Good Ws, Bad Ls and Buts

vicjagger

Well-Known Member
Messages
4,110
Reaction score
1,934
I was playing around with some NFL stats in Excel. I'm pondering the whole question of strength of schedule played, and I decide to look at the records of the teams beaten (and lost to) rather than the entire schedule played.

Some interesting things appear. Namely what I like to term good wins & bad losses (I'll get to the buts later).

A good win (GW), as I've defined it, is one in which you beat a team that has only lost to you. Conversely, a bad loss (BL) is one in which your opponent has beaten only your team.

There have been a total of 60 games so far this year. There have been 7 good wins and 8 bad losses, so just under 25% of the games played have fallen into one of these buckets. One game, PHI > DET, is the only game to count as a GW for the winner and a BL for the loser.

Good Wins:

Wk 1 TEN > JAX, SEA > T B
Wk 2 IND > TEN, ARI > SEA
Wk 3 NYG > WAS, PHI > DET
Wk 4 ARI > PIT

Arizona is the only team with 2 GWs.

Bad Losses:

Wk 1 BAL < CIN, CHI < S D, ATL < MIN
Wk 2 KC < CHI
Wk 3 MIA < NYJ, DET < PHI
Wk 4 NYJ < BUF, HOU < ATL

Now for the "but somebody has to lose" teams (the buts). Here are the top 12 teams that rank the highest in "win % of teams lost to".

Team---Total Losses--- Opp Record

1) NY Giants---2---8-0
1) Tennesee---1---4-0
3) San Diego---3--10-2
4) Denver------2---6-1
5) New Orleans-3---9-2
6) Cincy--------3---9-3
6) Buffalo-------3---9-3 <----
6) Minnesota----3---9-3
6) San Fran-----2---6-2
6) Tampa Bay---1---3-1
11) Philladelphia-3---8-3
12) St Louis-----4--11-5

Now some of these teams are downright lousy, but some others are certainly a product of the fact that they have played some tough schedules to date.

Notice that none of those teams have a "bad loss".
 
Interesting, but you forgot Rack always losses and it doesn't matter it's against. ;)
 
but its only week 4, who is to say any loss is the result of playing a good team? I mean aside from the 4-0 teams... who are the "good teams"? Its much too early. Say we beat Buffalo and they go 1-4 and then win out. Their record will be 12-4, will they still be labeled a bad team? etc.
 
DallasDW00ds0n;1685214 said:
but its only week 4, who is to say any loss is the result of playing a good team? I mean aside from the 4-0 teams... who are the "good teams"? Its much too early. Say we beat Buffalo and they go 1-4 and then win out. Their record will be 12-4, will they still be labeled a bad team? etc.

No the stats are adjusted each week. If you beat a team w/ a .750 record, no matter how many games you've played, it will likely be labeld a good win.

Since there are only 4 weeks played, the "good teams" are those that have no more than 1 loss.
 
I refuse to believe that KC losing to Chicago was a "bad loss," regardless of the fact that Chicago has subsequently lost 2 straight.

Oh and how is this criteria going to change as the season progresses? Eventually every team is going to have a loss and every team is going to get a win.
 
:umm:

So, the bottom line is that winning improves your record while hurting the record of your opponent....who'd a thunk it?

Seriously, I see you put a lot of effort into this and I appreciate the thought process but I don't see where it adds anything to the discussion. I don't think anyone would disagree with you that when an 0-3 team beats a 3-0 team it's good for the winner and bad for the loser.
 
peplaw06;1685301 said:
I refuse to believe that KC losing to Chicago was a "bad loss," regardless of the fact that Chicago has subsequently lost 2 straight.

Oh and how is this criteria going to change as the season progresses? Eventually every team is going to have a loss and every team is going to get a win.

You are correct on both points. Subjectively, we know that Chicago isn't going to finish the season 4-12, and KC isn't going to finish 12-4. But for now, 4 games in, they fit the mathematical definition. In a few weeks, their records will change and likely it would no longer fit any definition of a bad loss for KC.

The definitions of good wins and bad losses, as written in the OP, could really only work 3, 4 or 5 weeks into a season. Mathematically, the good wins are games where the losing team has a CURRENT winning percentage of .667 or greater. Conversely, the bad losses where games where the winning team has a current winning % of .333 or lower.

As to changing criteria, as the number of games increases to 7 or more, I may change the qualifying numbers a little. Instead of .667/.333, I might use .700/.300 or .750/.250. Obviously, the bit about "losing to a team that has beaten only you" won't apply...

onetrickpony said:
I don't think anyone would disagree with you that when an 0-3 team beats a 3-0 team it's good for the winner and bad for the loser.

No in fact any win is good for the winner and any loss is bad for the loser. But you seemed to miss the point that I'm referring to outliers here. I was using the words good and bad as modifiers to the concepts of wins and losses. Wins and losses that might be considered unexpected or extraordinary (at least mathematically, see Chi > KC above).

If a team beats a team that is currently 6-4 or 5-5 or worse, that is unimpressive. If they beat a team that is 7-3, or 8-2 or better, then we can call that a "good" win. See? (In retrospect impressive wins and unimpressive losses may have been better terms to use originally. Sorry)

This whole thing started with me listening to the arguments stating that Dallas and New England really hadn't beaten "anybody" yet. Beat a team with a good winning percentage, and you've beaten "somebody".

Finally, I'm not advocating winning percentage as the best method to determine the quality of an opponent, because, after all, who have they beaten? It is just one piece of the picture.

Maybe someday, I'll see if I can write a formula to calculate the winning % of your opponents' opponents' opponents' opponents' opponents' opponents.
 
peplaw06;1685301 said:
Oh and how is this criteria going to change as the season progresses? Eventually every team is going to have a loss and every team is going to get a win.


Maaaaaaaaaaaaaaaaaaaaaaaaaaaaaaybe.............:laugh1:
 
Whats interesting, is if you take the tiebreaker rules set by the NFL, The Packers would be seeded over the Cowboys based on Strength of Victory.

Of course its week 4....
 
they should change the tie breaker to who has the better looking fans, we would clearly win that one.
 

Staff online

  • Sarge
    Red, White and Brew...

Forum statistics

Threads
463,949
Messages
13,779,298
Members
23,770
Latest member
AnthonyDavis
Back
Top