Kirby Jackson Posted February 23, 2015 Share Posted February 23, 2015 ESPN is doing an interesting comparison on how different organizations are adopting analytics. It is not surprising at all to see the NFL as the last to sort of jump on board. They typically follow the NBA 5 years later when it comes to business operations. Nonetheless, this is an interesting read: http://espn.go.com/espn/feature/story/_/id/12331388/the-great-analytics-rankings Link to comment Share on other sites More sharing options...
BringBackFergy Posted February 23, 2015 Share Posted February 23, 2015 Our Analytics Dept. was run like a well oiled machine last year...Marrone will be missed. Link to comment Share on other sites More sharing options...
johnwalter Posted February 23, 2015 Share Posted February 23, 2015 keep in mind that it may be that the nfl is slower to adopt "analytics" because statistics are much less useful in football. in baseball, a player's statistics are almost a perfect representation of his performance. in the basketball, a player's statistics are slightly less useful, but still fairly representative of his performance. in football... statistics are incredibly difficult to interpret. does a WR have strong stats because of his strong performances, or because of strong performances by his QB (just look at how Decker / Sander's receiving stats changed when they left / joined Manning in Denver)? does an RB have good stats because of a his strong performances or because of the strong performances of his offensive line? it may not be that the nfl is behind, just that analytics are less applicable / interpretable in football. football is a true team sport. Link to comment Share on other sites More sharing options...
zonabb Posted February 23, 2015 Share Posted February 23, 2015 It started with baseball... why? Because in baseball, every single event has a given situation with it and far less randomness. A strike is a strike. One pitcher to one batter. One ball hit to one spot. Football is way more random and hockey even more so. I laugh at the WGR66 types who talk analytics all day long, in particular elementary measuers based on some simple probabilities. They stand behind them vociferously and when the improbable happens, they chalk it up to luck. Exactly. Luck is a huge component and it limits the use of prediction in football and hockey more so than in baseball. Link to comment Share on other sites More sharing options...
Kirby Jackson Posted February 23, 2015 Author Share Posted February 23, 2015 keep in mind that it may be that the nfl is slower to adopt "analytics" because statistics are much less useful in football. in baseball, a player's statistics are almost a perfect representation of his performance. in the basketball, a player's statistics are slightly less useful, but still fairly representative of his performance. in football... statistics are incredibly difficult to interpret. does a WR have strong stats because of his strong performances, or because of strong performances by his QB (just look at how Decker / Sander's receiving stats changed when they left / joined Manning in Denver)? does an RB have good stats because of a his strong performances or because of the strong performances of his offensive line? it may not be that the nfl is behind, just that analytics are less applicable / interpretable in football. football is a true team sport. That is all true and it wasn't meant as a shot at the NFL. In general though, the business operations practices of the NFL are about 2 years behind MLB & NHL and 5 years behind the NBA. Since Pete Rozelle they haven't really been pioneers. This is in part because they don't have the international influence of the other leagues. It is also in part because they are by far the lowest paying league in business ops (50% or less in many cases). The NBA in particular is attracting top business execs and Ivy League grads to run their business ops. That is why they are the ones doing the innovating. Link to comment Share on other sites More sharing options...
plenzmd1 Posted February 23, 2015 Share Posted February 23, 2015 . Football is way more random and hockey even more so. I laugh at the WGR66 types who talk analytics all day long, in particular elementary measuers based on some simple probabilities. They stand behind them vociferously and when the improbable happens, they chalk it up to luck. Exactly. Luck is a huge component and it limits the use of prediction in football and hockey more so than in baseball. Football may be its own bird, but analytics in hockey are not random, and do tell a story over an 82 game period. Possession stats, shots attempted stats, shooting percentages, save percentages are all HUGELY illuminating in hockey. Earlier in the year, Sabres won 10 of 13, but their analytics were so far from the norm there was no doubt a regression was coming. Same for a player with a hot year scoring goals, when his shooting percentage was off the charts.Or if 80% of his starts were in the offensive zone...just a ton of useful stats. Link to comment Share on other sites More sharing options...
jletha Posted February 23, 2015 Share Posted February 23, 2015 (edited) It started with baseball... why? Because in baseball, every single event has a given situation with it and far less randomness. A strike is a strike. One pitcher to one batter. One ball hit to one spot. Football is way more random and hockey even more so. I laugh at the WGR66 types who talk analytics all day long, in particular elementary measuers based on some simple probabilities. They stand behind them vociferously and when the improbable happens, they chalk it up to luck. Exactly. Luck is a huge component and it limits the use of prediction in football and hockey more so than in baseball. the biggest advantage that baseball has isnt the non-randomness; baseball has plenty of randomness. Its the fact that the sample size is so large that they can generate stats with small confidence intervals, thus eliminating the possibility of them being ambiguous. With 30 teams, 162 games per team and 9 innings per game and X many pitches per game every player faces so many different scenarios that it becomes very easy to eliminate outside variable and focus on one piece of information. For example a new stat in baseball is called strikes above average, which measures one single players influence on a pitch he throws or has thrown to him being called a strike. In order to quantify this for a batter you would need the same player to face multiple pitchers multiple times with different umps and different catchers each time. Football just doesnt have the volume of available data in a lot of areas so its harder to get large trends and really useful data right now. Edited February 23, 2015 by jletha Link to comment Share on other sites More sharing options...
