Roster Maximization: The Case for Tight End Streaming

Pat Thorman focuses on the strategy of tight end streaming and uncovers more supporting evidence for its usage.

| 2 years ago
Bennett

Roster Maximization: The Case for Tight End Streaming


In the never-ending quest to gain any edge over their opponents, fantasy owners have developed an affinity for the practice of “streaming” seemingly inferior options into their active lineups based on advantageous matchups. While this takes a bit more thought than blindly riding with your No. 1 option on a weekly basis, if done correctly the results tend to be favorable more often than not.

Streaming is not a new idea by any stretch – as fantasy baseball players, who have long utilized this method to squeeze every drop of value out of their pitching staffs, will attest to. Nor is it a revolutionary technique in fantasy football. Chances are many leagues in which you participated last season were won by an owner who streamed his defenses on the way to the title.

An increasing number of owners take this strategy well beyond merely inserting a widely available, mediocre fantasy defense into their lineups when Mark Sanchez or Matt Cassel happens to be their next opponent. As C.D. Carter delves into over at TheFakeFootball.com, there is a strong case to be made for streaming tight ends who are not part of the uber-exclusive elite.

Carter illustrates the futility of attempting to ride a single tight end all season long if he does not qualify as among the best of the best at his position. He also demonstrates how the highest tight end point totals are delivered by such a wide array of players as to make it foolish to expect consistent output from the non-elite. As we will demonstrate later on, he is right on the money.

The main issue for fantasy owners, in both redraft and dynasty leagues, is when to draft your tight end – or, more accurately, tight ends. The widely held elite at the position consist of the top five – Rob Gronkowski, Jimmy Graham, Tony Gonzalez, Aaron Hernandez and Jason Witten. However, if we look at last year’s average draft positions (courtesy of MyFantasyLeague.com), Antonio Gates was the third tight end off the board, and Vernon Davis was the fifth. Witten did not go until seventh (in part due to a preseason spleen injury) and Gonzalez was the ninth tight end chosen. It is tough to project if we will have appreciably changed our views of the elite by this time next year, or even sooner than that.

If you selected Gates, Hernandez or Davis last season, it would likely have cost you a pick with an ADP in the 50s, based on MFL’s average pick data. Other options that went in that range at the ultra-thin running back position included Doug Martin (54th), Reggie Bush (63rd), Willis McGahee (66th) and BenJarvus Green-Ellis (69). Hindsight is always 20-20, but if you were of a mind to stream tight ends based on matchups and ignore them during the meat of the draft, you would have been better off than if you had chosen one of the aforementioned pass-catchers.

Out of the top 25 scoring tight ends, 18 of them averaged more points scored against defenses that were in the bottom half of the league in defending fantasy tight ends. Oddly, only five of the top 10 tight ends averaged more fantasy points against bottom half defenses. However, from the 11th tight end through the 25th, just two averaged more points against top-half tight end defenses than they did against the bottom-half teams.

 

TE Group (by Point Total) PPG against top ½ TE defenses PPG vs. bottom ½ TE defenses
TE 1 – TE 10 8.16 average PPG 8.81 average PPG
TE 11 – TE 20 4.69 average PPG 7.41 average PPG
TE 21 – TE 25 4.15 average PPG 5.39 average PPG

*- Week 17 fantasy totals are not included

 

A quick look at each group begins to tell you just how advantageous it would have been if you had streamed tight ends from the middle group against bottom-half tight end defenses. Granted, it will take a few weeks to clearly identify those weaker units – but the especially poor ones should be somewhat obvious right away.

These numbers, and the ones below, help to illustrate a clear opportunity for fantasy owners to bypass the uncertainty of the non-elite tight ends ranked in the top 10 on draft day, in favor of building crucial depth at more shallow positions – namely, at running back. Why bother throwing away a valuable pick on the eighth tight end on your board if you can wait another seven rounds to pick up two players that could combine to reasonably approximate TE1 numbers?

