Talk about the NFL

[ponchi101]

And remember something else. There have been a lot, and it is a lot, of games this season that were decided by less than a FG. So, bravely going for that 4&1 in the 1st quarter instead of securing the 3 points may have real consequences this season.
If you are the Chiefs or the Bills, scoring 35 points per game, maybe not. But if you are the Jets or the Giants, you have no business going for it on 4&1, in the first half. Secure the points.
That would be another stat I would like to see. How many teams that lost by a FG had a FG chance earlier in the game and passed on it.

[ptmcmahon]

Yeah that's one thing i'm guessing the stats aren't differentiating... if one of the teams going for it is a Bills/Chiefs type offensive juggernaut... or a Giants/Jets one

[ti-amie]

[ti-amie]

[ponchi101]

That's not that bad.
In the CLE-LAC game, there was a roughing the passer that was literally one player patting Brisset.
I know you have to protect the players, but at least allow the coaches to challenge these calls.

[ponchi101]

And again.
The Bengals go for it on 4&1, at the 2 yard line. They don't convert, and of course, miss on the FG.
They lose by one FG.
It is insanity.

[patrick]

Analytics is playing a major role where games are decided by a FG or less

[ptmcmahon]

And again.
The Bengals go for it on 4&1, at the 2 yard line. They don't convert, and of course, miss on the FG.
They lose by one FG.
It is insanity.

Are you talking about the third quarter play? You can't directly corelate going for it on 4th down on that play and not converting with the loss. The rest of the game would have played out much differently. At the very least if everything else stays the same, Baltimore don't go for a field goal at the end and have to go for a touchdown.

I'm going to have to start writing down everytime a team that wins had a converted fourth down so I can say they only won because of analytics.

[ptmcmahon]

One interesting thing I found in my researching... the % chance to win models are all definitely a bit variable. My favourite example from yesterday - The Chargers went for it on 4th and 1 up 2 with a minute left and failed to convert. According to some stats models:

"Ben Baldwin’s model had the decision inherently giving the Chargers 11.7 percent in added win probability. ESPN’s model had the decision adding 5.2 percent in win probability. Next Gen Stats’ model had the Chargers losing 0.2 percent in win probability with the decision."

So quite the difference depending on the model.

Ps Can I claim this a win for stats? If the Charges had punted maybe Cleveland instead throw a couple deep passes and win with a shorter field goal? It's too bad the Chargers didn't convert because that would be a rare case where you could say the stats wound up sealing the win.

[ponchi101]

Of course you can claim any win. After all, how many NFL teams do I coach?
My point is this. Football is not a ZERO SUM game. It is a WIN game, in which the points you score have nothing to do with the points your opponent scores.
Let's say that you have the same situation, for the Bengals. The chances of converting on the FG are about 95%. So, as a FG is worth .95 x 3= 2.85.
But what were the chances of converting that TD? Work it back: ANYTHING that calculates under 2.85 is statistically the wrong choice. At 6 points/TD, you need a conversion rate of 47%.
I understand the stats. But as I said before: I want to know how many games are lost by less than a FG WHEN that team passed on a previous FG, by going for a TD.
In football, I think that the chance of losing one game out of 17 is too much to take that chance, especially if you are down in the score.

[ponchi101]

I will let you speak first now.
Did the Raiders need to go for 2, when the PAT would tie the game against a scoring machine?

[ptmcmahon]

Well the problem with your calculation is it's just getting expected points for that one play, but nothing that comes after. You can do that maybe if it's very late in the game, but a play that happens early in third quarter would vastly change what happens afterwards, so it's much harder to draw a direct result since there was so much game left.

The two point conversions last night were very interesting last night. I was hoping one of them would make it and I could say going for it won the game for a team...but alas neither was converted.

I think going for two for the Raiders was the right call giving them the chance for the lead against a team they hadn't stopped all half.

And then of course they stopped them. But if the Raiders take the lead, Kansas City probably play a little different since they are losing, instead of winning

So the key thing is it's so damn hard to say how one decision affects everything else. I'll have to watch next week again for a good winning example

[ti-amie]

[ti-amie]

[ti-amie]