Week 14 Discussion

General talk about Dynasty Leagues.
grooner
Player of the Year
Player of the Year
Posts: 2173
Joined: Fri Aug 10, 2012 10:01 am

Re: Week 14 Discussion

Postby grooner » Tue Dec 10, 2019 11:53 am

jenkins.math wrote: Tue Dec 10, 2019 9:55 am
honcho55 wrote: Mon Dec 09, 2019 11:43 pm Poking my head back in to this thread to comment on the PAT vs 2pt conversion.

PROv3’s posts are all spot on. That is all.
Yes Prov3's analysis is correct. But as I've learned both here and in daily life, most people are mathematically clueless and do not understand anything beyond basic arithmetic. People will argue the math but only because they don't understand or grasp it, so since they don't understand it, it must be incorrect. It's much easier to stick with the status quo than to data mine, think logically, and base a decision based on said logic.
His math is correct, but steelman's explanation was also spot on. Analytics are not nearly as useful in the NFL as they are in the MLB as there are so many less games and much higher variance, and as steelman said if you don't succeed quickly you will get axed. I am not saying that analytics are useless in the NFL, I think teams don't use them enough (such as going for it more on 4th down etc.).

Two quick examples:

Pat Odds 95%
Two odds 55%

Win Tie Loss Total
Pat 0.00% 90.25% 9.75% 100.00%
two 30.25% 49.50% 20.25% 100.00%


Pat Odds 95%
Two odds 50%

Win Tie Loss Total
Pat 0.00% 90.25% 9.75% 100.00%
two 25.00% 50.00% 25.00% 100.00%

So as you can see going for two twice gives you a much higher chance of winning outright, a much lower chance of tying and a much higher chance of losing. This is ignoring anything else such as time left, momentum, chance of getting the ball back etc.
If you want to win, go for two. If you think you have a significant advantage in OT, you may want to go for a Pat. Also, I just used 95% as an arbitrary number, I think the Pat success rate may be lower now that they moved it back.
There are numerous factors, but overall I agree that teams do not go for two enough.

User avatar
saw061600
Legend
Legend
Posts: 6768
Joined: Fri Oct 10, 2014 8:58 pm

Re: Week 14 Discussion

Postby saw061600 » Tue Dec 10, 2019 12:02 pm

In the end, all of the "go for 2 vs go for 1" doesn't really matter if the team subscribes to a much older and wiser philosophy of football:
Defense wins championships.

PSA
If you're not playing IDP, you're doing it wrong.
10 TM No PPR or waiver 85RST
Herbert Fields Garoppolo Ridder
JT Barkley Etienne Ingram Charb CEH
Evans Aiyuk Dionte Jeudy London JWill GWils EMoore JMyers Mims Moorex2
Andrews Kinkaid Pits Freiermuth
NBosa Quinnen DJJones Clark F-Myers Taylor Graham
Darius Okereke Kendricks DCampbell DJones Baker Kiser Brooks
Adams Simmons Vaccaro Joseph

12 TM .5ppr 45 RST
Herbert Stroud Mayfield
JT Achane Kamara Ford Pacheco Jaleel
AJB DK Godwin Aiyuk Kirk ZJones RMoore Thornton
Hock F-muth
LWilliams Payne Reed Greenard
Bernard Kendricks Warner Baker Williams Tranquil
Budda McKinney Clark Wilson

1-2QB 2-4RB 3-5WR 1-3TE 11OFF/DEF
Herbert Stafford Brock Dobbs
Taylor Jacobs Mattison Kyren Jaleel Ford Bigsby
AJB Diggs Evans Kirk McLaurin Dionte Boyd Renfrow JuJu
Kelce Pitts Deguara
Hutchinson Rousseau Greenard Travon Demarcus
QWilliams EJones TBernard Dean ShaqT Kyzir
Amos Bates Peppers Murray Fitz Clark

Jigga94
GOAT
GOAT
Posts: 16060
Joined: Thu Nov 09, 2017 8:38 pm

Re: Week 14 Discussion

Postby Jigga94 » Tue Dec 10, 2019 12:05 pm

When you go for 2, you only look good when it works

User avatar
Phaded
GOAT
GOAT
Posts: 11964
Joined: Thu Jul 03, 2014 2:32 pm
Location: Canada

Re: Week 14 Discussion

Postby Phaded » Tue Dec 10, 2019 12:10 pm

saw061600 wrote: Tue Dec 10, 2019 12:02 pmPSA
If you're not playing IDP, you're doing it wrong.
I wish we had up-voting sometimes.