The top 10 tight ends averaged 8.18 points per game last year. The middle group averaged 7.41 points per game against bottom-half tight end defenses, which is over 90 percent of the top tier’s average (90.6 percent). In many leagues, a handful of players from that middle group were available, at various points, on the free agent list.

Tight ends in that middle group also saw their point totals fluctuate the most, based on the caliber of defense they were facing. The 2.72 per-game point difference between when they played top- or bottom-half tight end defenses was by far the most of any of the three groups we looked at – over 45 percent of their per-game average.

 

TE Rank (Point Total) Average Points/Game Vs Bot ½  vs Top ½ Percent Difference
TE 1 – TE 10 8.18 PPG + 0.65 PPG +7.9%
TE 11 – TE 20 5.93 PPG +2.72 PPG +45.9%
TE 21 – TE 25 4.82 PPG +1.24 PPG +25.7%

*- Week 17 fantasy totals are not included

 

It is very tempting to treat that piece of data like sausage, and just enjoy it without asking too many questions about what goes into it – and risk being inaccurate in attempting an explanation. However, one thought that comes to mind would be their relative importance to game planning.

From the first group, Gronkowski, Witten, Dennis Pitta, Owen Daniels and Kyle Rudolph all averaged more points on a per-game basis against the tougher tight end defenses. Perhaps this is indicative of the fact that these players are so vital to their respective offenses that they tend to play prominent roles in the game plan on a weekly basis, regardless of the opponent.

Tight ends who play more of a bit part in their offenses are more apt to show up when the conditions are favorable. Martellus Bennett, for instance, saw 61 out of his 84 targets (72.6 percent) when facing bottom-half tight end defenses (60 percent of opponents, not counting Week 17 games). He was targeted 6.8 times per game against easier competition, as opposed to seeing only 3.8 passes come his way against top-half tight end defenses. Brandon Pettigrew was targeted 64 times against easier tight end defenses (10.7 per game), and 35 times against the tougher half (5.8 per game), in an equal amount of games with each (six). Bennett averaged 5.9 more points per game against easier competition (8.4 PPG vs. 2.5), and Pettegrew did 2.6 better (7.3 vs. 4.7).

If nothing else, we can use this information to help to project which teams’ offenses tend to take advantage of matchups – and use it to see which tight ends to target for streaming. Bennett is no longer in New York with Kevin Gilbride calling the shots, but Brandon Myers is a strong pass-catching tight end who should capably fill that void. Pettigrew is not sexy, but with a points-per-game average of 7.3 when up against easier tight end defenses, and an ADP of 123 (and falling) in early 2013 MFL mock drafts – he could represent an interesting streaming candidate. Norv Turner is taking his tight-end-friendly offense to Cleveland, and it should mean good things for popular sleeper option Jordan Cameron. Last season under Turner in San Diego, Gates saw 6.9 targets against easier tight end defenses, 3.8 per game against stiffer competition, and scored 1.6 points per game more during the former.

As draft season approaches we will attempt to dive a little deeper into the question of streaming at various positions. Although it is unconfirmed, the expectation is that the tight end spot will be the most consistently advantageous opportunity to utilize this strategy.

 

