FOF 6.1 Sandbox: TE Value
 

So, my first real attempt to provide something of value as far as FOF research.

I'm going to do my best to have sound logic and testing here, but this is my first attempt so let me know if there's something I should do differently.

This all came about because of something that Quiksand mentioned in passing. He was musing about what value a Stud TE really has to a team. I realized I had a SP team that was built towards having a VG QB, Stud WR, VG WR #2, and a Stud TE. I decided to take this team and try to determine how much value (namely in the passing game) a Stud TE has.

So, my attempt will follow.

TEST DATA

I will be simming 10 seasons of roster set. I will plug the data into the QuikTest Template (well, a version I am modifying for him). There will be 6 different roster sets:

Stud WR, VG #2 WR, Stud TE
Stud WR, Stud TE
VG WR, Stud TE
Stud TE
Stud WR, VG #2 WR
VG WR

So, I will cut some combination of these three players.

GAMEPLAN
Based on SkyDog's Sandbox on the short passing game, it appears that the "sweet spot" for TE's is the 5-8 yards and probably the 9-12 yards settings. So this gameplan will lean towards those kinds of passing, especially in short passes:

Short Pass: 0/0/100
Long Pass: 35/30/20/10/5

Pass Blocking: RB - 50, FB - 50, TE - 0

This offense is geared relatively high run and somewhat weighted to longer passing. The idea is I want to get geared towards the TE, but not so imbalanced that it becomes ineffective.

The Formation Use is Recommend and I'm not changing it between rosters (I actually don't think it changes that much anyways, if at all)

I will not change the gameplan at ALL between rosters.

ROSTER


This is the starting Roster, Paz, Clark, and Young are the guys that I will be toying with moving around, etc. I realize Paz and Young aren't fully developed - but we'll make it work.

Stud WR, VG #2 WR, Stud TE

Team Averages:

Player Averages:

Stud WR, Stud TE

Team Averages:

Player Averages:

VG WR, Stud TE

Team Averages:

Player Averages:

Stud TE

Team Averages:

Player Averages:

Stud WR, VG #2 WR

Team Averages:

Player Averages:

VG WR

Team Averages:

Player Averages:

Placeholder 3.

Placeholder 4.

Ok.. I'm going cross-eyed... so, I'll try to make what I can of this later... but anyone that has a desire to jump in with analysis - please do!

To me, the most interesting differential to see (which I don't think you've gotten to yet) is that between:

Stud WR, VG #2 WR, Stud TE
and
Stud WR, VG #2 WR

On a team that has standout talent at wide receiver, in real football terms, it seems like adding yet another serious weapon to the underneath passing game would be a major boost. Whether the defense is forced to leave the quality receiving TE relatively uncovered where he can be very productive underneath the deeper coverages, or whether they rotate coverages toward him and free up the outside receivers to be more effective -- you'd have to expect that it would help the team overall.

So, I guess once you post the set of numbers with the good receivers but without the good TE, we might have some sense of how that plays out in this scenario.

I will post here a compressed version of my findings of my own test team -- where I took a team that had pretty mediocre talent in both the WR corps and the receivers out of the backfield, and switched from basically a blocker-only at TE1 to a pretty solid quality guy overall. I found those results to be intriguing.

Actually, I posted that I just missed editing the label of my last two posts. Let me fix those.

Ok, Updated... the 5th set of data is the one you want to look at..

I should have sent myself the spreadsheet at work so that it's easier to compare side by side, but I had a rushed morning.

That being said, it appears to me that the team and QB performed marginally WORSE with the Stud TE added to the Stud WR and VG #2 WR - which as you said, does not seem logical in football terms.

For easier comparison:

With Stud TE : 11.0752 wins, 99.7 QB Rating, 31.8 Passing TD's, 25.48 PPG
Without Stud TE: 11.7744 wins, 100.0 QB Rating, 33.9 Passing TD's, 25.17 PPG

I do have to wonder whether part of the issue here is not tweaking the gameplan to the talent, but I was trying to reduce the number of variables...

For what it's worth (related note, but not about wade's test) here is a snapshot of what happened with my other team, in my playtesting dynasty thread:

- - -

I was running a FB and TE-based offnese, with a lot of short passing, and a lot fo the longer passing targeted to the 9-12 range, where the TE is a major player. I ran the exact same team, exact same gameplan, with two different TEs slotted as the TE1 with 75% playing time. Here are the two guys we used:



Superficially, we saw a HUGE jump in production when we moved from the blocker-only guy to the all-around TE... the TE1 position went from an average of around 20 targets and 150 yards receiving to one of the top targets on the team -- averaging maybe 65 targets and 450 yards receiving.

