Bills Fan888 Posted September 30, 2012 Share Posted September 30, 2012 I've been looking at the stats going into week 4 and noticed some interesting things from Ryan Fitzpatrick. Despite his horrible opening game, Fitz is ranked number 12 in passer rating, just two spots behind Tom Brady and 8 spots ahead of Peyton Manning, and tied for FIRST in touchdowns. On the other hand though, he's ranked 28th in passing yards and yards per game with 584 yards and 194 yards per game. It's also interesting to note that CJ Spiller is also tied for number 1 in touchdowns and Stevie Johnson is tied for 2nd in touchdowns. Link to comment Share on other sites More sharing options...
ganesh Posted September 30, 2012 Share Posted September 30, 2012 I've been looking at the stats going into week 4 and noticed some interesting things from Ryan Fitzpatrick. Despite his horrible opening game, Fitz is ranked number 12 in passer rating, just two spots behind Tom Brady and 8 spots ahead of Peyton Manning, and tied for FIRST in touchdowns. On the other hand though, he's ranked 28th in passing yards and yards per game with 584 yards and 194 yards per game. It's also interesting to note that CJ Spiller is also tied for number 1 in touchdowns and Stevie Johnson is tied for 2nd in touchdowns. The best stats for a QB is Yards-Per-Attempt... Passing Yards per Games, YPG is not a good ranking because the team is a Run-First focused team. Link to comment Share on other sites More sharing options...
Hapless Bills Fan Posted September 30, 2012 Share Posted September 30, 2012 The best stats for a QB is Yards-Per-Attempt... Passing Yards per Games, YPG is not a good ranking because the team is a Run-First focused team. So the NYTimes and our own EA do say. Except....Except....standard YPA statistic includes both air yards, and yards after catch. Thus, it is measuring both QB quality (longer passes, passes to a location which enables YAC), and WR quality (ability to switch quickly from pass catcher to runner and accumulate those yards without fumbles/miscues), and it also strongly reflects completion percentage. So I can't agree that YPA is the best QB statistic. It IS a good team statistic. Question: anyone got a free site where air yards/YAC stats are measured? Washington Post gives the leaders, not everyone far's I can tell. Passing YPG actually appears to have a NEGATIVE correlation with winning: I plunked the QB stats off NFL.com into a spreadsheet, added a column for wins and points per game, and looked for correlations. It's early in the season yet, and I expect if I repeat this at the halfway point, and at the end, things will become clearer. If someone knows how to put excel graphs into this bulletin board without converting them to graphic files first, let me know and I'll illustrate this post (PM is fine). 1. pts vs passing yards per game. Total scatterplot, passing yards per game have no discernable relationship to points scored. 2. wins vs passing yards per game Slight negative correlation: the 2 teams with 0 wins average 32 ypg more than the teams with 2 or 3 wins, and 24 ypg more than the teams with 1 win. TBD posters who bemoan Fitz' low ranking in yardage, have it completely backwards!!!!! Having a QB with low passing yards, in general, means you doin' good. 3. wins vs INTs. This is HUGE - strongest correlation. 0 wins: average 7 ints. 3 wins: average 1.3 ints. 1 and 2 wins: 2.3 and 2.9 ints, respectively. If you want to win in the NFL, you can NOT turn it over more than 1x per game. 1x per game, a D seems to be able to thwart/overcome. This seems likely to be causal. Message to Fitz: take the sack, just don't turn it over. 4. wins vs completion percentage - wide spread in the teams with 1 and 2 wins, but there is a correlation. 0 wins: average 54.3% completion (Brees, Weedon). 3 wins: average 66.4% (Ryan, Schaub, Flacco, Kolb - everyone's picture of the league's elite QBs right?). 1 and 2 wins: 61%. There is such a spread at the 1 win and 2 win spot not clear it's significant, and I suspect this is a correlation rather than a causation - in other words, if your team is winning, you don't take chances and make risky throws, so your completion percentage is higher. 5. wins vs points per game. Oh this must correlate, you say. FOOLED YA! Absolutely flat average of 23 pts per game, with wide spread, between 0,1, and 2 wins. 3 win teams did score more points, but less than a TD worth: average of 28 pts per game, and again there's a spread. Message: despite the blurtings of sports pundits last year when GB and NE were winning with sucky Ds, overall this is a team game, and Defense is critical. 6.(drumroll) YPA vs wins: correlation between this and 0 vs 3 wins, but not significant in between and there is nothing magic about the 7 ypa figure that is bruited about as the metric of a good QB. 0 win teams: 6.3 ypa 3 win teams: 7.7 ypa 1 and 2 win teams: 71 and 7.3 YPA. I suspect this is a correlation with completion percentage, not a causation: winning teams only taking the high-probability shots, losing teams trying to force something to happen. 7. YPA vs completion percentage: and indeed, graphed over every team, this is what we see. positive linear correlation, YPA get higher as completion % gets higher. Again, it's only 3 games - I thought it's interesting. I may save this spreadsheet and plug in the numbers after 8 games and at the end of the season, just for grins. Link to comment Share on other sites More sharing options...