Player Averge PPG Avg vs Top ½ Ds Avg vs Bot ½ Ds PPG Difference
D. Pitta 7.26 3.77 11.24 +7.47
A. Hernandez 9.30 7.04 13.1 +6.06
M. Bennett 6.07 2.53 8.43 +5.90
J. Graham 9.59 6.28 12.1 +5.82
M. Lewis 4.51 2.41 7.26 +4.85
H. Miller 8.64 6.53 11.0 +4.47
B. Watson 4.08 2.37 6.65 +4.28
S. Chandler 6.14 4.66 8.43 +3.70
J. Gresham 6.85 5.92 9.40 +3.48
J. Cook 5.83 4.68 7.70 +3.02
B. Pettigrew 5.97 4.68 7.25 +2.57
D. Clark 4.50 2.87 5.09 +2.22
L. Kendricks 4.71 3.75 6.15 +2.14
A. Gates 6.31 5.40 7.00 +1.60
D. Allen 4.84 2.37 6.65 +1.53
J. Finley 4.73 4.01 5.36 +1.35
G. Olsen 7.33 6.68 7.66 +0.98
T. Gonzalez 9.13 8.70 9.34 +0.64
B. Myers 6.62 7.01 6.17 -0.84
V. Davis 6.02 6.50 5.54 -0.96
O. Daniels 7.49 8.21 6.53 -1.68
K. Rudolph 6.75 7.77 5.86 -1.91
B. Celek 4.97 7.68 3.47 -4.20
J. Witten 7.35 9.32 5.10 -4.22
R. Gronkowski 13.4 19.2 12.1 -7.10

 

Follow Pat on Twitter: @Pat_Thorman … and our main feed: @PFF_Fantasy

  • http://twitter.com/FFMagicMan Micah James

    I can understand the concept, and see where the numbers bear it out. I have a very hard time letting my TE position depend on winning my waiver wire week to week. Streaming DEF and streaming K allow you to pick from a relatively-equal waiver pool of 6-8 at each position every week. Hinging my success on 1-2 TEs week-in and week-out? I’ll let someone else go that route.

    • Seth Gordon

      I agree. The strategy is nice in theory, but impractical, especially in dynasty.

  • Wintermute

    I like the article but would caution that the strategy may not work in larger leagues. If there are 10 to 12 teams then I would expect 14 to 16 TEs to be rostered leaving plenty of supply to enact the strategy. But with 14 and especially 16-team leagues the demand would outstrip supply as I would estimate 21 to 25 TEs to be on a roster. The larger the league the more self-reinforcing cycle of demand reducing supply creating yet more demand would drain the available TEs from the market.

    Roster size would similarly reduce TEs as the roster size limits were expanded. A 12-team league with 15 roster spots would allow for a streaming-TE strategy. A 16-team league with 25 roster spots would not.

    Thanks for writing the article, it was a good one. Good idea. But I would just like to suggest that it may be helpful to discuss when such a strategy could be implemented once you consider the supply-and-demand dynamics of league and roster-size.

  • Pat Thorman

    Those are fair criticisms/concerns – especially the question of how well it would work in dynasty leagues, depending on how the rules are set up.

    I am not advocating a strict waiver-wire only strategy. That is one route, especially in more shallow leagues – as Wintermute mentioned. Drafting one or two TEs, well beyond when the rest of your league picks their starters, would work as well (even in deeper leagues). Even in a 16 team league, it is unlikely that most, if not all, of your league mates will be selecting their backup TEs until very late. That still leaves an opportunity to bypass the initial TE runs and grab one or two in the TE15-20 range, several rounds after when you’d normally do so, but before the backup TEs dry up.

    I should have been more clear about that, and again – that this is not a solely waiver-wire streaming strategy, in the D/K mold that’s commonly used.

    I understand that this is a fairly uncomfortable method to use, and it may not work in all leagues. If nothing else, it is well worth considering – especially if you get caught at the end of a vicious TE run and are not excited by the scraps that remain (and want to deepen your RB corps.

    Thank you for reading/commenting.
    -Pat

  • Scott@Seattle

    I streamed QB’s and it worked fine. One team i drafted RG3(round 7) so that was fine, another i went with Fitzpatrick/Palmer until i grabbed Kaepernick from waivers. I really wouldnt draft anything other than RB in the first 3 rounds, and only Flex until round 6. I also nabbed Wilson in the later rounds, he was money in the fantasy playoffs.

  • Ryan

    Have you done this with QBs?

  • Pat

    I just did. The post went up last night.

  • trevor maltbie

    Does anyone else notice that Gronkowski absolutely DOMINATES the best defenses? And does just as well vs bottom half as Graham does. Holy!