If you are a statistical analyzer, and break down numbers on NFL games with any sort of regularity, you know that TO margin is an incredibly important statistic in the outcome of all games. Straight up, ATS – doesn’t matter. Perform well in this area, be it taking care of the ball, or taking the ball away from your opponent on a regular basis, and the likelihood you are playing football come January greatly increases. Most NFL fans, even the casual fan that doesn’t get too involved in breaking stats down is aware of this phenomenon. But how good a tool can it be to predict future outcomes, whether it be on a week to week basis trying to pick straight up winners in your office pool, picking a few games to wager your hard earned cash on, or even trying to predict the chances your team has at making the playoffs? It’s very useful, and below I will show you in Part I of my NFL Turnover Analysis one way it can be used, with more entries to follow discussing additional angles and ways to utilize this one stat.
In Part I let’s examine the relationship between turnovers, points scored vs. points against (which is a team’s points margin), and projected records based on those two stats.
Most handicappers and others who work with turnovers in their models will value these at approximately being worth 4 points either way – takeaway’s are worth +4pts, while giveaway’s impact points margin (4pts). Of course this number is not set in stone, and can be debated – you may find turnovers are worth a different amount, but since that is a sort of “industry standard”, I will use that today for this analysis. In reality, any number you select within reason (the number has to be worth anywhere between a minimum 2pts and a maximum 5pts because a turnover either way leads to the possibility of scoring or allowing points – working with estimated % chances of scoring/allowing a FG/TD will allow you to derive your own worth of a turnover) will work as long as it is consistent across the board for each team. By using that method of “valuing” turnovers, we can calculate a new points margin based on a team’s pure play performance – stripping away the advantage/disadvantage turnovers had on their points margin. This is a valuable way to place a barometer on how team’s truly performed, statistically speaking, in their games.
Let’s look at a matrix with 2010 information (which would have been used heading into last year, 2011) that shows this information:
2010
|
TO
|
*Normalized
|
2010
|
New Proj
|
Wins
| |
Margin
|
Adv/(Dis)
|
Margin
|
Record
|
Record
|
Impact
| |
ARI
|
(9.06)
|
(1.25)
|
(7.81)
|
5-11
|
4-12
|
(1)
|
ATL
|
7.69
|
3.50
|
4.19
|
13-3
|
10-6
|
(3)
|
BAL
|
5.50
|
1.75
|
3.75
|
12-4
|
10-6
|
(2)
|
BUF
|
(8.50)
|
(4.25)
|
(4.25)
|
4-12
|
6-10
|
2
|
CAR
|
(13.25)
|
(2.00)
|
(11.25)
|
2-14
|
3-13
|
1
|
CHI
|
3.00
|
1.00
|
2.00
|
11-5
|
9-7
|
(2)
|
CIN
|
(4.56)
|
(2.00)
|
(2.56)
|
4-12
|
6-10
|
2
|
CLE
|
(3.63)
|
(0.25)
|
(3.38)
|
5-11
|
6-10
|
1
|
DAL
|
(2.63)
|
0.00
|
(2.63)
|
6-10
|
6-10
| |
DEN
|
(7.94)
|
(2.25)
|
(5.69)
|
4-12
|
5-11
|
1
|
DET
|
(0.44)
|
1.00
|
(1.44)
|
6-10
|
7-9
|
1
|
GB
|
9.25
|
1.50
|
7.75
|
10-6
|
12-4
|
2
|
HOU
|
(2.31)
|
0.00
|
(2.31)
|
6-10
|
6-10
| |
IND
|
2.94
|
(1.00)
|
3.94
|
10-6
|
10-6
| |
JAC
|
(4.13)
|
(3.75)
|
(0.38)
|
8-8
|
8-8
| |
KC
|
2.50
|
2.25
|
0.25
|
10-6
|
8-8
|
(2)
|
MIA
|
(3.75)
|
(2.75)
|
(1.00)
|
7-9
|
7-9
| |
MIN
|
(4.56)
|
(2.75)
|
(1.81)
|
6-10
|
7-9
|
1
|
NE
|
12.81
|
7.00
|
5.81
|
14-2
|
11-5
|
(3)
|
NO
|
4.81
|
(1.50)
|
6.31
|
10-6
|
11-5
|
1
|
NYG
|
2.94
|
(0.75)
|
3.69
|
10-6
|
10-6
| |
NYJ
|
3.94
|
2.25
|
1.69
|
11-5
|
9-7
