re: To defer or not to defer? LSU Coin Toss Results (Kelly Era)

Well, by restricting it to coin toss results, with enough data one may make a reliable statement about its potential effect on the game.

quote:

---

Fwiw, it’s much easier to look at the effect of the coin toss, and being in offense or defense first in overtime because there are fewer variables - but there is definitely a “right vs wrong” decision. That decision is not altered by what actually does happen in OT. Ie if the first team with the all scores a TD, and the second team with the ball does not score, it doesn’t mean it was the wrong decision to play defense first.

The same applies with the decision before the game, there are simply more variables.

---

Overtime is completely different. I understand what you’re getting at when you say “there are simply more variables” affecting the pregame decision, but it’s nonsensical.

The fact that one specific variable - the number of possessions per team - doesn’t exist in overtime is the entire reason the overtime decision is objective. It creates a huge advantage for the team who plays defense first, and that advantage doesn’t exist in regulation.

Among all the reasons people give for deferring at the beginning of the game, I’ve never once heard someone use the argument that “it lets you know whether you need to go for it in 4th down.” Because it doesn’t. In fact, if receiving the ball to start a half meant that the other team was more likely to get the last possession, you would want the ball to start the game because the last possession of the game is more valuable than the last possession of the first half.

So for you to say…

quote:

---

And that hypothesis is absolutely wrong.

---

Is it?

In statistics this would be the “null hypothesis.” If you want to evaluate whether either opening coin toss decision (receive or defer) has an impact on win/loss outcomes, you would compare your data to the null hypothesis - the expected range of outcomes if the decision doesn’t have any impact.

I don’t know that you ever really prove the null hypothesis to be true, since there’s always the question of whether you have enough data to see through the noise. If you flip a coin 100 times, there’s a 46% chance to get heads at least 51 times. Flip a coin 1,000 times and the chance to get heads at least 510 times goes down to 27.4%. More data reduces the probability that your “edge” (one way or the other) is due to chance.

But the data we have certainly doesn’t show it to be “absolutely wrong.”

Which is why we really don't need to be electing to receive the kickoff when we win the toss. i don't mind starting on offense if we lose the toss and the opposing teams defer but when we win the toss and elect to receive it beyond frustrating. There is zero reason to do it.

quote:

---

In overtime, each team on gets 1 possession and there is no clock. There are still four downs to get ten yards.

---

in overtime you absolutely HAVE TO at a minimum match your opponent on every single possession. Going 2nd provides you with the advantage of knowing exactly what is needed to at least match them.