User avatar
killer_of_giants
Ring of Fame
Ring of Fame
Posts: 3281
Joined: Thu Sep 27, 2018 8:20 am

Re: Week 14 Discussion

Postby killer_of_giants » Wed Dec 11, 2019 7:12 am

PR0v3 wrote: Tue Dec 10, 2019 10:17 am The data is based on 2-point conversion rates over the history of the NFL. By this point, the sample size should be large enough to be statistically significant.
no, because the sample is not homogeneous: if you sample the heights of man and women from around the world in the past century, that's hardly going to be significant on what height you would expect from a caucasian male born after 2000.

regardless, that's not even the variance you need to consider. you can assume 50% or 60% conversion rates (or whatever) without uncertainties, you still have to consider what's the variance on repeating the same "experiment" (the 2pt conversion) a number of times (let's say 60 times a season if your offense is good) assuming those same exact expected conversion rates.
and this is already assuming that the conditions of the "experiment" are the same, which aren't, which adds to variance.


jenkins.math wrote: Tue Dec 10, 2019 9:55 am Yes Prov3's analysis is correct.
it lacks an analysis of the uncertainties. i would expect someone with "math" in his username to pick up on that.

ericanadian
Legend
Legend
Posts: 6519
Joined: Mon Jul 02, 2012 12:13 pm

Re: Week 14 Discussion

Postby ericanadian » Wed Dec 11, 2019 7:18 am

I think in a defensive battle, going for the PAT probably makes more sense. However, if your offense is dominating and it’s a shootout, I’d go for two at every opportunity.

As for team strategy, there are still teams that play prevent way too early due to the exact same mentality that leads them to go for the PAT every time.
All I Der Is Win - 16 Team IDP League (Pass TD 6pts)

QB - Stafford, Stroud, Tune
RB - Swift, Hall, Penny, Bigsby, Ford
WR - Pittman, Olave, Di. Johnson, G. Wilson, J. Williams, Metchie, Robinson, M. Wilson
TE - Okonkwo, Schoonmaker
LB - Brooks, R. Smith, Phillips
DL - Crosby, Allen, Simmons
DB - D. James, Baker, Delpit
K - Just a stupid kicker

steelman
Ring of Fame
Ring of Fame
Posts: 3074
Joined: Sun Jun 10, 2012 4:19 pm

Re: Week 14 Discussion

Postby steelman » Wed Dec 11, 2019 8:36 am

The only stats that matter when it comes to going for 2 is Team A vs Team B. All the stats in the history of the NFL are irrelevant. Even if the math is right, the data is irrelevant.

Take the Steelers for most of this season. They're terrible in short yardage situations. Put them against a team that has a good short yardage D and see what the odds end up being. I would bet it's not going to be anywhere close to 50%.

That's the thing though, we don't have the data we need to come up with the proper stats. There's no long term data of Team A vs Team B, at least not with the same players, same coaches, same scheme, etc.

What GB did vs the Jets in 87 has no bearing on how effective the Steelers would be vs SF. Analytics and math can only go so far in pro sports. Coaches need to realize their strengths and weaknesses as well as their opponents strengths and weaknesses. If the Steelers go for 2 every time and are pushing 65% against weaker D's but only 35% vs stronger D's, then perhaps going for 2 with the game on the line vs NE or SF isn't a good idea. Kick the PAT and hope to win the game in OT. Or perhaps it's the Bengals vs the Saints and the the Bengals coaches realize that the fact they even have a chance to win the game is a miracle and they don't want to push their luck by going to OT, so they go for the win right away.

Stats and math have their place, but in pro sports, there's much more than that. Situation, circumstances, the flow of the game, injuries, players hot streaks, players struggles, etc. all need to be taken into account.

jenkins.math
All Pro
All Pro
Posts: 1589
Joined: Sat Sep 29, 2018 10:56 am

Re: Week 14 Discussion

Postby jenkins.math » Wed Dec 11, 2019 9:37 am

killer_of_giants wrote: Wed Dec 11, 2019 7:12 am
PR0v3 wrote: Tue Dec 10, 2019 10:17 am The data is based on 2-point conversion rates over the history of the NFL. By this point, the sample size should be large enough to be statistically significant.
no, because the sample is not homogeneous: if you sample the heights of man and women from around the world in the past century, that's hardly going to be significant on what height you would expect from a caucasian male born after 2000.