However, the team results reveal that *absolutely nothing* happened to the team's overall effectiveness. Here's the summary of the two cycles, each with ten no-injury seasons:



Bottom line conclusion pretty much has to be that whatever we gained on that team from adding a pretty decent TE was cannibaliized 100 cents on the dollar from the other players on the team -- we didn't actually make our team better by improving the talent at a key position. The before and after stats, in every single area, are nearly identical, and without a doubt are not statistically separable.

I'm open the the general notion that ten trials just isn't enough to weed out some slight difference that truly does exist here -- but if that's the case, then you certainly couldn't feel comfortable over the course of, say, one normal multi-player season, feeling that the addition of one key player at TE actually makes a difference in the team's success.

Yeah, I think your tests bear about the same data as mine. The only real thing I take out of all of this is it seems that a Stud TE does do about the same for you as a VG #2 WR, but outside of that if you merely ADD a Stud TE to an already good roster, it seems to do nothing or potentially even hurt.

From your tests and my own experience, i must agree. You either have 2 good WR's or a good WR and good TE, but having all them doesn't make a huge difference.

In FOF, i see the WR's and TE just as possible pass targets. Given a set number of passes to throw, not for having more targets you will do better, you will just spread the receiving yards between those players.

Having more targets could work in real football, opening your passing game a lot more, creating mismatches etc, but in the FOF engine, i doubt that mismatches exist, or that X player is really covering Y player. I think FOF works more with overall formulas than with individual ones (of course i could be wrong, that is just my feeling not knowing how the engine really works).

Another feeling i have is that the TE's are just extra receivers in FOF, not having a lot to say in the Oline performance.

Unfortunately the Team Summary screen does not show Blocking numbers, but your statement would be supported by the fact that the team average is basically consistently at the same YPC.

Okay, while I agree that this may indeed be the "right" way to think about FOF -- I'm still not sure how we are to square that with what we see here.

If the engine for resolving passing plays relies on some sort of overall team-wide or unit-wide calculations, then presumably the success of the offense would depend on some amalgamated level of talent among the receiving targets in the game for that play, right? So, if that's true, then we should expect to see the *addition* of a high quality receiving target make that aggregate rating of the receiving group higher -- and it should cause the whole unit to perform better. So -- we ought to expect to see the whole offense get a bit better by adding a high quality TE, even to a team that features high quality receivers already (I'm thinking about Dallas Clark and the Colts as an easy example).

If this doesn't happen... and indeed a quality receiving corps is *harmed* by adding another quality player at TE, then I guess we need to modify that approach somewhat to explain what we think we're seeing here.

Is there any reason to believe that "blocking numbers" would give us additional insight into how well the team is rushing the ball? I think the "effect-and-cause" relationship there has been established pretty clearly.

Certain positions average less yards per throw in the receiving game regardless of talent. A great receiving FB will still average less yards per throw than even poor receivers, the reason why it is usually more effective to use a low route running fullback over a great fullback. You want those targets going to players who will be the most effective with them.

While a TE may be better than another as far as ratings, if he takes more targets away from the wide receivers he may actually lower your team's production.

That's not to say you don't need good players at those positions. If you use an unusually high amount of any particular passing distance you'll want enough options to keep the defense from being familiar with the plays. If the playcalls overly favor a specific receiver there won't be enough variety in the playcalling to keep the defense from being used to those plays.

If your formations are varied enough to keep the defense from being familiar with your playcalls you don't need as many options as you would if you only use a few formations. If you're going to use a lot of any formation you need, at least, a 3rd option whether WR2/TE or RB (or a 3rd-5th WR in a few formations). If you're going to use only a handful you might need a 4th passing option, and if your quarterback knows almost none and is dumber than a brick you may even benefit in production from a receiving fullback.


Of course it is better to have a player with better receiving skills outside of route running, as you want players that can do as much with each target as possible. You don't want guys that will not do very much with the ball, whether because of their position or poor receiving skills, who have high route running. They'll get more targets and be less effective with them than other players.

Good point.

jkat is definitely Jim. :)

Yeah seems pretty obvious. :)

I think the effect of a great TE depends greatly upon what sort of offense you are running. A great TE is perfect for a heavy run offense with some short passes on 3rd down to get the first (say 3rd and 4 or 5 yards). But, if you are not running that sort of offense, the TE is much less effective than having another stud WR. However, you can probably get a stud TE cheaper than you can a VG WR.

you're good at compiling numbers, but you must not be good at analyzing those numbers. how is the above "marginally WORSE" in your opinion? ok, so without a stud TE you averaged 2 td's more. that's understandable, without a stud TE with the necessary route running skills to take away catches - the stud WR's were able to get targeted more. perhaps out of those extra targets they used big play ability to score on avg a few more big play td's.

