I think you misunderstood my original post (which was the original post of our discussion). When I said people don't like what they don't understand, I was referring to people like those who don't like seeing coaches go for it on 4th down because they don't understand the reasoning behind it. Not some guy touting a stat on Twitter.
Gotcha. I responded talking about the analytics crowd, not analytics departments. My gripe isn't with teams using them, it's with fans and "experts" using them then telling you that you just don't understand if you disagree using actual, solid video evidence. I'm not big on them, I don't think you can rank players based off a number derived from a ton of different variables. I think in the age of social media and discussion boards, analytics have ruined the way we talk about basketball and you no longer need to understand the game to analyze it. I also have 0 issues with teams implementing them.
It’s just another tool in the toolbox of being a good franchise Can’t only use one tool to build a house.
Should I base my opinion on a guy who never coached the game or HOF coach like Parcells who said you should not chase the game until you have to? He always said let the game develop, when you have a feel for how it’s flowing then you can pull out the bag of tricks. I know who I would listen to. I’m not completely against it, it really depends on the blueprint of the GM for building a team and how he uses the information to select players and how the HC uses some of the data to influence his decision making. The data I never see is how much of a failure analytics is when the players selected because of it are out of the game, or HC that suck on game day because of it, that’s one stat I would love to see.
That’s exactly why I would support Parcells, this paper is based on a Risk Neutral approach in the first quarter, HC cannot think that way, it’s called a game plan, there are way too many variables for a HC to think about for them to base decisions on such a simplistic theory. A team with a bad offense and QB cannot be so cavalier with their opportunities to score points, especially when playing a road game against a better team. As I said before, I’m not against it, I’m just not sold on how great people think it is.
Nobody ever said coaches should base their decisions on that paper. The purpose of the paper was to show that on average, coaches were too conservative in low-leverage situations. It was published in 2006, and since then, other people have expanded it to take into account additional variables and to account for high-leverage situations. Having said that, a coach definitely realize something the numbers don't (maybe the QB smacked his throwing hand on a lineman's helmet on the previous play). But at the same time, there are a bunch of old-school coaches (like Bowles) who are consistently too conservative and would be better off changing their ways.
When will coaches learn? the 1st half should be when you are the most aggressive. the benefits far outweigh the risks.
there are PLENTY of people willing to compile stats/instances on the "failures" of analytics. Maybe you just haven't tried to find it? --- The problem is that there are fewer instances overall. the "nerds" have already won. the game is over. Its common sense if you think about it. Who is going to build a better strategy? A team of Ph.D.'s analyzing data from decades and employing artificial intelligence tools that Fortune 500 companies dream of. OR Industry "lifers".... (coaches, scouts) basing their decisions on gut feelings If the game was drilling oil wells, it will be no question which group you would want making the decisions. Why is it different when its football/baseball, etc??? The amazing advantage sports teams have is that these data scientist geniuses are willing to take less money because they love sports so much. And its shocking how little they take advantage of that. Since we are talking about drilling oil wells, Schlumberger would kill to have the data scientists that the Yankees, Cubs, Astros, etc. (most analytically minded teams) have, but they would go broke. -- Our Jets are showing up to the game and its already over. This dude they hired doesn't stand a chance. Its like the Jets are used to handwriting and they just bought a used printing press. It "bolsters" our operation, sure. But the competitors already have photoshop and laser printers.
What about a tape measure and stopwatch? All that the NFL has bred is bigger players that need more "help" to get bigger to play, doing goodness knows what to their bodies, guys that are nearly 300lb shouldn't be able to sprint as fast as these guys are doing right now. I say this as football over here went through a similar spell and bigger was better and it left no room for the skill players to get a look in, they only wanted athletes who could run and run and run. Yes, I am a small guy and it does piss me off that you have to be a fucking giant to be even classed as capable to do something sporty.