Luxy312 Posted February 23, 2015 Share Posted February 23, 2015 I think a lot of you guys got it wrong on a number of fronts. There's nothing random about analytics. There are more variables and interaction between the variables to explain than there are in baseball and other sports. To make a simple analogy, you can make macaroni & cheese with noodles, a little milk and the cheese. Pretty simple to figure out what you're going to get based on the ratio of those three things. Making the perfect pasta sauce from scratch is a lot more precise of an exercise. Considering the pasta sauce the allegory for the NFL, there isn't a "recipe" out there that tells you how to get to perfect. Then comes the issue in the NFL with the data itself. It changes year to year based on changes in rules year to year. So continuing the analogy, a tomato isn't "just" a tomato. One year it's a roma and the next a beefsteak. The raw data in the NFL is looked at as being the same, when it really is not. At the end of the day, these analytics guys are mining for the factors that matter and the one's that don't matter in predicting success at the individual player level. It's incredibly complicated and involves lots of moving parts. I love the stats to be honest, so I find this incredibly interesting. Link to comment Share on other sites More sharing options...
jletha Posted February 23, 2015 Share Posted February 23, 2015 I think a lot of you guys got it wrong on a number of fronts. There's nothing random about analytics. There are more variables and interaction between the variables to explain than there are in baseball and other sports. To make a simple analogy, you can make macaroni & cheese with noodles, a little milk and the cheese. Pretty simple to figure out what you're going to get based on the ratio of those three things. Making the perfect pasta sauce from scratch is a lot more precise of an exercise. Considering the pasta sauce the allegory for the NFL, there isn't a "recipe" out there that tells you how to get to perfect. Then comes the issue in the NFL with the data itself. It changes year to year based on changes in rules year to year. So continuing the analogy, a tomato isn't "just" a tomato. One year it's a roma and the next a beefsteak. The raw data in the NFL is looked at as being the same, when it really is not. At the end of the day, these analytics guys are mining for the factors that matter and the one's that don't matter in predicting success at the individual player level. It's incredibly complicated and involves lots of moving parts. I love the stats to be honest, so I find this incredibly interesting. I think youre kind of underselling how complicated baseball actually is. There are many interacting variables in baseball and a lot of things influence every pitch that is thrown and the outcome. The MLB also changes its rules year-to-year just like most leagues do. They make changes where they seem necessary and the way its played has drastically changed over the years like the NFL, look at stealing numbers for example. Link to comment Share on other sites More sharing options...
Lurker Posted February 23, 2015 Share Posted February 23, 2015 Nonetheless, this is an interesting read: http://espn.go.com/espn/feature/story/_/id/12331388/the-great-analytics-rankings I guess they didn't consult with Battling Bill on that story... http://www.nfl.com/news/story/0ap1000000121582/article/bill-polian-moneyball-does-not-work-in-the-nfl Link to comment Share on other sites More sharing options...
Luxy312 Posted February 23, 2015 Share Posted February 23, 2015 I think youre kind of underselling how complicated baseball actually is. There are many interacting variables in baseball and a lot of things influence every pitch that is thrown and the outcome. The MLB also changes its rules year-to-year just like most leagues do. They make changes where they seem necessary and the way its played has drastically changed over the years like the NFL, look at stealing numbers for example. I don't disagree and it was just an analogy. In baseball though the data is far more consistently gathered and maintained and there's techniques to adjust for small changes to strike zones, size of ball parks, etc. They've been doing it a lot longer to boot. Give the NFL another 4-5 years and they'll be much better at it than they are now. Link to comment Share on other sites More sharing options...
BuffalothruMyVeins Posted February 23, 2015 Share Posted February 23, 2015 Our Analytics Dept. was run like a well oiled machine last year...Marrone will be missed. Robust. Lockstep. Obviously. Analyze that. Link to comment Share on other sites More sharing options...
Kirby Jackson Posted February 23, 2015 Author Share Posted February 23, 2015 I guess they didn't consult with Battling Bill on that story... http://www.nfl.com/news/story/0ap1000000121582/article/bill-polian-moneyball-does-not-work-in-the-nfl I always find it funny when the old school scout type argues with numbers. The reality is that analytics is a tool. It does serve a purpose as does traditional scouting. It reminds me of the Clint Eastwood character in Trouble with the Curve (don't judge me for having seen it). Gather as much information as you can and make the best decision for your organization. Link to comment Share on other sites More sharing options...
Over 29 years of fanhood Posted February 24, 2015 Share Posted February 24, 2015 keep in mind that it may be that the nfl is slower to adopt "analytics" because statistics are much less useful in football. in baseball, a player's statistics are almost a perfect representation of his performance. in the basketball, a player's statistics are slightly less useful, but still fairly representative of his performance. in football... statistics are incredibly difficult to interpret. does a WR have strong stats because of his strong performances, or because of strong performances by his QB (just look at how Decker / Sander's receiving stats changed when they left / joined Manning in Denver)? does an RB have good stats because of a his strong performances or because of the strong performances of his offensive line? it may not be that the nfl is behind, just that analytics are less applicable / interpretable in football. football is a true team sport. It's not that it can't and wont be done, only that it is just harder because of a greater number of interdependencies. Link to comment Share on other sites More sharing options...
BuffOrange Posted February 24, 2015 Share Posted February 24, 2015 At the end of the day, these analytics guys are mining for the factors that matter and the one's that don't matter in predicting success at the individual player level. It's incredibly complicated and involves lots of moving parts. I love the stats to be honest, so I find this incredibly interesting. Ding ding. This is what the old guard can never comprehend for some reason. Any old baseball writer will tell you sabermetrics are idiotic for downplaying RBIs. They always misrepresent the point - things can be a big factor in wins & losses and still not be a skill. The best way to hit with guys on base is to be a good hitter in general and sometimes a hitter lucky/better with guys on base, some years not so much. A little bit like redzone and 3rd downs in football. In basketball it's 3pt defense %. It's pretty funny that my previous unpopular Rueben Amaro bashing is now en vogue here in Philly. Link to comment Share on other sites More sharing options...
Recommended Posts