.7 more wins without a stud TE isn't really earthshattering. and same amount of pts were scored per game. QB even performed about the same. so if anything one can infer that offenses without stud TE's one scores more often than with.

this is how i operate my team as well. i noticed when having a stud TE my WR's overall numbers suffered. i have since switched to seeking TE's with specific skills - good 3rd down catching ability, good big play ability, and good blocking. i don't want a TE with high route running. i pay my WR's signigicantly more than i pay my top TE - i want the majority of my receiving production to come from the guys i'm paying more. all stud TE's serve to do in FOF is steal catches away from your WR's. based on the skills i target in TE's, i just want a guy who gives my QB another solid option on 3rd downs, i want a guy who can maybe snag the occassional 20+ yard catch or two and overall i want him to stay behind and block and generally not get in the way of my WR's numbers.

Hold on a second, here...I don't think the Colts are an "easy" example. Prior to adding Dallas Clark, a first round pick, the Colts were already an explosive offense with Marcus Pollard at tight end...a guy who went undrafted and never played a snap of college football. Pollard put up some great numbers and if an undrafted guy can do that you'd think a first rounder would be able to absolutely explode in the same role but he hasn't...mind you, this year could prove to be the year.

Now, if Clark does continue to roll as he has this year, what do you think the end of the year numbers are going to look like for the Colts? Better? If so, why?

I think what jkat has hit on is that if the offensive scheme is going to dictate how many passes Manning throws in a game and if more of those passes are going to the tight end, then the WRs and RBs just aren't going to see as many balls. So, by the end of the year, maybe you'll see Harrison and Wayne with less than their usual stats while Clark had a career year...all while seeing PPG staying relatively the same.

So, what if Dungy were to go "Andy Reid" and pass, pass, pass? Would having a TE like Clark be more beneficial than Pollard and would you see the WRs number drop like they would in the previous example?

To really test the usefulness of a great TE and two good WRs, I would alter the gameplan to be a little more pass happy.

I don't think you can ever settle an argument using a real-life example, and I suppose I regret making any such mention. I don't think you can simply compare one year to another and say what exactly happened -- after all, the Colts had quite a bit of turnover last year to this year, wouldn't it be impossible to quantify that?

All I was trying to say was that it *seems* that Dallas Clark is a very viable receiving option, a guy who is pretty tough to cover. He seems like the type of tight end that would make *any* real football team harder to defend -- if you commit to stopping him, you'll leave the wideouts single covered more frequently, and are likely to get burned by them instead. My point wasn't so say "look at the Colts as an ironclad statistical example" but rather just that they seem like a team with quality talent at WR, and I can't imagine anyone *really* arguing that the team would be better off if, say, Dallas Clark broke his leg in practice and they were forced by cap reasons to trot out a no-talent blocker-only type of guy for the rest of this season.

What about getting a pass blocking guy and leaving the TE in to block?

Actually, who led the Colts' TEs in receptions and yardage last season? One hint...it wasn't Clark. It was Ben Utech...a blocking TE.

My point is that the gameplan is going to dictate the usefulness of a particular type of player. So, if you have more talented receivers (both TEs and WRs) I would think to maximize their potential you would need to give them more opportunities to perform...i.e. throw the ball more. So, again, a true test of the usefulness of a great TE paired with good WRs would be to alter the gameplan to try and take advantage of personnel.

mar·gin·al /ˈmɑrdʒənl/ Pronunciation Key - Show Spelled Pronunciation[mahr-juh-nl] Pronunciation Key - Show IPA Pronunciation
–adjective
1. pertaining to a margin.
2. situated on the border or edge.
3. at the outer or lower limits; minimal for requirements; almost insufficient: marginal subsistence; marginal ability.
4. written or printed in the margin of a page: a marginal note.
5. Sociology. marked by contact with disparate cultures, and acquiring some but not all the traits or values common to any one of them.
6. Economics.
a. selling goods at a price that just equals the additional cost of producing the last unit supplied.
b. of or pertaining to goods produced and marketed at margin: marginal profits.

I would be interested in seeing whether different kind of QBs use the tight end better. Common NFL perception is that young/inexperienced QBs benefit greatly from a stud TE in the lineup. I wonder if that might play out in FOF.

Stud TE vs. VG WR
 

Hi,I may be missing something obvious here however:
A stud WR + very good WR had an ave/att of 7.6 yds
A stud WR + a Stud TE had an ave/att of 7.6 yds.

I believe yards/att is the best statistic to measue the difference.
So, my take home between a very good WR and a Stud TE is take the cheaper, when both are available.

If it's the TE dont run many formations where he is off the field. ( This could be a problem with a low formation QB.)

Comments appreciated.
Merry:)

Well, this was two years ago - so I don't remember my exact thinking at the time.

I can see where you're coming from, but you need to look at more than just one stat.

That being said, I wouldn't assume that any of this is even close to accurate under 6.3a. Anecdotaly, with the increased usefulness of the short passing game, a stud TE seems to be even more useful than in the past.

Was Skydog's Short Passing Offense ever made available for download?

would love to see a 6.3a version of this