for display only
Big Blue Interactive The Corner Forum  
Back to the Corner

Archived Thread

Some thoughts on NFL Tiebreaking

ray in arlington : 12/3/2020 12:33 pm
We’re watching December football, and that means teams trying to get playoff spots. As someone who has had interest in figuring out NFL tiebreak scenarios since I wrote my first computer program to do it back in 1982, I have had a chance to see it play out for decades.

There’s certain phrases you’re going to hear this time of year. Here’s some thoughts about them. Also I have some comments on how the media presents this stuff and the tools that are provided for fans to use.

=======================================
“Team X controls their own destiny”
======================================
This is fine, as it just appears to mean that if a team wins the rest of their games, they are in the playoffs regardless of other results. So “no help” is required. However, it doesn’t mean that if you are tied with a team with 4 games to go, and both of you go 3-1, you will get in. I think this is well understood – “control your own destiny” just deals with the case where you win out.

The quibble I have about it is that it can be a weak statement compared to more precise statements. For example, the Giants were said to “control their own destiny” with 4 games to go in the 2011 season. But their situation was actually better than that. They had to beat DAL twice and either WAS or NYJ. They could drop the WAS game and still get in without needing help. That is what happened. The Giants lost for the second time that season to Rex Grossman and the 5-11 Redskins. And that Giants team won the super bowl.

===================================
“Team X owns the tiebreak over Team Y”
===================================
This is usually invoked when there is a possibility of finishing with the same record as a team you have beaten. That’s fine but may not matter if team Z gets into the tie as well. Team Z might win the 3 way tiebreak. Note that if team Z didn’t play Teams X or Y the head-ot-head result can’t be applied (until team Z advances or is eliminated). I was listening to a Giants radio broadcast in 2016 when Bob Papa pointed out that the Giants “owned the tiebreak over Detroit”. The problem was that, with 2 games to go, the only scenarios where the Giants could get eliminated involved finishing with the same record as Detroit. On BBI a couple of folks posted that they wanted GB to win the NFC North because we “owned the tiebreak over Detroit.” What was happening was that the only way for the Giants to get eliminated was to get in a 3-way tie with DET and ATL.

This year the Giants “own the tiebreak against WAS” due to having swept WAS but don’t necessarily win a 3-way tiebreak with WAS and DAL.

=================================
“If the season ended today”
=================================

This is the way that media likes to give an idea of who is going to make the playoffs. That’s ok as a possible outcome, but it sometimes leads to the wrong conclusion. In 2015, the Vikings had the #6 seed “if the season ended today” with 2 games to go. It was being said on broadcasts if they won they would make the playoffs. The problem was that they had the #6 seed because the Giants were in 3rd place with the same record as the 2nd place team in the NFC east. A 3rd place team can’t get a spot (unless the 2nd place of them gets in and there is another spot to fill). But a NYG win would automatically get them out of 3rd place. So it was NYG that would get in, not MIN, with 2 more wins. Unfortunately, the Giants laid an egg against the Ravens and got knocked out.

Generally speaking, if you have a seed “if the season ended today”, it doesn’t always mean you get in if you win out.

========================================
“It’s just a non-conference game”
========================================
Before the NFL went to 32 teams, the tiebreaking had a simple aspect to it. There was a hierarchy of games. You had the head-to-head, the division games, then conference games to look at. The Giants games against the AFC would have the least impact on tiebreak. But the NFL slipped in a trick when they went to 32 games. For ties in the division, they made it head-to-head, division, common opponents – and then conference. (I presume that was because with the new schedule set up the common opponents were are larger set of games for teams in the same division.) Since the AFC games are common opponents with the team in your division, that changes the picture. The “least important” games become the 2 non-common opponents games. These are NFC games for the Giants.

Actually, this tweak is why DAL can’t win a 3-way tiebreak with NYG and WAS this year if they lose to NYG. DAL already won a non-common game against ATL. The Giants lost both of theirs, to TB and CHI. So DAL can’t win the common opponents tiebreak vs NYG, and they can’t take the earlier tiebreaks.
Now for the wild card, it is true that the non-conference games are the least important. But not always because if 2 teams from the same division are part of the wild card tie, you have to use the division tiebreakers to determine which one is the “upper team” to compare against the team from the other division. Arggh….

