Week 9 BEST & WORST Decisions

For this week’s Best and Worst Decisions, we’ll focus on the back-and-forth ARI vs MIA game, specifically the second half of this game and the five pivotal 4^th down decisions these teams faced.

BEST Decisions – ARI vs MIA

In terms of correct decisions there were three, and all were by ARI.

In the 3^rd quarter ARI made a decision that increased their win probability by +7.7%.  And, in the 4^th quarter ARI made two more correct 4^th down decisions (on the same drive), that increased their win probability by +10.4% and +10.3%, respectively.  In these three decisions, ARI increased their (pre-snap) win probability by +28.4%.

Great calls Coach Kingsbury! 

WORST Decisions – ARI vs MIA

In terms of errors, there were two. 

In the 4^th quarter with 3:35 remaining, a MIA decision to attempt the FG cost them -7.0% in (pre-snap) win probability.  And, on the very next drive a similar ARI decision to attempt the FG with 1:58 remaining, cost them -21.2% in (pre-snap) win probability.

Decisions in context…

Let’s take a closer look at the ARI error, in context.  Prior to his decision-making error, Coach Kingsbury faced three other critical decisions and got them all right.  The first two decisions resulted in successful 4^th down conversion attempts.  The third decision (Q4 5:20 on the clock) resulted in a failed 4^th down conversion attempt – and, it’s very common to see coaches incorrectly change their decision-making strategy based on a recent play’s result (i.e. recency bias).

Empirical data suggests that an average NFL team is expected to successfully convert on 4^th and 1 about 70% of the time – which also means they are expected to fail to convert about 30% of the time.  Based on our Edj Power Indexes (EPIs), we had ARI as substantial favorites to convert on what ended up being an unsuccessful attempt – a prime example of a good/correct decision that had a bad execution/result.

The analytics behind our recommendations are based on an in-depth understanding and analysis of more than 20 years of play-by-play data, which includes the realization that not every attempt results in a successful conversion.  Analytics are about looking at the best available information at the time of decision (i.e. prior to the snap) and making the decision that gives your team the best chance to go on to win the game – as opposed to judging decisions purely on their outcomes (i.e. whether or not the play was successfully executed).   

On the 4^th down attempt that ARI failed to convert, the decision to go for it came with an expectation to go on to win the game 68.9% of the time and a decision to attempt a FG came with an expectation to go on to win the game 58.6% of the time.  The 10.3% difference in win probability between these two options makes the decision to go for it the correct call, regardless of play execution. 

However, it’s not hard to imagine how a good decision that had a bad result, played a role in Coach Kingsbury’s ultimate decision-making error to attempt a FG instead of going for it (Q4 1:58 on the clock)  – a decision that cost his team 21.2% in (pre-snap) win probability.

We recognize that it’s easier said than done, but analytics can give coaches the confidence to know they are making decisions that give their team best chance to go on to win the game, and not get caught up second-guessing decisions based on play results. 

Notable High School Calls of the Week

Cadillac High School Vikings (MI) at Petoskey High School Northmen (MI)

With 8:37 left in regulation and the game tied 14-14, Coach Mallory and the Vikings faced a 4th and 3 at the Petoskey 27-yard line. By deciding to go for it instead of attempting a FG, Coach Mallory increased the Viking’s (pre-snap) win probability by +9%. The Vikings would go on to defeat the Northmen 27-14.