The Big Cat Posted September 30, 2012 Share Posted September 30, 2012 So the NYTimes and our own EA do say. Except....Except....standard YPA statistic includes both air yards, and yards after catch. Thus, it is measuring both QB quality (longer passes, passes to a location which enables YAC), and WR quality (ability to switch quickly from pass catcher to runner and accumulate those yards without fumbles/miscues), and it also strongly reflects completion percentage. So I can't agree that YPA is the best QB statistic. It IS a good team statistic. Question: anyone got a free site where air yards/YAC stats are measured? Washington Post gives the leaders, not everyone far's I can tell. Passing YPG actually appears to have a NEGATIVE correlation with winning: I plunked the QB stats off NFL.com into a spreadsheet, added a column for wins and points per game, and looked for correlations. It's early in the season yet, and I expect if I repeat this at the halfway point, and at the end, things will become clearer. If someone knows how to put excel graphs into this bulletin board without converting them to graphic files first, let me know and I'll illustrate this post (PM is fine). 1. pts vs passing yards per game. Total scatterplot, passing yards per game have no discernable relationship to points scored. 2. wins vs passing yards per game Slight negative correlation: the 2 teams with 0 wins average 32 ypg more than the teams with 2 or 3 wins, and 24 ypg more than the teams with 1 win. TBD posters who bemoan Fitz' low ranking in yardage, have it completely backwards!!!!! Having a QB with low passing yards, in general, means you doin' good. 3. wins vs INTs. This is HUGE - strongest correlation. 0 wins: average 7 ints. 3 wins: average 1.3 ints. 1 and 2 wins: 2.3 and 2.9 ints, respectively. If you want to win in the NFL, you can NOT turn it over more than 1x per game. 1x per game, a D seems to be able to thwart/overcome. This seems likely to be causal. Message to Fitz: take the sack, just don't turn it over. 4. wins vs completion percentage - wide spread in the teams with 1 and 2 wins, but there is a correlation. 0 wins: average 54.3% completion (Brees, Weedon). 3 wins: average 66.4% (Ryan, Schaub, Flacco, Kolb - everyone's picture of the league's elite QBs right?). 1 and 2 wins: 61%. There is such a spread at the 1 win and 2 win spot not clear it's significant, and I suspect this is a correlation rather than a causation - in other words, if your team is winning, you don't take chances and make risky throws, so your completion percentage is higher. 5. wins vs points per game. Oh this must correlate, you say. FOOLED YA! Absolutely flat average of 23 pts per game, with wide spread, between 0,1, and 2 wins. 3 win teams did score more points, but less than a TD worth: average of 28 pts per game, and again there's a spread. Message: despite the blurtings of sports pundits last year when GB and NE were winning with sucky Ds, overall this is a team game, and Defense is critical. 6.(drumroll) YPA vs wins: correlation between this and 0 vs 3 wins, but not significant in between and there is nothing magic about the 7 ypa figure that is bruited about as the metric of a good QB. 0 win teams: 6.3 ypa 3 win teams: 7.7 ypa 1 and 2 win teams: 71 and 7.3 YPA. I suspect this is a correlation with completion percentage, not a causation: winning teams only taking the high-probability shots, losing teams trying to force something to happen. 7. YPA vs completion percentage: and indeed, graphed over every team, this is what we see. positive linear correlation, YPA get higher as completion % gets higher. Again, it's only 3 games - I thought it's interesting. I may save this spreadsheet and plug in the numbers after 8 games and at the end of the season, just for grins. QBs put Wrs in a position to catch & run. Link to comment Share on other sites More sharing options...
Rob's House Posted September 30, 2012 Share Posted September 30, 2012 The best SINGLE stats for a QB is Yards-Per-Attempt... Passing Yards per Games, YPG is not a good ranking because the team is a Run-First focused team. Fixed YPA may be better than any other single statistic (that's debatable) but by itself doesn't necessarily tell you much either. It's subject to manipulation just like any other statistic. That's why statistics suck. Picking up 2 yards on 3rd and 1 looks bad on a RBs ypc and tossing a 15 yard pass on 3rd and 20 after you just took a sack for a 10 yard loss looks good on ypa (think Rob Johnson). Point is, no one statistic means dick without context. Link to comment Share on other sites More sharing options...
CardinalScotts Posted September 30, 2012 Share Posted September 30, 2012 2 and 1 moving to 3 and 1 today Link to comment Share on other sites More sharing options...
Over 29 years of fanhood Posted September 30, 2012 Share Posted September 30, 2012 True that if you get for examle; a 90 yard RAC then go incomplete on 9 passes in a row your YPA doesn't reflect poor play but in aggregate over the season that stuff averages out. It's akin to yards per carry for a RB. Link to comment Share on other sites More sharing options...
shibuya Posted September 30, 2012 Share Posted September 30, 2012 (edited) If you look at last season vs the Pat's Fitz threw for over 670 yds, A 65% completion percentage. He had 6 TD's and 6 INT's, however 4 of those INT's were in a meannigless game at the end of the season and the D had pretty much given up a 21+ lead by the 4th quarter when Fitz tossed 2 of his INT's in desperation time.. I agree Fitz needs to eliminate some INT's, but he has had some of his best games last season against New England. Edited September 30, 2012 by shibuya Link to comment Share on other sites More sharing options...
Recommended Posts