| Buckeye Football Ohio State football, moderated. Consider this the general community forum, but with a theme. Completely off-topic chatter should start at, or move to, the Open Discussion forum. |

06-20-2006, 03:16 PM
|
 |
Grey Beard = True Buckeye
Moderator
|
|
|
|
Join Date: Apr 2004
Posts: 4,549
Points: 4,203,027.54
Bank: 0.00
Total Points: 4,203,027.54
|
|
|
Ideal number of pass attempts per game
Last night I was sorting data on my passing efficiency spread sheets, looking for more ways to reflect the blazing glory that is Troy Smith; and something interesting fell out of the numbers.
For every year going back to 2000 I did the following: - Sorted every top 100 QB (efficiency) w.r.t. Attempts per Game
- Performed a 10-point moving average on the data
- Graphed the data and located the peaks
If you don't really get the analysis techniques, don't worry about it. Not everyone was born to be a stat geek with zero social skills.
The upshot of the analysis is that there is a narrow range of pass attempts per game that produces the highest passing efficiency with incredible regularity. Here are the data: <TABLE style="WIDTH: 216pt; BORDER-COLLAPSE: collapse" cellSpacing=0 cellPadding=0 width=287 border=0 x:str><COLGROUP><COL style="WIDTH: 38pt; mso-width-source: userset; mso-width-alt: 1828" width=50><COL style="WIDTH: 86pt; mso-width-source: userset; mso-width-alt: 4205" width=115><COL style="WIDTH: 92pt; mso-width-source: userset; mso-width-alt: 4461" width=122><TBODY><TR style="HEIGHT: 13.5pt" height=18><TD class=xl36 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext 1pt solid; BORDER-LEFT: windowtext 1pt solid; WIDTH: 38pt; BORDER-BOTTOM: windowtext 1pt solid; HEIGHT: 13.5pt; BACKGROUND-COLOR: transparent" width=50 height=18>Yr</TD><TD class=xl30 style="BORDER-RIGHT: black 1pt solid; BORDER-TOP: windowtext 1pt solid; BORDER-LEFT: #ece9d8; WIDTH: 178pt; BORDER-BOTTOM: windowtext 1pt solid; BACKGROUND-COLOR: transparent" width=237 colSpan=2>Number of Attempts - Range</TD></TR><TR style="HEIGHT: 12.75pt" height=17><TD class=xl37 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: #ece9d8; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: windowtext 0.5pt solid; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" height=17 x:num>2000</TD><TD class=xl31 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: #ece9d8; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num>29.70</TD><TD class=xl25 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: #ece9d8; BORDER-LEFT: windowtext; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="32.272727272727273">32.27</TD></TR><TR style="HEIGHT: 12.75pt" height=17><TD class=xl38 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: windowtext 0.5pt solid; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" height=17 x:num>2001</TD><TD class=xl32 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: windowtext; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="28.818181818181817">28.82</TD><TD class=xl24 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="31.384615384615383">31.38</TD></TR><TR style="HEIGHT: 12.75pt" height=17><TD class=xl38 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: windowtext 0.5pt solid; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" height=17 x:num>2002</TD><TD class=xl32 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: windowtext; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="28.727272727272727">28.73</TD><TD class=xl24 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="30.666666666666668">30.67</TD></TR><TR style="HEIGHT: 12.75pt" height=17><TD class=xl38 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: windowtext 0.5pt solid; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" height=17 x:num>2003</TD><TD class=xl32 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: windowtext; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num>29.38</TD><TD class=xl24 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num>33.17</TD></TR><TR style="HEIGHT: 12.75pt" height=17><TD class=xl38 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: windowtext 0.5pt solid; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" height=17 x:num>2004</TD><TD class=xl32 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: windowtext; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="28.666666666666668">28.67</TD><TD class=xl24 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="32.272727272727273">32.27</TD></TR><TR style="HEIGHT: 13.5pt" height=18><TD class=xl39 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: #ece9d8; HEIGHT: 13.5pt; BACKGROUND-COLOR: transparent" height=18 x:num>2005</TD><TD class=xl33 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: windowtext; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: #ece9d8; BACKGROUND-COLOR: transparent" x:num>28.00</TD><TD class=xl28 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext; BORDER-BOTTOM: #ece9d8; BACKGROUND-COLOR: transparent" x:num>31.25</TD></TR><TR style="HEIGHT: 12.75pt" height=17><TD class=xl40 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext 1pt solid; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: windowtext 0.5pt solid; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" height=17>AVG</TD><TD class=xl34 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: windowtext 1pt solid; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="28.8820202020202" x:fmla="=AVERAGE(B2:B7)">28.88</TD><TD class=xl29 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext 1pt solid; BORDER-LEFT: windowtext; BORDER-BOTTOM: windowtext 0.5pt solid; BACKGROUND-COLOR: transparent" x:num="31.836122766122767" x:fmla="=AVERAGE(C2:C7)">31.84</TD></TR><TR style="HEIGHT: 13.5pt" height=18><TD class=xl41 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext 1pt solid; BORDER-BOTTOM: windowtext 1pt solid; HEIGHT: 13.5pt; BACKGROUND-COLOR: transparent" height=18>STDEV</TD><TD class=xl35 style="BORDER-RIGHT: windowtext 0.5pt solid; BORDER-TOP: windowtext; BORDER-LEFT: #ece9d8; BORDER-BOTTOM: windowtext 1pt solid; BACKGROUND-COLOR: transparent" x:num="0.59488380903371718" x:fmla="=STDEV(B2:B7)">0.59</TD><TD class=xl27 style="BORDER-RIGHT: windowtext 1pt solid; BORDER-TOP: windowtext; BORDER-LEFT: windowtext; BORDER-BOTTOM: windowtext 1pt solid; BACKGROUND-COLOR: transparent" x:num="0.90282909196699279" x:fmla="=STDEV(C2:C7)">0.90</TD></TR></TBODY></TABLE>
What This Means:- The ideal range of attempts per game, on average, was 28.88 to 31.84.
- By ideal, this means that it provided for a higher average passing efficiency for quarterbacks whose attempts/game fell within this range.
- In 2005, QBs within the ideal range had an average efficiency 8.9 points higher than those outside the range.
- The Standard Deviation being less than 1 for both the ends of the range is astounding. This means that there was surprisingly very little variability in the ideal range from year to year.
What This Does Not MeanNone of this should be taken to imply that I believe that Ohio State should pass this much. This range of passing attempts provides for a slight bump in passing efficiency. It does not seem to have any effect on winning and losing. Every season, the teams that fell in this range were a wide mix of good and bad, winners and losers.
The data here say nothing about what amount of passing would be ideal for Troy Smith's passing efficiency. If we have a dynamite running game, it is likely that 21.55 attempts per game (last year's number) would be just fine.
So why post this then? Primarily, because I'd like to see if any of you geniuses have an explanation for this. As far as I'm concerned, it's counter-intuitive that there is a narrow range of pass attempts that lends itself to the highest efficiency every year for 6 seasons. Further, it is astonishing that the boundaries of the range varied so little from year to year.
__________________
That's right Bo. Five in a row.
|

