Shame of the Game - Washington

[Dave Gray]

Yes it does make sense.  Those percentages factors assume both scoring opportunities were successful.

They do not.

The percentage means that at the decision point, the team that choose to attempt to go for it will win the game 3.5% more than the team that chooses to attempt the kick.  Then you have to factor in all the possibilities of missed TDs, going for 2, missed kicks, onside kicks, etc.  There are a lot of factors.

The real argument against is that the Dolphins are uniquely bad at going for it on short yardage as opposed to the average.

[Spider-Dan]

It's pretty easy to look at every scenario after a game and punch in the numbers and give the probability. These coaches are facing a play clock,  are not sitting in front of a computer, etc.

Any coach that actually cares about analytics and statistical probability would have to be a complete moron NOT to have a guy in the booth upstairs punching down and distance into a computer after every play; a person that can instantly tell him on his headset what his expected outcomes are.

[Phishfan]

I think that depends on exactly what is allowed. The internet isn't exactly clear on the subject but what I see is that only NFL approved hardware and software can be used during a game. What is included in those tools I don't know. I don't expect they have real time access to number crunching real time for every single scenario but they do probably have an analyst with a bunch of general situations already played out.

[Spider-Dan]

The point is that it's not McDaniel whipping out a calculator and figuring out the percentages himself while the play clock is running.  Teams that care about analytics will have a person in charge of that, and the head coach will simply ask them "What are the percentages on going for it vs. kicking the FG?", quickly receive an answer, and make a decision.  He's not just spitballing it.

[fyo]

It's pretty easy to look at every scenario after a game and punch in the numbers and give the probability. These coaches are facing a play clock,  are not sitting in front of a computer, etc. If you think McDaniel had any clue there was a 3.5% sway either way when he made his decision I think you are delusional. I may be naive.

He absolutely knew the percentage for the general case. Each team uses their own models, but if you don't think they have a chart or someone in the booth available with the numbers, I do think you are naive. 4th down, X yards to go, up Y points is a set of numbers that are very easy to run beforehand.

The 3.5% is obviously for the average case across the league and I would seriously hope that the Dolphins use a model that is optimized for their offense. Hopefully, they adjust the numbers every game for their opponent.

I would also hope that the McDaniel uses what he has seen all game long to tip the scales one way or another when deciding.

There's a lot of hope there and I obviously have zero clue what the Dolphins actually do.

Personally, it *felt* like the wrong call, but the Dolphins have actually been quite good in "power" situations this year, so I'm not able to point to easy numbers that back up my *feelings*.

[fyo]

Any coach that actually cares about analytics and statistical probability would have to be a complete moron NOT to have a guy in the booth upstairs punching down and distance into a computer after every play; a person that can instantly tell him on his headset what his expected outcomes are.

In this case, it's just a table. You could easily have that printed out and laminated nicely if there are limitations on software that teams can use during the game.

[Phishfan]

In this case, it's just a table. You could easily have that printed out and laminated nicely if there are limitations on software that teams can use during the game.

It would be a pretty big table. Consider how many variables need to be plugged in (remember this all started because we are talking about 3.5% difference) down, distance,  field position, time on the clock, etc. All being run at multiple variables prior to a game.