opinions about fast food and but we're going to trust your thoughts and opinions when it comes to statistics because you wrote a really interesting article about Jimmy Garoppolo and you compare them to some of the players that have been rumored to to replace if you were going to go out and find a replacement for someone like Jimmy Garoppolo and you really grounded urinalysis and something. You did a couple of times before with one of your other articles and that's in Beijing. Statistics that's kind of where I wanted to start because it's something that you're going to see. I think it has been recently a bit more and more in the football space and so the first question off the top is really. What is our busy statistics? And why did you think that was a good methodology to use to evaluate the future performance or really to predict the future performance of quarterbacks? Yeah I would say. There are two major schools as far as Statistics in how. You're going to use them for predictive Trying to figure out predictions on on outcomes on basin and you have more of like a frequent tests sort of perspective and the big difference here is one would be gathering. A large enough sample is sound frequent. Aside where you're gathering a large enough sample the once. You have a large enough sample. You make a prediction based upon that sample. And you're hoping for something like a p value or one of these measures you can say this sadistically significant and that is really highly dependent upon just how big a sample that you have in other words. How many Throws someone has how many games they've played it depends on what on what you're measuring their whereas for Beijing and Statistics why. I think that it really works well with football which is notoriously has smaller samples because of the shorter season. And you don't have as many plays as you have possessions like you would in something like basketball. The reason that it works there. Is You have a few different elements but you really have a strong will what they call a prior. Which is your expectation before you start in each time. You see a piece of evidence. You adjust slightly based upon that. And then you have a new prediction which is based upon which was comes off of that evidence so the reason that it works well with football as you can have a prediction based upon very small bit of of sample size as opposed to if you just had someone who had thrown the ball. Let's say a hundred times and they're having such a career amazing years a rookie You still wouldn't be able to necessarily get anything that interesting off of the frequent perspective. Because you don't have enough sample to go on whereas you could actually take that in plug it in and try to properly weigh all these things in the basin statistics. So that's why I think it's interesting. Ultimately you've got a prior. Which is some evidence base belief? And then you add some current evidence and it gives you kind of like an updated projection or what sometimes called a posterior belief and it sounds like you are also able to draw on a much larger samples that because when you're talking about frequent tests you're saying okay. Well that one rookie. All you're looking at. His is his sample size. But when you're using something like basins that you can actually take the performance of all rookies or all quarterbacks and say that's GonNa actually inform my prior and I can use that as a basis to compare what this individual rookie is going to do. So you can get access to more information don't you? Yeah Yeah you get that so yeah I think just to build on a little bit further and some of the work that I've been doing on it. You not only have a prior for what? You're what you're exact. Prediction is but you also have a range of outcomes and that range is based upon the range that we've ended up seeing so we'll get into it more about rob lowe later but just to use the the quarterback for an example. You could say here are all..