Overall, I’d still say the AFC games are “less important” for the Giants, but when they end up in a situation where realistically the division title is the only way to get in (as in 2011 or 2020), that’s not the case.

=============================
The Simulators and the media
======================================================
I don’t think the NFL has ever been wrong in the tiebreaking information they provide. But they only talk about what can happen in the upcoming week. They don’t present all the scenarios for the rest of the season with 2 or 3 weeks to go. It’s just a matter of saying who can clinch week-to-week.

I am used to hearing wrong stuff from broadcasters and also seeing wrong stuff in the NY Post. The NY Post had wrong information about tiebreakers back in 2016, the last time the Giants made it. Even last year the NY Post had an article saying our 2020 schedule was set and that we would be playing Atlanta this year. That didn’t happen. (The article was re-edited later).

So to look into this stuff, you can use the simulators that are available. The ESPN one has had multiple mistakes over the years. I ventured once into a blog where folks were discussing the errors in the ESPN simulator. The folks there were on a higher level than I. I think the bug in their program has been cases where a team still has 2 games against the same opponent before the year is out (like the Giants with 4 games to go in 2011). So I don’t trust the ESPN simulator. There is also a YAHOO playoff generator. It gives you multiple ways to start the simulator before you select your results, like picking all the home teams to win. The problem I had one year the result I got by putting in my specific scenario depended on how I started the simulator. That shouldn’t happen.

I haven’t seen any bugs in the NY Times simulator. When it was computing how your team could get the first 3 picks in the draft, it was not handling it correctly a few years ago. (I think they just assumed it was like doing the playoff tiebreak, but only in reverse – which doesn’t work.) I notice that they removed the draft predictions for a couple of years, but came back with it this year. But I haven’t noticed a mistake in how it handles playoffs. Somebody just posted a playoff generator from fivethirtyeight. I haven’t had a chance to try that one yet.

Anyway, that’s a pretty tough read, but hopefully gives an idea of how involved some of this can be.
WOW! That's a lot of work (and typing) Ray  
pivo : 12/3/2020 12:59 pm : link
Many thanks for all you do
RE: WOW! That's a lot of work (and typing) Ray  
ray in arlington : 12/3/2020 1:09 pm : link
In comment 15065124 pivo said:
Quote:
Many thanks for all you do


Thank you pivo. I hope someone finds some value in it. Wish I could explain better.
RE: RE: WOW! That's a lot of work (and typing) Ray  
Big Blue '56 : 12/3/2020 1:17 pm : link
In comment 15065135 ray in arlington said:
Quote:
In comment 15065124 pivo said:


Quote:


Many thanks for all you do



Thank you pivo. I hope someone finds some value in it. Wish I could explain better.


You’ve explained it perfectly, per usual..Thank you
Haha  
ray in arlington : 12/3/2020 1:20 pm : link
I wrote "32 games" instead of "32 teams".
RE: Haha  
Big Blue '56 : 12/3/2020 1:23 pm : link
In comment 15065144 ray in arlington said:
Quote:
I wrote "32 games" instead of "32 teams".


Perfect!
RE: RE: WOW! That's a lot of work (and typing) Ray  
FatMan in Charlotte : 12/3/2020 1:26 pm : link
In comment 15065135 ray in arlington said:
Quote:
In comment 15065124 pivo said:


Quote:


Many thanks for all you do



Thank you pivo. I hope someone finds some value in it. Wish I could explain better.


Every year we find value in the playoff scenario breakdowns.

One of the best regular posts on BBI!
These posts are priceless  
Go Terps : 12/3/2020 1:39 pm : link
I read every one.
RE: These posts are priceless  
djm : 12/3/2020 2:42 pm : link
In comment 15065168 Go Terps said:
Quote:
I read every one.


I would throw up in my mouth a little when reading the draft scenarios and inevitably stopped doing so. These are much more palatable. Let's hope we see a lot more of these and less of the former threads going forward.
Good Chance 6-10 Is Good Enough  
LTIsTheGreatest : 12/3/2020 2:56 pm : link
as long AS one of the wins is against Dallas. I'd say they have a good chance to beat Cleveland and maybe AZ. Don't think they beat Seattle though it will be a close game. Same with Baltimore as long as Jackson plays
Back to the Corner