Five Thirty Eight's breakdown of the SB interception - Statistical data

[RiffRaff]

I love five thirty eight and the statistics fit the case (I believe top level offensive coordinators keep up with advanced statistics in cases like this which would explain the call in the first place), but the fact of the matter is, you still had a timeout. The case for running is made because it is overwhelmingly a much safer play to try and punch the ball in. You either score or you get stopped short and have a timeout to burn, the chances of you fumbling are slim. If you still want to pass on the next two downs you still have that option, but at least give the run a try because like I said you have a timeout to use. Not only did they not try to run the ball with a timeout left, but the type of pass play they decided to call went against all logic of what you do on a goal line. You have one of, if not the most mobile QB in the league that could he himself probably could have scored on a read option or naked bootleg that wasn't utilized. You called a slant in an area where the defensive backs can actually be aggressive with receivers and there isn't a lot of room to work with. Also you called a slant for a receiver that went straight across the middle . Either the receiver ran a bad route or you have to split him out farther so you can actually see whether or not the DB has a realistic shot of jumping the route. The speed of the play and the fact that it was in the most congested area of the field basically ensured that Russ wouldn't have known the DB was coming.

 

[Greenstrings]

Well if you read the cotdamn article you would see they factored Lynch's edge into the stat and it still came out pretty even to pass or run. We can all make an opinion on if it was a good or bad call but the numbers tell that probability was about even. That is all.

Nah first few paras is enough. I've reached my quota of 1 for longreads applying rigorous statistical analysis to a what if play. Bill Barnwell did it better without disingenuously making out as if his conclusions were definitive.

The probability wasn't even, its pretty clear that the author decided on the conclusion he was looking for before seeking out stats that reflect it.

That is how most statistical analysis works which is why you always take it with a truckload of sodium. It can be just as coloured by opinion or bias in pursuit of justification as anything else.

 

[ghostwriterx]

Nah first few paras is enough. I've reached my quota of 1 for longreads applying rigorous statistical analysis to a what if play. Bill Barnwell did it better without disingenuously making out as if his conclusions were definitive.

The probability wasn't even, its pretty clear that the author decided on the conclusion he was looking for before seeking out stats that reflect it.

That is how most statistical analysis works which is why you always take it with a truckload of sodium. It can be just as coloured by opinion or bias in pursuit of justification as anything else.

Basically. All these advanced stats and "probabilities" are interesting, but a football play is not a flip of the coin. Way too many variables that that are damn near impossible to properly account for.