regardless, that's not even the variance you need to consider. you can assume 50% or 60% conversion rates (or whatever) without uncertainties, you still have to consider what's the variance on repeating the same "experiment" (the 2pt conversion) a number of times (let's say 60 times a season if your offense is good) assuming those same exact expected conversion rates.
and this is already assuming that the conditions of the "experiment" are the same, which aren't, which adds to variance.


jenkins.math wrote: Tue Dec 10, 2019 9:55 am Yes Prov3's analysis is correct.
it lacks an analysis of the uncertainties. i would expect someone with "math" in his username to pick up on that.
As someone with a math degree, that specialized in probability and statistics, and uses this on a daily basis, I'm well aware of any shortcomings it may have. If you really want to go down the rabbit hole mathematically we can take this as far as you can handle.

The problem with what you're asking for is that it hasn't been done. No team has gone for 2 60 times in a season, so you have to use the data that is available. That doesn't make it perfect, but that doesn't discredit it either. It would be awesome if we had that, but we don't. The data used was 2 pt conversion percentages over the history of the game, which doesn't care about weather, dome, home/road splits, offense rank vs defense rank in that particular game, etc. The variance is essentially already baked in to the sample size. It makes those individual breakdowns irrelevant for the discussion at hand. Unless you are in a lab environment where you have full control over everything in your experiment there will always be a level of variance. Since the NFL is played all over the country (and outside of the country a few times a year), and is officiated by humans, and played by humans, your "variance" stance basically forces you into a corner proclaiming that all data and advanced analytics when it comes to sports are completely irrelevant and false. I highly doubt you are trying to die on that hill.

The original premise was that you would have to convert at the historical league average or higher for this to be effective. Obviously I would assume that a team with a good offense that can utilize both the run and pass (the Ravens, 49ers, Seahawks immediately come to mind) would be much more likely to achieve the historical league average than a team like the Dolphins. If you want to say his original point totals of going for 2 every single time yielding somewhere between 90-120 might have a lower floor overall, I could see that stance, due to your "variance" claim. I would also say those numbers would only be for top half offenses (maybe only top 10, not sure since I haven't gone that in depth), and that bottom tier teams would have a much larger range of outcomes, with a much lower floor.


I'm well aware there are plenty of breakpoints within the data that would make for some interesting talking points, but I don't have the time nor inclination to do all that work. If you have that much free time feel free to get after it because I would be interested to see the results. But just pooping on the premise because the data isn't broken down enough for you and using the term "variance" doesn't mean the data was irrelevant or the math is meaningless like you're trying to imply.

User avatar
killer_of_giants
Ring of Fame
Ring of Fame
Posts: 3281
Joined: Thu Sep 27, 2018 8:20 am

Re: Week 14 Discussion

Postby killer_of_giants » Wed Dec 11, 2019 10:29 am

as i said earlier, i don't give a funk about the percentage of conversion and its variance: let the head coach assess his chances and feed them to the analyst, who could also do some calculations for every percentage and say which is the minimum one needed for the trick to work, so we don't really need to know that number, especially if pulled by a non-representative sample.

whatever methods the analyst uses, it will give an expected outcome (piss easy to do, basically what the head coach fancies his conversion rate to be times the TDs) and most important of all an uncertainty over this value, let's forget the term variance.
whatever the method, given the small amount of tries, this uncertainty will be significant.
so having an expected outcome which is marginally better than for the 1pt conversion, but with an uncertainty much larger than the gap between the two, means that you can't really tell what method is better or worse (especially considering that there are a lot of variables that we are not considering).

jenkins.math wrote: Wed Dec 11, 2019 9:37 amBut just pooping on the premise because the data isn't broken down enough for you and using the term "variance" doesn't mean the data was irrelevant or the math is meaningless like you're trying to imply.
math is meaningless if wrongly applied. applying gaussian statistics to NFL samples and not even assessing uncertainties will mostly result in irrelevant conclusions.

ericanadian
Legend
Legend
Posts: 6519
Joined: Mon Jul 02, 2012 12:13 pm

Re: Week 14 Discussion

Postby ericanadian » Wed Dec 11, 2019 11:02 am

Are we actually that short on data points here? I get that two point conversions are in short supply, but I would think we have substantially more data on plays with goal to go from the two. I think it’s fair to say that the end goals of those plays are so similar as to make them effectively the same for such an analysis. I get you’re still dealing with fairly small sample sizes, but they would be significantly less small grouped.
All I Der Is Win - 16 Team IDP League (Pass TD 6pts)

