By: Nick Zylak
A complete record of success by year since 2015 can be found by clicking here
Quick Intro:
I highly recommend everyone reads the running back draft strategy article first. Or at the very least reads the introduction to that article. It goes over how I collected the data for this study, and what exactly the data is. It’s not the end of the world if you just skip to the findings, but I do think it’s important to understand the limitations of the underlying data. Also I think this study is really cool, and I’m sure many of you will just be interested in how it was done. Below I have a quick summary of the data if you’re on a time crunch.
Where Did the Data Come From TL:DR
- I spent 4 months scraping ESPN public leagues
- I removed:
- Keeper leagues
- In-Person drafts
- 2 QB leagues
- TE Premium Leagues
- Leagues with impossible value picks (unreal steals and reaches)
- Leagues with clear draft pick trading
- The final dataset includes close to a million teams
- I used a random sample of 110,000 teams from each year between 2015-2021 (770,000 teams)
When to draft your first TE:
This graphic is fascinating to me because it displays clear draft strategy trends right away! The 5 most optimal rounds to draft your first TE are rounds 1-5, with a steep drop off as you enter rounds 6, 7 and 8. I love that the data confirms beliefs we’ve held in the past. Round 1 does result in an above average playoff rate, but it’s minor. Rounds 2-4 lead to more success and are also less risky overall. So my answer to this question is rounds 2-4.
But that’s not where the story ends! Sometimes it’s not possible to draft a TE early, or maybe you’re not a big fan of doing so. If that’s the case, this graph does offer more insights. The worst time to attack the TE position is rounds 6-9 and after round 15. Why is that? Well, what happens in these rounds logically? In rounds 6-9 you’ve missed out on the elite TE’s and have likely either 1) Been left with the least desirable mid round option or 2) Are reaching on a late round TE. Both of these are bad bets over time, and that’s why we see such a dip, bottoming out in rounds 7 and 8. So, if you miss out on the elite TE’s, don’t draft one in rounds 6-9.
After round 9 we see success rate flatline at slightly below average in rounds 10-13. Why is that? Well, teams that waited this long have spent at least 8 picks at RB and WR already. They’re loaded at those positions and can now choose their favorite of the late round TE options. If these teams miss at TE then it’ll have very little impact on their teams. And if they hit, it’s really beneficial.
Now, there are two things I want to note:
- Yes, the round 15 number is stable
- You’re hurting your team by not taking a TE in the top 4 rounds
Let’s start with 1. I initially thought the success in round 15 was just due to the sample I chose. That can happen sometimes if you select a sample and one of the areas of that data has a smaller sample size. I tested the data by taking 10 different samples and round 15 always produced a spike in success. I don’t know that there’s a big takeaway here, but it sort of lends itself to the idea of just waiting as long as possible for a TE and taking one in the final few rounds. My only concern is that we see pretty low success rates in rounds 14 and 16. If waiting on TE was such a good strategy then why would round 15 be the only one to see a spike? That’s what has me leaning towards #2. You’re hurting your team by not taking a TE in the top 4 rounds. Over the last 7 years these are the TE’s that have been drafted in the first 9 rounds and their respective finishes overall and in PPG. As you can see, the success rate of drafting a TE 1 in rounds 1-4 is SIGNIFICANTLY higher than trying to do so in rounds 5-9. You’re basically just guessing in those rounds. Also…there’s no clear trend in rounds 5-9. Every one of those rounds has a fairly even split of hits and misses, with no single round having a significant edge in hit rate or even league winning seasons.
That’s why I think the first graphic is so important to understand! Once the early round TE’s are gone, you hurt your team by being the first to select from the next group. Whether we like it or not, there are a TON of busts at the tight end position in these middle rounds. The general strategy seems to be trying to get an early round TE, and if you can’t…wait.
This is further confirmed when we look at playoff and championship teams from last season. This chart is available for all players here, and I recommend you read that page to understand what it’s showing. But in a nutshell, we want to look at the D+F PR and D+F CR columns. Those show us the playoff and championship rates for teams that drafted and finished with each respective player. Last season was actually a semi miss for early round TE’s. Kittle missed about a month, Kelce had a down year by his standards and Waller was only healthy for 9 games during the fantasy season (with just two 100 yard games). Yet despite this, all three ranked inside the top 8 most valuable TE’s. In fact, they were more valuable than numerous tight ends that were picked up for free off FA and finished as low end 1’s to high end 2’s! Because the thing is…even though the elite tight ends had more duds than we’d like last season…that’s just how it goes with tight ends! Even Mark Andrews (who was the clear TE 1 and most valuable TE by a mile) had 4 games with fewer than 10 PPR points. Every single tight end is going to bust a bunch. The only difference is that the early round tight ends provide access to a weekly ceiling that simply can’t be matched by the late round (and often times even mid round) options.
Now again…this isn’t to say you can’t win by selecting a mid or late round tight end. That would be absurd to think! These are averages, and many rounds provide a close to average playoff rate. Just understand that success rate falls off a cliff for TE’s after round 4…and early round tight ends don’t need to have record breaking seasons to return value. Kittle was the 2nd most valuable TE last regular season despite a final stat line of 71/910/6.
