♫ Run Run Pass PUUUUUNT ... Run Run Pass PUUUUUNT ♫

I'm curious as to where you got that data or what criteria went into compiling it because I have similar results using my own play by play data, but not exactly, and I'd like to figure out if the error is mine or theirs. I assume theirs , but it's not off by much so it could be as simple as I'm not including 1 play series that were wiped out by penalty. Like an incomplete pass on 1st down where there was defensive holding. That would technically count as its own "series", but I only include series were a play was actually run.

Here are my results for your chart.

      YOURS  MINE  DIFF   RRP     PCT
HOU    418    408   -10    41   10.05%
TEN    492    493     1    46    9.33%
PHI    478    470    -8    39    8.30%
NOS    433    451    18    37    8.20%
MIN    482    478    -4    38    7.95%
GRB    483    485     2    38    7.84%
NYG    457    455    -2    35    7.69%
IND    476    475    -1    36    7.58%
CLE    471    468    -3    35    7.48%
JAX    449    445    -4    33    7.42%
SAF    485    485     0    34    7.01%
ATL    444    439    -5    30    6.83%
MIA    469    470     1    32    6.81%
NYJ    457    456    -1    30    6.58%
CAR    475    474    -1    31    6.54%
WAS    480    475    -5    31    6.53%
DEN    458    450    -8    29    6.44%
DET    466    457    -9    29    6.35%
SEA    435    426    -9    26    6.10%
CHI    477    466   -11    28    6.01%
LAV    493    489    -4    29    5.93%
CIN    469    466    -3    27    5.79%
NEP    498    475   -23    27    5.68%
DAL    519    511    -8    29    5.68%
KAN    524    522    -2    27    5.17%
PIT    476    474    -2    24    5.06%
LOR    473    477     4    24    5.03%
ARZ    494    490    -4    24    4.90%
BUF    522    520    -2    23    4.42%
LOC    512    503    -9    22    4.37%
BAL    541    539    -2    22    4.08%
TAM    526    523    -3    19    3.63%

Yeah that one would be a bit tricky since there can be multiple series per drive, so the multiple series drives would have to be filtered out from what I've done.

I was working on a similar chart that broke down every 3 play combination, to maybe find out which one led to the most success, but I learned there was a problem with looking at it that way. As I mentioned earlier, a 3 play series has to automatically assume the 1st and 2nd play failed to pickup a first down, so it would then maybe boil down to average yards to go on 3rd down and what play was run on 3rd, but then you have to take into consideration the NFL's tendencies on 3rd down, which is to pass if it's long, or run if it's short, so that can make the results biased and misleading, and it did as you can see on the charts below lol.

Up first, I got the data broken down by each of the possible combinations. Teams sorted alphabetically.



ATT = total number of that series
1ST = 1st downs on 3rd down (doesn't count if team went for it on 4th and converted)
PUN = punts on 4th down
FG = field goals on 4th down
CONV% = how often the 3rd down play was converted (so again not 4th down)

And for this one it's broken down by down. So for example, if a team ran on 1st down it would be a total of what they did on the 3 play series regardless of what they did on the other 2 downs. Fun fact: 3 play series begin with a run 55.9% of the time. Which is probably why they end up taking 3 plays .



So according to the data, it would look like run, run, pass isn't conservative enough, and you should actually run, run, run lol. But no, if you take into consideration everything I mentioned earlier you can quickly see how it's all misleading. This leads you to believe running is a much better strategy. Hey, maybe these are the kind of charts Zimmer's been looking at?

All I know is the Vikings were below average in almost every category except run, run, pass ironically lol, or just running on 2nd down in general. Their 3rd down TO-GO avg was also a lot higher than avg in almost all categories too.

Conclusion: The chart probably needs to include the 1 and 2 play series too, to get a better picture. For example, if you look at all NFL series that began with a pass play (remember sacks count as pass plays), 28.4% resulted in 1st down on the first play, while only 13.5% of them .... ahhh hell this will look better in chart form instead lol. We'll even do a column without sacks for comparison.

         INCLUDING   IGNORING
                SACKS      SACKS
pass            28.4%      29.8%
run             13.5%      13.5%

pass/pass       32.9%      36.8%
run /run        34.3%      34.3%

pass/run        32.4%      33.6%
run /pass       33.8%      35.8%

pass/pass/run   52.0%      57.7%
pass/run /run   62.4%      63.2%
run /pass/run   54.6%      57.1%
run /run /run   49.4%      49.4%
run /run /pass  41.1%      45.2%
run /pass/pass  35.9%      40.9%
pass/run /pass  38.8%      44.5%
pass/pass/pass  33.8%      39.8%

I guess sacks don't move the needle too much, and hmmm maybe it is better to run on 1st down? Nah, the very first 2 rows tell you which is better. Running on first down is how you end up with more 2 and 3 play drives to begin with. In other words, you gotta fail on 1st down to succeed at running lol. Makes sense.