SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.
Also, our existing PFR blog rss feed will be redirected to the new site's feed.
Pro-Football-Reference.com ยป Sports Reference
For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.
ShareThisShutdown defenses II
Yesterday's post was about how teams did against their opponents' best and second-best wide receivers.
From a fantasy football standpoint, this could be worthwhile information to have. If you can't decide between, say, Steve Smith and Chad Johnson, you might first look at their 2006 slate of opponents and see if one is expected to be playing against tougher pass defenses than the other. If you wanted to dig deeper, you could check to see if one of them is expected to play against defenses that were tougher against wide receivers specifically. Those ideas have been around for a long time. But as far as I know it's not common practice to take it a step further and examine whether Smith or Johnson figures to be playing against a slate of defenses that will be tougher against #1 receivers specifically.
The table at the end of yesterday's post shows enormous variation in how teams fared last year against WR1s and WR2s. Some teams, like the Bears, shut down WR2s while being eaten alive by WR1s. The Redskins, on the other hand, actually allowed more production (in terms of raw numbers) against opposing WR2s than the corresponding WR1s. If these tendencies are caused by personel --- like the mythical shutdown corner --- or by defensive scheme, then we would expect them to persist from year to year. If that's the case, then we've got ourselves a valuable bit of fantasy football information.
But it's not and we don't.
I gathered six years of this data and checked the year-to-year correlation. The correlation coefficient is .10 and is not significantly different from zero. If you don't know what that means, it means roughly this: there is not sufficient evidence to conclude that a team's 2005 "DIFF" will be related in any way to it's 2006 "DIFF." (Yeah, yeah, I know, it merely means it's not related in any linear way.) I guess it's possible that there is some year-to-year carryover among teams that maintain the same coaching staff and mostly the same group of players in the secondary, but that that carryover is being diluted by less stable teams to the point where we can't see it in the data. More likely, in my mind, is that random variation accounts for the differences we saw among teams' relative performances against WR1s and WR2s and random variation will also account for the differences we see in 2006.
This entry was posted on Thursday, June 22nd, 2006 at 4:46 am and is filed under Fantasy. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
Perhaps look at the correlation coeff. between 2005 1st 8 games DIFF and last 8 games DIFF. If it really is random, this should be low as well. If it is not, this could still be useful from a fantasy perspective, even if year to year there is little carry over.
Also, what were the correlation coeff. in 2005 for the WR1 Fpts given up vs average and team wins, and WR2 Fpts given up vs average and team wins? For example, all other things being equal, does it matter if you start a WR2 vs a good team instead of a bad team, or is it just team-specific?
noisy td's
I'm not a statistical guy at all, but is it possible that the variation is due (at least in part) to variation in schedule? From year to year, a team is likely to play 10 different games from the previous year, and 6 different teams (if I'm remembering the schedule correctly). I don't know if you can correct for that, but it seems to me that if you face the NFC East one year and the NFC West another, you are facing very different qualities of offense, particularly when it comes to passing.
wrmjr, that ought to be accounted for because I'm not looking at raw production vs. WR1s and WR2s. I'm adjusting for the quality of WR1s and WR2s that the team actually faced each year.
monkeytime, from a fantasy football standpoint, we want the TDs in there, noisy or not. Don't we?
There may be something real there that's being drowned out by the noisy TD data. Or they may be nothing there. As a fantasy football player, I don't see why you'd care. The end result is the same: the 2005 data (alone) doesn't help you project the 2006 data.
Or are you saying that maybe the Year N-1 yardage data would help predict the Year N overall data. Is that the claim?
A football is a collection of random events, right? I am much less interested in the number of "X"WR touchdowns allowed vs the number of "opportunity to score touchdowns". If we were talking about 30-40 "X"WR td per season, thats one thing, but we are talking about much fewer, 1-10 in most cases. This goes back to the whole arguement about ESPN's Hector and Victor, Fantasy Sharks, or any of the other ridiculus services that "assign" touch downs to projected weeks. Just a made up example: Going into the game the Chiefs expect 0.75 1WR touchdowns. Well, that week then had three 1WR touchdowns. What does that mean to its indicators? Look it as average rainfall. Once you've gotten the 10 inch deluge, it doesn't make the rest of the month any less likely to experience average rainfall. Am I making ANY SENSE HERE? That is why I think this should be studied on a reception basis, not a TD basis, then facter a TD per reception value to the expected receptions after the fact. Obviously, its never that simple - but i argue (haven't proved yet) that it would provide a better predictor, if a predictor at all. So, if TO gets 10 recepts and no touchdowns, and Pinkton gets 3 recepts and a TD, I am not going to start Pinkston next time they are looking at those predicted splits. If I throw the TD in there, the numbers might tell me too. Take the TD out and factor its probability in after the fact. I am talking about wr's here - consider defenese the "bizarro wr" and treat the defense as if it was a wr. what is the recepts given to 1wr, 2wr, 3wr. apply those recepts to the wr's probability to score td. I think saying what is the probability of td recepts given to 1wr, 2wr, 3wr is just plain noisy.
yardage production could be used rather than recepts too. A team that gives up 300 yards passing gives up more "opportunities to score touch downs" than a team that gives up 150 yards passing, theoretically.
which begs the question: What is the correlation between yards allowed and touchdowns and recepts allowed and touch downs. Which is a stronger predictor of potential tds. use team wide stats. we are getting too caught up right now between whether its wr1 or wr2 that scores the td. the wropen scores the touchdown. then lets see going into a game which wrx is more likely to be wropen. i am arguing that granting wrx "y" td's clouds the issue. lets simply grant him yards or recepts and then apply his own probability of td times his yards or recept variable.
An interesting idea, monkeytime. Basically, I think you're saying that a WR's TDs are based on two things: 1. His opportunities to score TDs (receptions), and 2. His normal likeliehood of TDs, based upon his ability and his role with the team (for example, some WRs are their team's go to guys in the Red Zone, while others are not).
Could we take it one step further and say that their opportunities are the number of passes likely to be thrown their way, then take into account their catch % to figure out how many they should normally catch, then multiply by their TD % to figure out how many TDs they are likely to get?
As long as catch % is relatively consistent (which I am not sure about) from year to year, I think you could. Let me look at Football Outsider's numbers and see if it is.
I do not think that the idea of looking at how defenses fare against 1WR, 2WR, etc. should be discarded because of the small annual correlation coeffecients.
I agree with JKL. Perhaps we can take the idea further and look at the number of games it takes a defense to "settle in" on an annual basis and take meaningful measurements of their ability to shut down receivers.
Just my two cents...
Just pick talented players with cool names and you'll be all set.
I think biggest thing "specialist" use to predict numbers is match ups in man and end zone coverage. You got a tall guy that can jump going against a bunch of short quick guys he is likely to get endzone targets. So you try to think they'll target the guy x times in the end zone and he should catach atleast X. You also gotta look at other matchups to figure out odds of you guy getting looks. Also fast wide reciever vs slow corners gives you a good match up.