Guest Post: Marginal Drops

by Chase Stuart on December 10, 2014

in Quarterbacks, Statistics

Munir Mohamed, a reader of Football Perspective, has agreed to write this guest post for us. And I thank him for it.

Regular readers are familiar with Adam Steele’s three–part series here on Marginal YAC; today, I want to look at drops, and marginal drops. As Adam noted, Sportingcharts.com keeps track of dropped passes. ^[1]Some fine print: Unfortunately, that data is only recorded on a team level, not at the individual passer level. As a result, I gave each quarterback his pro rata portion of his team’s dropped … Continue reading

Her’s how to read the table below, which is sorted by career Marginal Drops. Over the course of this data set, Eli Manning completed 2,929 of his 5,008 passes, for a completion percentage of 58.5%. Manning’s Giants dropped an estimated 299.4 of his passes; if we add those to his 2,929 completions, Manning was therefore “On Target” with 64.5% of his throws. Relative to league average, Manning had 44 more drops than we would expect. Manning’s drop percentage — i.e., his number of drops divided by his total number of completions and drops, was 9.3%, which represents his percentage of catchable balls that were dropped. Manning lost 516.5 yards from his marginal drops, or 52.9 yards last from marginal drops per 300 completions.

entries per page

Search:

Player	Cmp	Att	Comp%	Drops	On Target%	Mar Drops	Drop%	YLfMD	YL/300Cmp
Eli Manning	2929	5008	58.5%	299.4	64.5%	44.0	9.3%	516.5	52.9
Joey Harrington	1424	2538	56.1%	169.0	62.8%	34.6	10.6%	351.6	74.1
Gus Frerotte	1699	3106	54.7%	203.9	61.3%	30.7	10.7%	385.8	68.1
Derek Anderson	760	1440	52.8%	100.5	59.8%	28.7	11.7%	336.2	132.7
Dan Marino	2169	3628	59.8%	273.8	67.3%	27.8	11.2%	328.1	45.4
Matthew Stafford	1485	2497	59.5%	156.0	65.7%	27.1	9.5%	321.9	65.0
Kerry Collins	3487	6261	55.7%	366.1	61.5%	26.7	9.5%	329.6	28.4
Trent Dilfer	1759	3172	55.5%	208.6	62.0%	26.6	10.6%	313.8	53.5
Tony Banks	1278	2356	54.2%	159.7	61.0%	23.9	11.1%	300.2	70.5
Boomer Esiason	1216	2105	57.8%	167.3	65.7%	23.8	12.1%	267.1	65.9
Trent Green	2266	3740	60.6%	240.1	67.0%	23.5	9.6%	296.3	39.2
Donovan McNabb	3170	5374	59.0%	313.3	64.8%	19.6	9.0%	230.8	21.8
Rich Gannon	2131	3482	61.2%	245.0	68.2%	19.4	10.3%	243.0	34.2
Craig Erickson	591	1092	54.1%	90.2	62.4%	19.0	13.2%	251.7	127.8
Matt Cassel	1356	2298	59.0%	135.3	64.9%	16.2	9.1%	170.6	37.7
Byron Leftwich	930	1605	57.9%	100.5	64.2%	15.2	9.8%	169.3	54.6
Michael Vick	1703	3030	56.2%	166.8	61.7%	14.9	8.9%	193.1	34.0
Aaron Brooks	1673	2963	56.5%	172.9	62.3%	13.3	9.4%	163.7	29.3
Matt Hasselbeck	3036	5030	60.4%	285.3	66.0%	12.5	8.6%	163.8	16.2
Jake Plummer	2484	4350	57.1%	253.0	62.9%	12.2	9.2%	138.1	16.7
Jeff Blake	1827	3241	56.4%	202.8	62.6%	11.8	10.0%	151.3	24.8
Mark Rypien	657	1204	54.6%	95.6	62.5%	11.3	12.7%	122.7	56.0
Kent Graham	694	1339	51.8%	86.6	58.3%	9.9	11.1%	116.9	50.5
Chris Miller	635	1122	56.6%	85.4	64.2%	9.4	11.9%	109.9	51.9
Tommy Maddox	686	1200	57.2%	79.8	63.8%	9.3	10.4%	102.5	44.8

Showing 1 to 25 of 102 entries

The top 10 in career Marginal Drops is an interesting mix. The list features a Hall of Famer (Dan Marino), bad quarterbacks (Joey Harrington and Derek Anderson), and average quarterbacks (Matthew Stafford and Kerry Collins). It’s hard to find a pattern in that list. But the bottom 10 in Marginal Drops is disproportionately filled with quarterbacks who were pretty good. ^[2]Query whether the chicken or the egg came first here: are these quarterbacks considered good because their receivers didn’t drop passes? The worst QB in the bottom 10 is either Brad Johnson or Matt Schaub. Seven of the bottom 10 were MVP caliber players during their prime (Steve Young, Brett Favre, Daunte Culpepper, Kurt Warner, Chad Pennington, Drew Brees, and Philip Rivers), and quarterbacks known for their accuracy seem to populate the bottom of the list.

