Zim’s Grades For The Game: Week 6 Edition

[royalshank]

Anyone who disagrees with his assessment obviously is delirious and a homer. Zim is spot on!

Well I disagree and stand by the notion that when you only surrender 234 yards, hold the other team to 20 min or whatever of TOP, force 7, 3 and outs your defense gets an A or an A- (for not coming up w a takeaway). So the individual grades for the defense made no sense to me. I’ll take that defensive performance every week and we’d win 15-17 games in the regular season. 16% on 3rd down? Teams don’t win when they convert at that terrible rate. Unless your opponent beats itself w penalties, missed FGs and surrendering 22 QB hits.

 

[FINFANFOREVER1972]

Well I disagree and stand by the notion that when you only surrender 234 yards, hold the other team to 20 min or whatever of TOP, force 7, 3 and outs your defense gets an A or an A- (for not coming up w a takeaway). So the individual grades for the defense made no sense to me. I’ll take that defensive performance every week and we’d win 15-17 games in the regular season. 16% on 3rd down? Teams don’t win when they convert at that terrible rate. Unless your opponent beats itself w penalties, missed FGs and surrendering 22 QB hits.

You obviously missed the 4th quarter when the run defense imploded. The defense has been God-awful all year.

 

[where's th'fish]

So my profession is statistician and have degrees in math and statistics so yep, I understand the math here. The flaw in this math logic is that you can’t apply a 50% make rate with certainty like you could if we were talking coin flip. Remember, this is nuanced and each event is isolated. Not dissimilar to the coin flip - “each flip is 50/50” but getting two heads (or tails) in a row on 2 consecutive flips is .5 x .5 = .25

And the above is a controlled environment. A football game is not. I don’t know what the make rate is for 2-pt plays - is it 50/50? What about when your backup is in? What about when your oline is giving up 20 pressures or a pressure on half of all plays? What about home vs away etc? So many variables and at the end of the day whether you go for two and get it on the first TD or second, theoretically the odds of making one of them are the same. The only advantage I can see in trying the first one is it takes the decision as to what to do on the second one off of the table

The average historical conversion rate across the league is 50%, but you're right that if the defense is dominating, it's a bad idea. I think the break-even point is around 38%-39% conversion rate.

The advantage of going for it on the first is information. If you go on the second, you win or lose depending on the outcome because you're out of time and options. If you go on the first, the information you gain gives you the ability to adapt your next choices: if you made it, you kick PAT the second time. If you fail, you can make up for it with another 2-pt conversion.

 

[royalshank]

You obviously missed the 4th quarter when the run defense imploded. The defense has been God-awful all year.

It gave up a big run. We shouldn’t have even been in that situation.

 

[where's th'fish]

Stupid losing math, pluck analytics, it didn’t make sense

Regarding the 2 point conversion attempt. I understand the math but still didn't like it. The team was not consistently executing cleanly on offense. The question isn't "Is it the proper call on average?"

The question is "Is this team on this day likely to make the conversion?"

If you can make less than 40% of the time it's a bad call, so yeah, it's possible for it to be a bad strategy if your offense is having a very bad day for whatever reason.

 

[royalshank]

The average historical conversion rate across the league is 50%, but you're right that if the defense is dominating, it's a bad idea. I think the break-even point is around 38%-39% conversion rate.

The advantage of going for it on the first is information. If you go on the second, you win or lose depending on the outcome because you're out of time and options. If you go on the first, the information you gain gives you the ability to adapt your next choices: if you made it, you kick PAT the second time. If you fail, you can make up for it with another 2-pt conversion.

Agreed. I don’t like the idea of being down 8 when scoring that 2nd TD though. A ton of pressure there and w your backup QB etc. But yes, I do see the benefit of having the info. I don’t hate the decision - in general I tend to lean toward “take your points and don’t reach until you have to” but that’s purely preference and not math speaking

 

[royalshank]

If you can make less than 40% of the time it's a bad call, so yeah, it's possible for it to be a bad strategy if your offense is having a very bad day for whatever reason.

I think w your starting QB I’d feel better. It wasn’t stupid at all though - just not how I like to go - I prefer to take my easier points all game and then reach when I have to but that’s not grounded in science here

 

[McDolphins]

I’d disagree on X. That whole defense held up really well except for the possession where we lost Needham and the 2 after turnovers late. Yeah, he got beat on a well designed rub route covering one of the best young WRs in the league. He still had a solid day.
But you’re right. I do want to see those picks happening again.

