boratt Posted July 29, 2019 Share Posted July 29, 2019 Oh yeah and if you combine fumbles and interceptions, turn over is more important than passing so I would have guessed right. Don’t need to over complicate common sense. 1 Quote Link to comment Share on other sites More sharing options...
TransientTexan Posted July 30, 2019 Share Posted July 30, 2019 5 hours ago, skibrett15 said: On a related separate note, why not run it 20% of the time? I would guess that the problem is, all those stats took place in an environment where offenses ran a lot more than 20% (and were more unpredictable). It’s baked into the results. If you change that environment, then defenses will change accordingly, which will change the success rates. In short, the passing successes don’t happen in a vacuum. 1 Quote Link to comment Share on other sites More sharing options...
HyponGrey Posted July 30, 2019 Share Posted July 30, 2019 9 hours ago, skibrett15 said: actually having a passable run game doesn't really matter. Having a poor run game hurts a little, and having a very efficient run game helps a little. That's kind of the point of the analysis. If there's one thing we learned from baseball, it's that the human arm can only pitch so often before breaking down. Bad run games get abandoned early. Quote Link to comment Share on other sites More sharing options...
ny92mike Posted July 30, 2019 Share Posted July 30, 2019 (edited) On 7/29/2019 at 1:50 PM, squire12 said: @ny92mike This might be something you would be interested in for you sim league. Definitely up my ally. Most of my work with projecting wins deals with correlation values (stat line vs total win). Which I get a pretty solid number in terms of straight win / loss if I've got the full regular season stats, but it drops when running a weekly data collection of stats. Following. Edited July 30, 2019 by ny92mike Quote Link to comment Share on other sites More sharing options...
Malfatron Posted July 30, 2019 Share Posted July 30, 2019 On 7/29/2019 at 12:51 PM, HyponGrey said: It's a passing league. It helps a lot to have a passable run game, though. like halfback passes? 2 Quote Link to comment Share on other sites More sharing options...
HyponGrey Posted July 31, 2019 Share Posted July 31, 2019 (edited) 11 hours ago, Malfatron said: like halfback passes? More like not completely useless. Only for QB pitch count purposes though. Edited July 31, 2019 by HyponGrey Quote Link to comment Share on other sites More sharing options...
Malfatron Posted July 31, 2019 Share Posted July 31, 2019 7 hours ago, HyponGrey said: More like not completely useless. Only for QB pitch count purposes though. Gotcha. I think other runs besides QB pitches should count though. HB Dive and Offtackles are useful too. Quote Link to comment Share on other sites More sharing options...
ny92mike Posted July 31, 2019 Share Posted July 31, 2019 Perhaps I'm overlooking it or not fully understanding what I'm looking at, but where are the actual touchdown numbers for this, I see yards, interceptions and the likes but not TD? The correlation to wins and rushing td's is 68.7% and even higher at 71% when correlated to Net Pts, with passing being more important than rushing which I do agree with, I thought it was cool that the rushing TD's carried a higher correlation value than passing TD's 52.8% and 60% for net point correl (which is identical for receiving stats). For just straight line stats correlation values for passing and rushing. Passing Stat Line Correlation % Rate 67.4% Avg 61.2% 1st% 56.6% Pct 54.2% TD 52.8% Yds 33.9% Yds/G 33.9% 1st 31.2% 20+ 27.7% Comp 19.7% 40+ 19.3% Lng 9.4% Att -6.0% Att/G -6.0% Int -44.2% Sck -46.1% Rushing Stat line Correlation % TD 68.7% 1st 61.6% Yds 53.7% Yds/G 53.7% Att 50.4% Att/G 50.4% 1st% 47.1% 20+ 25.7% Avg 21.8% FUM -9.8% Lng -21.0% 40+ -36.1% Quote Link to comment Share on other sites More sharing options...
ny92mike Posted July 31, 2019 Share Posted July 31, 2019 On 7/29/2019 at 12:46 PM, skibrett15 said: no, not predictive. A better thread title (not mine btw, just pulled from twitter) might be "what happens to NFL teams when they win" rather than what makes them win. Is this your work? Or something you stumbled across? Quote Link to comment Share on other sites More sharing options...
HyponGrey Posted July 31, 2019 Share Posted July 31, 2019 3 hours ago, Malfatron said: Gotcha. I think other runs besides QB pitches should count though. HB Dive and Offtackles are useful too. Well played. Quote Link to comment Share on other sites More sharing options...
skibrett15 Posted July 31, 2019 Author Share Posted July 31, 2019 (edited) 1 hour ago, ny92mike said: Is this your work? Or something you stumbled across? not mine It's an update to something brian burke did before he was consumed by ESPN and stopped publishing relevant things. It looks at the modern era data set. I don't think that current team building strategies are quite there from a top-down perspective. I think there's some really good innovation at the scheme and using personnel creatively at the middle levels like coaching. It will take someone who really believes in this and is willing to test the limits of the passing game at the ownership and GM level. In the next 10 years, I expect there to be a complete abandonment of the running game outside of short yardage and clock killing scenarios to a degree similar to that of the mid-range and long 2 pointer in the NBA. Once the box is light enough, running plays should start to average 6 or so y/a, and things will be at something of an equilibrium. We aren't even close to that yet. Edited July 31, 2019 by skibrett15 Quote Link to comment Share on other sites More sharing options...
ny92mike Posted July 31, 2019 Share Posted July 31, 2019 56 minutes ago, skibrett15 said: not mine It's an update to something brian burke did before he was consumed by ESPN and stopped publishing relevant things. It looks at the modern era data set. I don't think that current team building strategies are quite there from a top-down perspective. I think there's some really good innovation at the scheme and using personnel creatively at the middle levels like coaching. It will take someone who really believes in this and is willing to test the limits of the passing game at the ownership and GM level. In the next 10 years, I expect there to be a complete abandonment of the running game outside of short yardage and clock killing scenarios to a degree similar to that of the mid-range and long 2 pointer in the NBA. Once the box is light enough, running plays should start to average 6 or so y/a, and things will be at something of an equilibrium. We aren't even close to that yet. What I'm doing with data similar to this is projecting scores based on correlation data collected so the conversation isnt the same but how the data is looked at is similar. Most of my work is after the fact but the same correlation data works for previous years too. So I do enjoy reading about this stuff. So appreciate you sharing. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.