QB - Stafford, Stroud, Tune
RB - Swift, Hall, Penny, Bigsby, Ford
WR - Pittman, Olave, Di. Johnson, G. Wilson, J. Williams, Metchie, Robinson, M. Wilson
TE - Okonkwo, Schoonmaker
LB - Brooks, R. Smith, Phillips
DL - Crosby, Allen, Simmons
DB - D. James, Baker, Delpit
K - Just a stupid kicker

bjd5211
Legend
Legend
Posts: 5615
Joined: Wed May 03, 2017 11:50 am

Re: Week 14 Discussion

Postby bjd5211 » Wed Dec 11, 2019 11:12 am

ericanadian wrote: Wed Dec 11, 2019 11:02 am Are we actually that short on data points here? I get that two point conversions are in short supply, but I would think we have substantially more data on plays with goal to go from the two. I think it’s fair to say that the end goals of those plays are so similar as to make them effectively the same for such an analysis. I get you’re still dealing with fairly small sample sizes, but they would be significantly less small grouped.
They are not the same thing though, because the vast majority of goal to go situations the offense has another option or opportunity if it doesn’t get into the endzone on that one play, so it's played differently than a 2 point conversion where it's one and done.

jenkins.math
All Pro
All Pro
Posts: 1589
Joined: Sat Sep 29, 2018 10:56 am

Re: Week 14 Discussion

Postby jenkins.math » Wed Dec 11, 2019 12:29 pm

killer_of_giants wrote: Wed Dec 11, 2019 10:29 am as i said earlier, i don't give a funk about the percentage of conversion and its variance: let the head coach assess his chances and feed them to the analyst, who could also do some calculations for every percentage and say which is the minimum one needed for the trick to work, so we don't really need to know that number, especially if pulled by a non-representative sample.

whatever methods the analyst uses, it will give an expected outcome (piss easy to do, basically what the head coach fancies his conversion rate to be times the TDs) and most important of all an uncertainty over this value, let's forget the term variance.
whatever the method, given the small amount of tries, this uncertainty will be significant.
so having an expected outcome which is marginally better than for the 1pt conversion, but with an uncertainty much larger than the gap between the two, means that you can't really tell what method is better or worse (especially considering that there are a lot of variables that we are not considering).

jenkins.math wrote: Wed Dec 11, 2019 9:37 amBut just pooping on the premise because the data isn't broken down enough for you and using the term "variance" doesn't mean the data was irrelevant or the math is meaningless like you're trying to imply.
math is meaningless if wrongly applied. applying gaussian statistics to NFL samples and not even assessing uncertainties will mostly result in irrelevant conclusions.
So you're essentially saying that we should stick with the status quo just because. What more uncertainty do you have on a 2 pt conversion that you don't have on any other play throughout the course of a game? What amount of uncertainty has suddenly crept into the 2 pt conversion that renders 25 years of its existence and success/failure irrelevant? Are we not properly valuing the uncertainty of Bane burrowing tunnels underneath Gotham and blowing up the field right as the snap occurs?

The math is properly applied to the data set given. Obviously you would want to break that set into different subsets to get a clearer picture for an individual team or whatever data split you want to look at, but the overall numbers and "expected points for" are what they are when you look at the last 25 years. The change hasn't necessarily come from the 2 pt conversion itself, but the drop in certainty of the XP from the move back. The overall probability of an event and finding the probability of an event given (insert additional event/data here) would yield a clearer picture for whatever you're looking for specifically. However, just looking at the conversion rates over the last 5 seasons including this one, 4 out of the 5 seasons yielded at least half of the league converting their 2 pt attempts at a rate of at least 50%. Prior to that it was rare to have over half the league be successful at that clip, and typically about a third of the league never converted one over the course of the season. So the current trends (probably has a lot to do with the changes made to allow for more offense) are lending one to believe the 2 pt conversion is the way to go.

Each individual team would need to look at their ability to succeed and how the coach wants to play it moving forward, but you wouldn't just ignore the previous history and success rates when you are trying to formulate your gameplan if you have gotten that far. The issue is that it is much easier to do what everyone else is doing than think outside of the box, no matter how good the data is. If you kick the PAT and lose in overtime, it is what it is. If you go for 2 and aren't successful and lose at the end of the game because of it, people will call for your head. Everyone loved Pederson when he went for it on 4th down all the time, and he was successful and won a SB using that strategy all year. As soon as it doesn't work though, everyone turns on you awfully quick. Most can't take the emotion out of sports to make decisions like that. That's why you get the classic answer of "I went with my gut" sound bite. Most people believe perception is reality, and the long standing perception is to just kick the PAT. It made sense mathematically when it was a 15 yard kick, but it doesn't make as much sense any more mathematically.

