Strength of schedule / points vs. wins
Posted Jun. 17 at 08:12 AM
I get e-mails from time to time from readers wanting to see strength of schedule information not for wins and losses but for points scored – or for rushing or production allowed. The reasoning is sound enough: Fantasy leagues aren’t based on wins and losses – they’re based on yards and touchdowns. So why not look at schedules in those terms.
As part of an extensive study on this topic last year, however, I spent hours examining all of the strength of schedule data since 1993 – both expected and actual. And I blended that information with how teams actually performed. And I found that it makes far more sense to look at schedules in terms of wins and losses (even if you’re more interested in points and yards).
In the feature article on Strength of Schedule in the 2008 edition of Fantasy Football Index, I outlined this point in a three-paragraph sidebar, summarizing a few of the findings. I didn’t want to throw out too many numbers, however, in fear of confusing/alienating readers. But with this still being an important topic, I’ll offer a more detailed explanation below.
Conclusion: For fantasy purposes, it’s more use to look at wins/losses in strength of schedule, rather than potentially weak defenses.
Those who are statisfied with that conclusion are excused for the rest of the day. Those who want to see the supporting data can plow on.
THE 20-WIN RULE
Consider teams with 20-game gaps in their strength of schedule. That is, their opponents in the previous combined to post a record equal to or worst than 118-138, or those same opponents went 138-118 or better. There are the teams that tend to rank in about the top 5 (or bottom 5) in strength of schedule. That’s when you tend to see a reliable swing of about 10 percent in offensive production.
Since 2002, 31 teams (about 4-5 per year) have entered the season with schedules of 118-138 or easier. Those teams combined to score 1054 more points than they had the previous year, an average of 34.0 per team. So that’s maybe 4 more touchdowns and 2 more field goals per team. Maybe the quarterback throws 2-3 more TDs and those key offensive guys score 1-2 more TDs. (I never said strength of schedule was a silver bullet – we’re just talking about picking up a little extra production on the margins). Of the 31 teams, 24 scored more points and 7 scored fewer points.
Compare that to 31 teams that since 2002 have projected to face the softest defenses. I set the level at an average of 22.3 points (or more allowed by opponents), so that there would be exactly 31 teams in the comparison group as well. Those 31 teams scored only 433 more points – an average of 14.0 per team. Of these 31 teams, 19 scored more points, 11 scored fewer points and one scored the exact same amount.
So in that comparision, the strength of schedule numbers based on wins and losses was more meaning – by an average of 20 points per team (per season).
As an aside, I’ll point out that I expected those teams playing schedules clocking in short of 118-138 to be even better. Collectively, those 31 teams won only 6 more games. Of those 31 teams, 14 actually finished with lesser records. If you look at the same types of schedules from 1993 to 2001, you only 8 of those 35 teams finishing with worst win-loss records; collectively, that group combined to win 61 more games. So on average, I expect an easy schedule to translate into an extra win for those teams. If you go off the defensive numbers (for 2002-2008), those 31 teams combined won only one more game.
Let me know turn the spotlight on the other end of the spectrum – teams with hard schedules.
From 2002-2008, 26 teams played schedules clocking out at 138-118 or harder. Those 26 teams combined to win 29.5 fewer games and score 734 fewer points – an average of 28.2 fewer points per team (per season). Of those 26 teams, 18 scored fewer points in the year they had the hard schedule. Only 8 of those 26 teams won more games.
Compare that to the 26 teams playing hard schedules in terms of points. (Here, we’re looking at teams playing opponents that on average allowed fewer than 19.9 per game – I again picked a seemingly weird number so that number of teams would match up exactly). The 26 teams in this group, despite playing seemingly tough schedules, actually scored 13.8 more points on average. They combined to win 12 fewer games. Again, far less severe statistical shifts than what we saw with the wins and losses chart.
Note, by the way, that if you base your strength of schedule decisions on defenses, there is essentially no difference between “easy” and “hard” schedules. Both of those groups need to score about 14 more points (per team) per season). And only a half win per team difference between those schedules as well – not what you want playing a big difference in your fantasy preparations.
Compare, those the differences following that 20-win rule. Teams with schedules south of 118-138 scored 34 more points, versus teams on the opposite end of the wins-losses charts at 138-118. So there, you’re looking at a difference of over 60 points per team. And the difference in wins is over one per team. That’s change you can believe in.
SCHEDULE: LOOKING AT 2009
Based on the findings above, I suggest that when considering the impact of scheduling, you focus on teams outside that 20-win gap (and the far removed from the 20-win mark, the best).
Teams that have a good chance of seeing their offensive production improve by about 10 percent (beyond any improvement caused by offseason injuries and player movement):
109-146-1 Green Bay
113-142-1 San Francisco
Teams that have a good chance of seeing their offensive production decline by about 10 percent (beyond any change caused by other offseason developments). And note here that the first four teams – Miami, New England, Carolina and Atlanta – play not only hard schedules but notably hard schedules (in comparision with those of the last 17 years).:
151-105-0 New England
148-107-1 Tampa Bay
145-110-1 NY Jets
142-113-1 New Orleans
Posted by Travis Billman | Jun. 18 at 01:31 AM
With the understanding that Pittsburgh had one of the hardest schedules last year, and this year has one of the easiest, I really like Pittsburgh's players to have a bounceback year and be pretty good sleepers. Theoretically, you could add approximately 20% to last years numbers (with the assumption that there was a 10% drop from expectations due to scheduling), right? Oh, and as an aside. . .change your picture Ian! You get to study fantasy sports as an occupation! You should look happier than your pic would indicate. :-)
Posted by Duane Stay | Jun. 19 at 06:39 AM
Go to facebook and you'll see his true identity.
Add a Comment
Already a registered user? Please sign in to add comments.
To add comments, you must become a registered user of our site. To register, please click here.