|
(2)
|
OAK
|
2.44
|
(0.50)
|
2.94
|
8-8
|
10-6
|
2
|
PHI
|
3.88
|
2.25
|
1.63
|
10-6
|
9-7
|
(1)
|
PIT
|
8.94
|
4.00
|
4.94
|
12-4
|
11-5
|
(1)
|
SD
|
3.56
|
(1.50)
|
5.06
|
9-7
|
11-5
|
2
|
SF
|
(2.56)
|
(0.25)
|
(2.31)
|
7-9
|
6-10
|
(1)
|
SEA
|
(6.06)
|
(2.25)
|
(3.81)
|
7-9
|
6-10
|
(1)
|
STL
|
(2.44)
|
1.25
|
(3.69)
|
7-9
|
6-10
|
(1)
|
TB
|
1.44
|
2.25
|
(0.81)
|
10-6
|
7-9
|
(3)
|
TEN
|
1.06
|
(1.00)
|
2.06
|
6-10
|
9-7
|
3
|
WAS
|
(4.69)
|
(1.50)
|
(3.19)
|
6-10
|
6-10
|
Legend:
2010 Margin: points scored – points against
TO Adv/(Dis): represents the per game impact each team’s turnovers had on their points margin. This number is derived by taking the total TO Margin on the season, dividing by 16 to get a “per game” TO Margin, then multiplying by 4 because as mentioned above, I am valuing each turnover to be worth 4pts
Normalized Margin: calculated by taking 2010 Margin – TO Adv/(Dis). This figure represents the points margin each team would have w/o the impact of turnovers
2010 Record: straight forward, each team’s true SU record
New Projected Record: uses the Normalized Margin calculation, and fits each into a matrix (that is relatively standard across the industry) which plugs teams into certain records based on their points margin. It is generally assumed that team’s who outscore their opponents by 1.5ppg will go 9-7, 3ppg 10-6, 5.5ppg 11-5 and so on increasing ppg by 2.5pts for each win – and using the reciprocal of each of these ppg marks for losing records. Note, since these figures are quoted in decimals, the sum of 252 wins does not equal 256, the amount we see in an entire NFL season if there are no ties.
Wins Impact: calculated by taking New Projected Record – 2010 Record. This column basically shows the impact turnovers had on each team’s actual SU wins and losses.
Now that you understand the data, here is where it gets useful. As mentioned, a “model” or any analysis is only good if you backtest it, and prove that it has worked in the past. You can read about models, statistical impacts of various items on games, but unless it has been proven to be a solid indicator of performance in the past, you will be wasting your time.
Let’s first focus on the teams I have highlighted in red, those are the team’s that achieved a record in 2010 that was above and beyond their actual performance disregarding the impact of turnovers. These team’s we forecasted to drop in wins from 2010 to 2011 because as we know, turnovers typically, but not always, revert back to the mean – so a team’s performance that was positively impacted by a strong TO margin the prior season often sees the opposite occur in the very next season. And wouldn’t you know, of the 7 teams we forecast to have a drop in wins last year, 6 actually did drop while one stayed the same:
- ATL: 13 to 10
- BAL: 12 to 12
- CHI: 11 to 8
- KC: 10 to 7
- NE: 14 to 13
- NYJ: 11 to 8
- TB: 10 to 4
Now let’s move onto the team’s that were negatively impacted by TO margin in 2010, which means we expected this group to have a strong record in 2011 comparing to 2010 – and once again, of the 6 teams we forecast to have an increase to their win total, 4 did while another stayed the same:
- BUF: 4 to 6
- CIN: 4 to 9
- GB: 10 to 15
- OAK: 8 to 8
- SD: 9 to 8 – the only team of the 13 that were projected to move one way or another that went opposite
- TEN: 6 to 9
Looking back on 2010 figures used to project 2011 standings, this analysis had 13 teams sliding one way or another in wins – and of those 13, 10 moved the way predicted, 2 stayed the same, and only 1 moved opposite – and that was by just one game.