06-20-2006, 03:24 PM
|
 |
I'd rather be napping!!
|
|
|
|
Join Date: Jun 2003
Posts: 6,684
Points: 6,008,867.64
Bank: 0.00
Total Points: 6,008,867.64
|
|
|
higher # of passes leads to a passing system, short, ball control passes used in place of an effective running game.
|

06-20-2006, 03:32 PM
|
 |
Loves Buckeye History
Senior Moderator
|
|
|
|
Join Date: Apr 2004
Posts: 26,318
Points: 1,501,646,434.65
Bank: 0.00
Total Points: 1,501,646,434.65
|
|
|
I was thinking that about 25 attempts would have the best ratings. A result of an effecient passing attack mixed in with a powerful running game.
But one factor that may explain why the numbers you've researched are closer to 30 is that the more a team passes, the more proficient they become.
__________________
"In your veins runs Scarlet and Gray blood with which there is no quit, no stopping. It makes you relentless and powerful, more so than your opponent."
SFC Ryan Poetsch
101st ABN DIV, Afghanistan
|

06-20-2006, 03:32 PM
|
 |
Bench Warmer
|
|
|
|
Join Date: Jul 2004
Posts: 935
Points: 5,222.06
Bank: 0.00
Total Points: 5,222.06
|
|
|
I would guess the reason why a low number of pass attempts correlates to low passing efficiency is that teams which pass infrequently are so run-oriented that they have not developed particularly good passing proficiency - and likely have QB's who are not good passers. As to the teams which pass the football very frequently, they likely inspire opposing defenses to play a lot of nickle and dime packages. To distill, ~29 pass attempts per game would be the number where your offense is sufficiently pass-oriented to enable you to be proficient at it (and to recruit good passing QB's), but not so pass-oriented that opposing defenses will invariably play a predomantly pass-oriented defensive scheme. As to why the optimum range is so narrow, I can't think of any "principled" explanation; just that each of the factors above kicks in very steeply.
edit: One thing I would wonder is whether having a quarterback who is a good runner might expand this range. I'm not talking about an option quarterback, or a quarterback who does a lot of designed runs, but a quarterback who has the ability and inclination to run from a called passing play when nobody's open. I'm sure you could think of an example of such a quarterback.
Last edited by zincfinger; 06-20-2006 at 05:28 PM.
|

06-20-2006, 03:50 PM
|
 |
I think. Therefore, I am
Administrator
|
|
|
|
Join Date: Jul 2004
Posts: 28,104
Points: 22,562,769.27
Bank: 0.53
Total Points: 22,562,769.80
|
|
|
What is the average number of total plays run by an offense? I'm guessing somewhere in the neighborhood of 70 - 80 per game. This means, that for every 1 pass (of a team in the range) there are also nearly 2 runs. What this tells me is that the chances are good that the teams averaging about 30 passes a game, are throwing high percentage balls (read: short yardage)
Just a guess. Still, I'm reminded of the San Jose State game in 02, I think it was. Their QB was something like 23 - 26, but he only had slightly more than 100 yards. Well... I guess there is more to efficiency than simply compl. %, but with the short passes he was throwing, picks weren't much of an issue.. conversely, neither were touchdowns.
Anyway... I think it's a good balance to run the ball 2 times to every throw, and it seems to me your data supports that idea in terms of efficiency. That is, if teams have to respect the run, the passing game is a bit easier to accomplish successfully.
__________________
"Reality is merely a framework for imagination" 9/25/05
"I am firmly convinced that Charlie Weis didn't have any idea as to how good the Buckeyes were defensively." - Brent Musburger
I'd rather be a prison whore than root for Michigan --- Long live the PCP
|

06-20-2006, 04:49 PM
|
 |
"Down in a hole........."
|
|
|
|
Join Date: Sep 2004
Posts: 3,688
Points: 307,650.21
Bank: 0.00
Total Points: 307,650.21
| | |