Apples and Oranges, that’s the comparison you just made. Ozzie Newsome can’t build a blueprint for building an oil rig but he drafted some excellent football players. I’m sure the Rooney’s can’t build an oil drilling rig either but they can sure draft football players as well. It’s all dependent on the correct blueprint and philosophy and having the right people to carry it out. I want an engineer with a PhD helping to design the ship I’m taking a vacation on or the car I’m driving my family in, I don’t want Parcells or Mike Macagnan any where near those design teams. How about the PHDs that designed the Boeing 737 Max airplane? You want to give those guys a handshake for a job well done? Data is needed for everything we do in life, it’s fundamental to all decisions made but the over analyzing of data can screw up the simplest decision and have terrible consequences. No one selects players on gut feeling in the draft, scouts use film, visits, and interviews to determine if such player fits their scheme and character profile. The teams that are successful don’t need a PhD to tell them that a good offensive line is the key to a successful offense. It’s not rocket science, like you said it’s common sense. Over analyzing things is not always the answer either. I said it before, I’m not against it, I’m just not buying the hype that’s the greatest thing since sliced bread.
Was that supposed to counter my point? Ozzie Newsome was the biggest believer in analytics than any GM in the entire NFL. He hired Yale & Harvard grads from Wall Street to set up the Ravens extensive scouting database and they were the 1st, and for a time, the only, organization where the head coach was working directly with the analytics department on in-game strategy. -- seems like you have a lot to learn on analytics in general
I don't know squat about analytics other than watching Moneyball once. Can someone give me the "Sports Analytics for Dummies" version? Is it's focus on player acquisition or in game decision making?
Both and more. It's about using data to find undervalued skills in the marketplace, developing players, improving player performance, better in game decision making, etc..... Here's an article about the Astros using analytics. Not about football, but still a good read IMO. https://www.theringer.com/mlb/2019/6/3/18644512/mvp-machine-how-houston-astros-became-great-scouting
Apples and Oranges, that’s the comparison you just made. Ozzie Newsome can’t build a blueprint for building an oil rig but he drafted some excellent football players. I’m sure the Rooney’s can’t build an oil drilling rig either but they can sure draft football players as well. It’s all dependent on the correct blueprint and philosophy and having the right people to carry it out. I want an engineer with a PhD helping to design the ship I’m taking a vacation on or the car I’m driving my family in, I don’t want Parcells or Mike Macagnan any where near those design teams. Data is needed for everything we do in life, it’s fundamental to all decisions made but the wrong data used can screw up the simplest decision and have terrible consequences. Just look at the airplanes that are being grounded the No one selects players on gut feeling in the draft, scouts use film, visits, and interviews to determine if such player fits their scheme and character profile. The teams that are successful don’t need a PhD to tell them that a good offensive line is the key to a successful offense. It’s not rocket science, like you said it’s common sense. We will have to agree to disagree on this topic.
Analytics is about trying to do a good job at something, in this case something football-related. Could be player acquisition, in game decision making, or something else. Say it's in game decisions for example. A coach will use English and his football knowledge (watching players, play-calling, etc) to call make those decisions. An analyst will use the language of math (writing out equations, working with datasets) to do the same thing. Each language has its own strengths and weaknesses. An analogy is communicating in English vs Morse code. Either way, the purpose is to communicate, but sometimes Morse code lets you do things you can't in English (like if you're in a lighthouse and need to communicate with faraway ships). An example of when math lets you do things in football that you sometimes can't do as well in English is going for it on 4th downs. A coach can consider the relevant factors in his head and make an educated guess as to whether it's worth going for it. But he can't crunch all the numbers in his head, so it's possible his intuition could be wrong. Specifically, it's well known in behavioral sciences that many people are risk-averse even to the point of hurting their bottom line. On the other hand, an analytics guy can crunch the numbers, which aren't risk-averse and reveal that on average, coaches are in fact too conservative. Now this doesn't mean an analytics guy can make 4th down decisions as well as a coach. There are too many variables for the analytics guy to put in his model; the math ends up turning into an uncomputable clusterfuck. But he can tell the coach, "The numbers say you should go for it here. Doesn't mean you should since you may know something that isn't in the numbers I have access to. But be aware that your inclination to punt might be a result of being scared to fail. If that's the case, then don't be afraid to go for it."
Thanks for taking the time to write that. It's hard to imagine some old school coach being convinced to make an in-game decision based on what some nerd up in the booth says over the headset. Sean McVay...maybe. Andy Reid?...I'm not so sure about that.
