1 Comment

Life is full of surprises...including the apparent agreement with the "analysis" of Greg Olsen on the 4th down decision. I could not disagree more.

First, let's establish a high-level understanding of what Olsen is saying here. His thoughts are not "analytics" as they has been mischaracterized. He provided no data in support of his viewpoint. It is at most – and I am being generous - unsubstantiated pop psychology. This was nothing more than his personal theory and gut instinct for what "should" happen. It's an approach to decision-making that sits uncomfortably near astrology, ESP, fortune tellers, and palm readers. You know, other gut instinct predictors.

Olsen instructs us that if the Cardinals go up by 6 instead of 3, they are in a worse position in terms of winning. This will come as a surprise to anyone with an arithmetic background who has foolishly thought for years that 6 is more than 3. Greg is here to tell you that what you have believed all these years is wrong when you step onto a football field. Let's call this the Olsen Paradox.

The proffered "explanation" for his Paradox is that Kevin O'Connell will be more aggressive on his ensuing possession which makes a TD and a resulting loss more likely. Put another way, he believes the Cardinals are in a superior position by keeping O'Connell passive in his play calls and getting to overtime. All of this, with zero data or evidence to support it.

There is so much obviously wrong here.

First, he advocates going for it on 4th and goal from the four yard line. Yet, he doesn't tell you how likely that strategy is to be successful on the road, with the defense backed up to the end zone with crowd noise - ironically, something for which there is data. My assumption is that the numbers do not favor going for it here, but Olsen doesn't bother to provide the data for the tradeoff at issue, so we don't know. No matter, Olsen then tells us that even if they don't get the TD, the psychology of the situation ensures that the Cardinals are still better off versus having more points.

You shouldn't believe that for a second.

There are at least two glaring problems with the Olsen Paradox. First, is reaching overtime a desirable potential outcome for Arizona under the circumstances? Again, what does the data tell us about the tradeoffs around getting to overtime here versus forcing your opponent to drive the length of the field and score a TD? Put another way, is getting to the outcome that Olsen advocates - being an underdog on the road in OT - more or less likely to result in a win? I'm reasonably sure the answer is less likely.

Beyond not understanding the tradeoffs at issue, the Olsen Paradox then fails to appreciate the context of the game and how that impacts the Paradox. There were over 3 minutes left in the game and the Cardinals had all three timeouts. O'Connell would know that if he kicked a game-tying field goal, there would still likely be between 1:30 and 2:00 left for the Cardinals with all three timeouts. Even following the Vikings TD, the Cards still had 1:13 left. O'Connell would know that tying the game in that context was far from a win when all Arizona had to do was drive 35 yards to kick the game winning FG with three timeouts in hand.

What that means is that the extra time on the clock would incentivize O'Connell to take the exact same approach of being aggressive and trying to score a TD to put the game out of reach. He would only settle for a tying FG and hopefully moving to OT if his hand was forced. Ultimately, you gained nothing from the alleged psychological “benefits” of eschewing three points. You are objectively worse off.

This is a classic example of poor reasoning driven by what poker player Annie Duke refers to as "resulting" - where you look at the outcome of a decision or advice retrospectively to judge the decision/advice instead of its quality when made. Yes, Olsen gave the advice before the outcome, but so what? He knew that if his theory didn't work out, nobody would remember or call him out for it. Instead, he was lauded when it just so happened to work out. Heads I win, tails you lose.

The Olsen Paradox is bad critical thinking on a number of dimensions. That is disappointing. What's more disappointing, however, was how so many gullible people swallowed his nonsense because he casts himself as the "analytics dude" on broadcasts.

Thanks for the great work as always.

Expand full comment