Not bad, huh? So you are probably saying, hey, let’s see this data for the upcoming season – is my team on this list? Well, of course I will show that data, so without further ado, here is data based on last season that can be used to forecast the 2012 NFL season:
2011
|
TO
|
*Normalized
|
2011
|
New Proj
|
Wins
| |
Margin
|
Adv/(Dis)
|
Margin
|
Record
|
Record
|
Impact
| |
ARI
|
(2.06)
|
(3.25)
|
1.19
|
8-8
|
9-7
|
1
|
ATL
|
3.25
|
2.00
|
1.25
|
10-6
|
9-7
|
(1)
|
BAL
|
7.00
|
0.50
|
6.50
|
12-4
|
11-5
|
(1)
|
BUF
|
(3.88)
|
0.50
|
(4.38)
|
6-10
|
5-11
|
(1)
|
CAR
|
(1.44)
|
0.25
|
(1.69)
|
6-10
|
7-9
|
1
|
CHI
|
0.75
|
0.50
|
0.25
|
8-8
|
8-8
| |
CIN
|
1.31
|
0.00
|
1.31
|
9-7
|
9-7
| |
CLE
|
(5.56)
|
0.25
|
(5.81)
|
4-12
|
5-11
|
1
|
DAL
|
1.38
|
1.50
|
(0.13)
|
8-8
|
8-8
| |
DEN
|
(5.06)
|
(2.50)
|
(2.56)
|
8-8
|
6-10
|
(2)
|
DET
|
5.13
|
2.75
|
2.38
|
10-6
|
10-6
| |
GB
|
12.56
|
6.00
|
6.56
|
15-1
|
11-5
|
(4)
|
HOU
|
6.44
|
1.75
|
4.69
|
10-6
|
11-5
|
1
|
IND
|
(11.69)
|
(3.00)
|
(8.69)
|
2-14
|
4-12
|
2
|
JAC
|
(5.38)
|
1.25
|
(6.63)
|
5-11
|
5-11
| |
KC
|
(7.88)
|
(0.50)
|
(7.38)
|
7-9
|
4-12
|
(3)
|
MIA
|
1.00
|
(1.50)
|
2.50
|
6-10
|
10-6
|
4
|
MIN
|
(6.81)
|
(0.75)
|
(6.06)
|
3-13
|
5-11
|
2
|
NE
|
10.50
|
4.25
|
6.25
|
13-3
|
11-5
|
(2)
|
NO
|
13.00
|
(0.75)
|
13.75
|
13-3
|
14-2
|
1
|
NYG
|
(0.38)
|
1.75
|
(2.13)
|
9-7
|
7-9
|
(2)
|
NYJ
|
0.88
|
(1.25)
|
2.13
|
8-8
|
9-7
|
1
|
OAK
|
(4.63)
|
(1.00)
|
(3.63)
|
8-8
|
6-10
|
(2)
|
PHI
|
4.06
|
(3.50)
|
7.56
|
8-8
|
12-4
|
4
|
PIT
|
6.13
|
(3.25)
|
9.38
|
12-4
|
13-3
|
1
|
SD
|
1.81
|
(1.75)
|
3.56
|
8-8
|
10-6
|
2
|
SF
|
9.44
|
7.00
|
2.44
|
13-3
|
10-6
|
(3)
|
SEA
|
0.38
|
2.00
|
(1.63)
|
7-9
|
7-9
| |
STL
|
(13.38)
|
(1.25)
|
(12.13)
|
2-14
|
2-14
| |
TB
|
(12.94)
|
(4.00)
|
(8.94)
|
4-12
|
4-12
| |
TEN
|
0.50
|
0.25
|
0.25
|
9-7
|
8-8
|
(1)
|
WAS
|
(4.94)
|
(3.75)
|
(1.19)
|
5-11
|
7-9
|
2
|
Team’s that are likely to see a drop in their 2012 record with a brief point regarding each:
- DEN (8): the Broncos are likely to be an outlier as far as this analysis goes based on the signing of Peyton Manning, one of the best QB’s of all time. While it is true almost all teams change year to year, with a lot of big free agent moves, this is a special one as we rarely see a QB of Peyton’s caliber changing teams. So we are likely to ignore this projection
- GB (15): unless they go 16-0 this one will at least stay even, with it almost a certain lock to drop and support our analysis especially considering they were 4 games higher than projected record w/o turnover impact
- KC (7): I know many are bullish on the Chiefs this season, especially if they get all their injured players back and healthy for 2012. But I will agree with this analysis and project they win 6 or less in 2012, again especially considering they were projected to be a 4-12 squad without the favorable impact of turnovers last season
- NE (13): their schedule will likely be tougher in 2012, and after finishing 3rd in TO Margin for 2011, I expect NE will win 12 games max this season
- NYG (9): the Giants under Coughlin have been the antithesis of up and down one year to the next. In a rapidly improving NFC East, I expect the Giants to be at best 9-7 this season, quite possibly .500 or worse
- OAK (8): with Carson Palmer not looking so hot last time we saw him, and Darren McFadden seemingly still banged up, along with a tough slate, the Raiders are likely to finish under .500 this season