If you are expecting a resolution to the Peyton Manning/Tom Brady debate, you won’t get it here. Brady and Manning are, quite amazingly, side by side by in career drops above average. They are separated by less than 1 Marginal Drop during their whole careers. Surprisingly, 5 of Brady’s 6 best seasons in drop “luck” happened from 2001-2005. The Patriots star quarterback has been great since 2007 despite his receivers getting worse relative to league average when it comes to drops. From 2001-2006, Brady had 32 less drops than expected. From 2007-2013, he had 14 more drops than expected. While Brady had better receiver help since 2007, this would seem to indicate that his receivers had less reliable hands than the ones he had when he was winning Super Bowls.

What about the best and worst single seasons?

entries per page

Search:

Year	QB	Tm	Age	G	Cmp	Att	Cmp%	Yds	TD	Int	PR	Drops	MD	D%	YLfMD	YL / 300Cmp
2009	Matt Cassel	KAN	27	15	271	493	55.0%	2924	16	16	69.9	45.3	16.9	14.3%	181.9	18.7
2013	Matthew Stafford	DET	25	16	371	634	58.5%	4650	29	19	84.2	44.0	13.9	10.6%	174.3	11.2
1993	Craig Erickson	TAM	24	16	233	457	51.0%	3054	18	21	66.4	42.8	12.6	15.5%	165.2	16.2
2007	Eli Manning	NYG	26	16	297	529	56.1%	3336	23	20	73.9	40.3	12.3	11.9%	138.3	12.4
1993	Boomer Esiason	NYJ	32	16	288	473	60.9%	3421	16	11	84.5	47.4	11.6	14.1%	138.0	12.1
2003	Joey Harrington	DET	25	16	309	554	55.8%	2880	17	22	63.9	46.4	14.1	13.1%	131.5	13.7
2001	Tony Banks	WAS	28	15	198	370	53.5%	2386	10	10	71.3	34.1	10.9	14.7%	131.0	16.5
2005	Eli Manning	NYG	24	16	294	557	52.8%	3762	24	17	75.9	36.9	10.0	11.1%	128.4	10.2
1992	Dave Krieg	KAN	34	16	230	413	55.7%	3115	15	12	79.9	46.0	9.4	16.7%	127.2	12.3
2013	Tom Brady	NWE	36	16	380	628	60.5%	4343	25	11	87.3	41.0	11.0	9.7%	125.3	8.7

Showing 1 to 10 of 656 entries

Matt Cassel’s 2009 season tops the list for most yards lost due to Marginal Drops. Cassel’s 182 yards lost from Marginal Drops equates to about 0.4 in Yards/Attempt. That might seem like a lot but that would only move him up from 28th in Y/A to 25th. Drew Brees dominates the bottom of the list. His 2013 season was the best in fewest yards lost due to Marginal Drops. Brees owns 3 of the top 11 seasons by this measure. The 2nd and 3rd fewest belong to Peyton Manning and Daunte Culpepper in 2004. That drop luck helped fuel their historically great seasons. Dan Marino is the best QB who seemed to have a lot of bad luck with drops. The most drops in a season from 1992-2013 was Marino’s 1992 season with 52.4 dropped passes. Marino has 3 of the top 12 seasons with most dropped passes.

Drop Percentage Over Time

The league average drop rate in 2013 was almost half the rate it was in 1992. The drop could either be attributed to better receiver play or a change in what counts as a drop. One theory I had was that the lower Yards per Completion average today in comparison to the past might play a role, since it should be easier to catch a short pass than a longer pass. But, for what it’s worth, there is no correlation on the individual level between drops and Yards per Completion. Do you have any ideas?

Finally, here’s some more food for thought: passers do seem to have some control for drops by their receivers. Among quarterbacks with 250+ attempts in back to back seasons, there is an R value of .237 for Marginal Drops per 300 Completions. Some quarterbacks might simply throw more “catchable” balls. ^[3]Chase note: This was a common refrain by fans, the local media, and the Jets organlization during the Pennington era. He was known to throw a “soft” or “catchable” ball, and … Continue reading The variable that seems to correlate the most to drops is QB accuracy. For example, the On Target % variable, defined as (completions + drops)/ Attempts, has a -.447 correlation with Marginal Drops. That means that accurate quarterbacks should be expected to have a lower drop rate, although this may be a more complicated question than you might think. The top 10 in most Marginal Drops had an On Target% average of 63.2% while the 10 QB’s with the fewest Marginal Drops had an average On Target% of 68.6%.

What else would you like to see done with the Drops data? Leave your thoughts in the comments.

References[+]

References
↑1	Some fine print: Unfortunately, that data is only recorded on a team level, not at the individual passer level. As a result, I gave each quarterback his pro rata portion of his team’s dropped passes relative to the percentage of team incompletions for the entire team. For example, let’s say the Jaguars have 30 dropped passes. Assume QB A for the Jaguars has 200 incompletions, and QB B has 100 incompletions. My methodology handled this by crediting QB A with 20 dropped passes and QB B with 10 drops. The numbers in this article are from 1992-2013. In the table below, “Marginal Drops” represents how many drops above average a quarterback had compared to league average rate. If a passer has positive Marginal Drops, this means he had more drops than expected.
↑2	Query whether the chicken or the egg came first here: are these quarterbacks considered good because their receivers didn’t drop passes?
↑3	Chase note: This was a common refrain by fans, the local media, and the Jets organlization during the Pennington era. He was known to throw a “soft” or “catchable” ball, and it seems as though the data here supports that theory.

Tagged as: Drops, Guest Posts, Munir Mohamed

{ 10 comments }