 

[McDolphins]

At this point I would honestly promote Skylar to QB2. If we’re talking about who gives us the best chance to compete it’s not even close anymore. Teddy needs to save what’s left of his body and retire after this year because he’s just not effective at even managing an offense.

 

[royalshank]

If you can make less than 40% of the time it's a bad call, so yeah, it's possible for it to be a bad strategy if your offense is having a very bad day for whatever reason.

Part of my gut feel thing here was probably not being confident enough in Teddy - rightly or wrongly

 

[royalshank]

I’d disagree on X. That whole defense held up really well except for the possession where we lost Needham and the 2 after turnovers late. Yeah, he got beat on a well designed run route covering his me of the best young WRs in the league. He still had a solid day.
But you’re right. I do want to see those picks happening again.

X erased Jefferson most of the game. The guy made one play. It was a big one but I’ll take that week in and out. As a whole we gave up just 234 yards of offense- and 100 came on that play and the Cook run - which we shouldn’t have even been in that spot. Minny couldn’t do anything on O all day

 

[where's th'fish]

Part of my gut feel thing here was probably not being confident enough in Teddy - rightly or wrongly

That's valid. I was only able to watch until we made it 7-3, so I have no feel for what the game was like. If the offense inspired zero confidence, then it's better to lock up the tie and kick the PATs. The 2-pt conversion early is not a strategy designed to tie, but to go ahead by a point.

 

[royalshank]

That's valid. I was only able to watch until we made it 7-3, so I have no feel for what the game was like. If the offense inspired zero confidence, then it's better to lock up the tie and kick the PATs. The 2-pt conversion early is not a strategy designed to tie, but to go ahead by a point.

To be fair he was stronger in the 4th so he had momentum going for him.

 

[DrMultimedia]

So my profession is statistician and have degrees in math and statistics so yep, I understand the math here. The flaw in this math logic is that you can’t apply a 50% make rate with certainty like you could if we were talking coin flip. Remember, this is nuanced and each event is isolated. Not dissimilar to the coin flip - “each flip is 50/50” but getting two heads (or tails) in a row on 2 consecutive flips is .5 x .5 = .25

And the above is a controlled environment. A football game is not. I don’t know what the make rate is for 2-pt plays - is it 50/50? What about when your backup is in? What about when your oline is giving up 20 pressures or a pressure on half of all plays? What about home vs away etc? So many variables and at the end of the day whether you go for two and get it on the first TD or second, theoretically the odds of making one of them are the same. The only advantage I can see in trying the first one is it takes the decision as to what to do on the second one off of the table

Thanks. I was a physicist back in my day and even taught an undergrad game theory class every so often. Glad someone posted the proper math.

And as you pointed out, the top issue is the fact that it's not 50/50. Various show stats that for an "average" team it's well below 50%, and as you said we aren't exactly a team that pushes people around (or protects the QB) when we need two yards. Almost last in third down conversion in general.

If you try on the first TD I find the psychology of not making it rather meaningful. You suddenly feel the pressure of needing the TD and a two point just to tie. The other team knows that and won't be as stressed about losing.

I'm really sick of so many coaches starting to say the call was due to analytics. If it's purey analytic there is no need for a coach at all.

Nice write up on Wikipedia. As they state, it's only mathematical favorable if you convert more than 38% of the time. I don't know if we do (and that is yet another issue...to have true stats you need a large data set).

Two-point conversion - Wikipedia

en.m.wikipedia.org

And a fun write up on 3rd down conversions (college data).

Comparing Third Down Conversion Rate by Distance to Go

In almost every game, announcers find a point to mention each team’s third down conversion rate. Sometimes it’s their conversion rate for that game only, and sometimes it’s their conversion rate fo…

staturdays.com

Ok... Enough math and psychology discussion for the morning. The topic got me so revved up I wrote a simulation for it and fired up my brain cells for the first time in awhile.

 

[BigNastyFish]

unlike 20 years ago -- you don't win in this league with defense you win with O. We all know that.

In today's game you just hope your D can slow the best teams down enough so you can outscore them.

Based on the stats yesterday -- our O must have defecated profusely and consistently all over the field.

Complimentary football means your O has to show up and score baby score!

That's our perennial problem. Can't score. Can't win. Rack 'em up and try again!