Statistic of the Week: 45.9%

I want to be clear right off the bat, in no way am I placing the blame for this game on the defense. They allowed 13 points and just one touchdown, this loss is on the offense. In fact, the fourth quarter play-calling was almost terrible enough to deter me from maintaining the focus of this post which I had been planning on writing for a while.

Feel free to skip right over the next paragraph.

Warning: rant coming….

RANT: What 4^th quarter play-calling you say (I realize nobody is actually saying this, just go with it)? CJ Spiller carried the ball just twice in the 4^th quarter, including not a single carry in the final 13:30 of the 4^th quarter. Chan Gailey called 12 pass plays and just 5 run plays in the final quarter. This all after Spiller averaged 8.58 yards per carry (103 yards/12 carries) and Fitzpatrick averaged 5.09 yards per attempt (107 yards/21 attempts) over the first three quarters of the game. I could go on for days…

Luckily, I decided not to devote this entire post to play-calling/getting Spiller the ball as it has been beaten into the ground enough by now and will surely be covered by countless others. 45.9% represents the Bills league worst 3^rd down failure rate. One of the most glaring issues in this week’s game against the Colts was the Bills complete inability to stop the Colts on crucial 3^rd downs. Be it 3^rd and inches or 3^rd and 12, the Colts were able to convert. Over the course of the game, they were able to convert on 3^rd and long several times including a conversion on a monumental 3^rd and 17 in the 3^rd quarter. The Colts averaged 6.7 yards per play on the 15 3^rd downs they faced today (not counting the kneel down on the last play of the game).

The ability to convert on third downs and the ability to stop opponents from converting on third downs has been shown to exhibit close ties to team’s win totals. No single underlying statistic has been shown to have overwhelming predictive power in determining win totals but 3^rd down success rate has been shown to be one of the better predictors. Below is the 3^rd down success rate of team’s opponents plotted vs the team’s win totals for every team over the last three years. Since we’re plotting the opponent’s success rate, we would expect wins to decrease as 3^rd down success rate (as listed here) increases.

The trend may not be perfect but it is a fairly small sample (96 sample points). There are three noticeable outliers in the upper-right side here with teams that had fairly high opponents’ 3^rd down success rate and still managed 14 or 15 wins. Those teams were the 2011 Packers, 2010 Patriots, and the 2009 Colts. The Packers and Patriots of those years are two of the most prolific offenses of recent memory, enough to make up for their porous defenses. We might be looking at a different picture with a more noticeable trend with those three outliers eliminated. Given the variance displayed here, this holds little predictive power but that doesn’t matter as all we’re looking for is a correlation. I won’t be as naïve as to say that if the Bills simply lowered their 3^rd down conversion rate on defense from 49% to 40%, they would be 6-5 instead of 4-7. However, the results here certainly show that there is some kind of correlation.

The defense has been better in the last few weeks and I want to be clear, the defense was far from the reason the team lost this game, but until they learn how to stop teams on 3^rd down, they will not be able to take that next step. Getting off the field on 3^rd downs is so crucial, and it’s something Buffalo has not been able to do all season long.