- SF (13): nobody benefitted from turnovers more in 2011 than the Niners – and that will not continue this season, especially considering they will not sneak up on anyone this year. I think even Niners fans would admit another 13 win season is not likely
To summarize the likely decliners, just like heading into last season, we had 7 teams this analysis projected were in for a drop. However, one interesting angle is last year all 7 were double digit win teams in 2010 – this time only 3 of the 7 won double digit games – and obviously it is more likely a team will drop 1+ wins when they won double digits vs. winning around 8. But nevertheless, I have backtested this for additional season’s beyond just last year, and it is a solid gauge of projected record.
Now let’s focus on the team’s that should see at least one more win than last year:
- IND (2): hard to go down from here, but with Andrew Luck in the huddle, and everyone healthier, the Colts should be able to win 3+ games in 2012
- MIA (6): the Fish actually outscored their opponents last season yet only wound up 6-10 – removing turnover impact they were projected to be a 10-6 squad. That record is about 2.5 games short of where their points (+1.00 per game, column A) suggested they should be – so with the coaching change, and some FA signings, MIA can certainly creep towards .500 this year
- MIN (3): now that Ponder has a year under his belt he should only improve, that is, if you buy into him being an NFL QB which I am not sure I do. There really isn’t anywhere down to go here, but we could see MIN be right around 3 wins again next year.
- PHI (8): a lot of things went wrong for the “Dream Team” in 2011, but a lot of it statistically speaking, should revert in their favor once 2012 gets going. I am certainly bullish on the Eagles to potentially reach double digit wins this season, especially taking into account they played like a 12-4 team last season w/o negative impact of turnovers
- SD (8): the Chargers could be on their way down as they have failed to win double digit games in 3 of the last 4 seasons. With Peyton Manning joining the division they are unlikely to be the preseason favorites for the first time in quite some time. SD could possibly approach 9/10 wins, but the AFC West is tougher than it’s been in 5+ years. I could see them at flat to +1 game.
- WAS (5): RG3 brings his talent to DC, looking to revive a Redskins franchise that has struggled mightily for almost 20 years now. Shanahan knows a thing or two about QBs which will help the offense, but does the defense have enough in a strong division. I will support this analysis and believe WAS will win 6+ in 2012.
In summary, heading into 2011 there were 13 teams that were projected to slide up or down in wins based on turnover impact – 11 moved the way projected, 2 stayed the same, and only 1 moved opposite, SD, whom was projected to go up from 9 wins in 2010 but only got to 8 wins in 2011.
Just like the number of likely decliners matched 2010 to 2011, there are also 6 teams projected to improve in the coming season, which is same as last year. Look for these teams to get more of the “breaks” in 2012, and parlay those into a better record.
That will conclude Part I of our Turnover Margin analysis. Trust me when I say I have a lot more great analysis related to turnovers that will be discussed in the coming week’s right here on my blog.
Thanks for reading, just a few weeks till preseason football gets going!
COPYRIGHT: THE SPORTSBOSS, 2012
No comments:
Post a Comment