ericanadian
Legend
Legend
Posts: 6519
Joined: Mon Jul 02, 2012 12:13 pm

Re: Week 14 Discussion

Postby ericanadian » Wed Dec 11, 2019 12:54 pm

bjd5211 wrote: Wed Dec 11, 2019 11:12 am
ericanadian wrote: Wed Dec 11, 2019 11:02 am Are we actually that short on data points here? I get that two point conversions are in short supply, but I would think we have substantially more data on plays with goal to go from the two. I think it’s fair to say that the end goals of those plays are so similar as to make them effectively the same for such an analysis. I get you’re still dealing with fairly small sample sizes, but they would be significantly less small grouped.
They are not the same thing though, because the vast majority of goal to go situations the offense has another option or opportunity if it doesn’t get into the endzone on that one play, so it's played differently than a 2 point conversion where it's one and done.
Why would having another shot change your approach? You want to score. You’re going to choose the play that is most likely to result in a score. What alternative goal would a coach choose to pursue from two yards out?
All I Der Is Win - 16 Team IDP League (Pass TD 6pts)

QB - Stafford, Stroud, Tune
RB - Swift, Hall, Penny, Bigsby, Ford
WR - Pittman, Olave, Di. Johnson, G. Wilson, J. Williams, Metchie, Robinson, M. Wilson
TE - Okonkwo, Schoonmaker
LB - Brooks, R. Smith, Phillips
DL - Crosby, Allen, Simmons
DB - D. James, Baker, Delpit
K - Just a stupid kicker

Jigga94
GOAT
GOAT
Posts: 16060
Joined: Thu Nov 09, 2017 8:38 pm

Re: Week 14 Discussion

Postby Jigga94 » Wed Dec 11, 2019 1:10 pm

ericanadian wrote: Wed Dec 11, 2019 12:54 pm
bjd5211 wrote: Wed Dec 11, 2019 11:12 am
ericanadian wrote: Wed Dec 11, 2019 11:02 am Are we actually that short on data points here? I get that two point conversions are in short supply, but I would think we have substantially more data on plays with goal to go from the two. I think it’s fair to say that the end goals of those plays are so similar as to make them effectively the same for such an analysis. I get you’re still dealing with fairly small sample sizes, but they would be significantly less small grouped.
They are not the same thing though, because the vast majority of goal to go situations the offense has another option or opportunity if it doesn’t get into the endzone on that one play, so it's played differently than a 2 point conversion where it's one and done.
Why would having another shot change your approach? You want to score. You’re going to choose the play that is most likely to result in a score. What alternative goal would a coach choose to pursue from two yards out?
1st and goal on the 2. The QB takes the snap, rolls out of the pocket, being chased, no one's open... throws it away to play 2nd down

2 pt conversion. The QB takes the snap, rolls out of the pocket, being chased, no one's open... Hucks it into the endzone hoping for a reception because there is no next down.

bjd5211
Legend
Legend
Posts: 5615
Joined: Wed May 03, 2017 11:50 am

Re: Week 14 Discussion

Postby bjd5211 » Wed Dec 11, 2019 1:16 pm

Say it's 2nd down from the exact same some spot you would take a 2 point conversion, you have a pass play with the only receiver that gets open is across the middle, but it's in a very tight window between two defenders that could pick it off, but completing it is the only chance of converting on that specific play. The smarter play in this instance is to just throw it out the back of the endzone and have another 2 downs to give you a better opportunity to put points on the board than risk turning it over and coming away with nothing.

Now you have the exact same scenario but it's a 2 point conversion rather than goal to go. Your only chance of coming away with points is on this one play, and you will be kicking off to the other team no matter the outcome. You make that throw into the tight window and hope the receiver can make the play and you can salvage the only opportunity you will have to come away with points. If it gets picked off it's almost certainly the same result as if it's incomplete, because while yes it's possible they return it for a 2 of their own, that is highly unlikely especially from the middle of the field.
You have almost nothing to lose and everything to gain by doing this on the 2 point conversion attempt, but you that risk is monumental on the TD attempt.


There are just so many factors that make them different, and there are multiple positive outcomes on most goal to go opportunities that don't necessarily have to result in a score on that one play. On a 2 point conversion though there is only one positive outcome possible on the play, and you only get the one chance to complete it.


Who is online

Users browsing this forum: Ahrefs [Bot], Bing [Bot] and 34 guests