Also this is a very important point. These numbers are reflecting ADP! So getting a TE in the 6th round is still fantastic if their ADP is round 4. I am by no means saying that if a TE that has an ADP of round 4 is a bad pick in the 6th round. I’m saying that historically, people drafting TE’s after round 4 have done worse on average. And they’ve had the least success in rounds 6-9. But that’s an average. The vast majority of those teams selected a TE with a 6th round ADP. Not a 4th round ADP.
When to draft your second TE:
Now this is interesting…and I’m genuinely not sure how I feel about it. I’ve long been in the camp that you just draft 1 TE. If it’s early then you can make up for the loss of early round draft capital at RB/WR by gaining an extra round late. If you take one late, then you can easily find a replacement on FA if they fail. This data however conflicts with that belief. No matter how you look at it, drafting two TE’s is almost always optimal, and the second TE should come in the double digit rounds (graph 1). The one caveat is when you select a TE in rounds 1-3. As I’ve suspected, teams that select a TE in these rounds, and then 0 after that, have an above average playoff rate.
Also, this should put to bed the idea of drafting multiple early or mid-round TE’s. Stop doing that and stop asking me about it. DO NOT DRAFT MULTIPLE TE’S IN THE FIRST 7 ROUNDS (unless you’re in a TE Premium league, which is not included in this data). Early and mid-round TE’s have a premium because they’re TE’s (and you get to use them in the TE slot). Taking multiple early effectively forces you to use one as a WR. So you’ve paid a premium and can’t take advantage. Again, only take 1 TE in the first 7 rounds.
Also for those who will ask, yes, this even holds if you draft an early TE and a mid-round QB (graph 2). Even using a ton of early and mid-draft capital at TE and QB (and thus not at RB/WR), teams have an above average playoff rate if they draft a second TE in the double digit rounds. And in this case, it’s best to do so in rounds 13-15. The same trend as only an early TE still does apply though. Teams that drafted an early TE and a mid-round QB had an above average playoff rate. So either don’t take a second, or draft another in rounds 11+. For this it comes down to personal preference since the playoff rates are virtually identical.
One other interesting point is that round 16 sees a dip. That’s the last round for the vast majority of ESPN public leagues. At this point you’re likely last to pick your tight end but were among the first to pick defense and kicker. Don’t do that! There’s a reason round 16 sees a drop off. Defense and kicker don’t need to be reached on at the expense of other positions. One of those two should be left for the final round at the very least.
Oh and for the crowd that still believes in going with one late QB and one late TE (something I myself used to do), it’s just not great (graph 7). Yes you open up a ton of upside at RB and WR…but you lose too much upside and success at QB and TE. Teams that did this made the playoffs at a below average rate no mater what double digit round their first and only TE came in.
Finally…as should go without saying…don’t draft 3 TE’s. Teams that did so had a below average playoff rate no matter what round their third TE came in (graph 3).
These graphs help illustrate the findings above:
A Better Way to Display Specific Draft Decisions
As with the running back and wide receiver articles, I wanted to construct a graphic that best displayed individual draft decisions. Please read those articles first to get a full understanding of what these graphics are saying, and especially what they’re not saying. I’m only going to include two because there’s no case where you should draft 3 TE’s in a non TE Premium league. As you can see, early TE is much more optimal than mid TE. In fact, even if you haven’t drafted a TE yet, you should avoid doing so in rounds 6-9. Also, if you don’t have a TE in rounds 13-15 then you’re doing a lot of damage to your playoff rate by not drafting one. When it comes to drafting your second, wait until round 11. You can do so at any time starting then, though you hurt your team the most if you take one in the final round. Please still draft defense/kicker last.
Also remember that you should pretty much always draft 2 TE’s. The one exception is if you drafted one in the first 3 to 4 rounds. If you did this then it’s ok to only take one…and if you want to draft a second…then wait until at least round 11 before doing so (and preferably take your second in rounds 13-15).
The Limitations of Tight End Research:
Sample sizes. And I’m not talking about the number of teams in the study. With tight end and quarterback there are very few players that fall into round buckets. So for example…in this entire 7 year sample…the only first round tight ends have been Travis Kelce and Rob Gronkowski. That’s it! The success rate for round 1 is good because they both did well in the season where they went in the first round. But if even one of them had failed, then first round success rate would have cratered. Whenever the result of a single player has a massive impact on a draft strategy, you can’t really say you have enough data.
The problem is that it would probably take over 100 years for us to have enough records of first round tight ends to consider a large enough sample. So we kind of have to work with what we’ve got! I only bring this up because I want everyone to understand how fragile fantasy can be. We can’t take away Kelce and Gronk from analysis because they exist! And other TE’s will post similar numbers to them in the future (Kyle Pits?). But at the same time, we need to recognize that early round success is largely built off the success of just a few players. That’s something we never need to consider at RB and WR. I still absolutely trust the data, especially since it all makes sense logically. I just want to raise this point because I do believe it’s important to understand.
I also want to point out that this is mostly an issue with the first round or two. After that we do have a pretty good and unique sample size. And we’ll really ramp that up next season when we add our 8th year, and also potentially triple the size of the dataset. So while we need to take the results of the first few rounds of TE and QB data with a grain of salt…pretty much every other round is perfectly fine for those positions, and every round is diverse enough for RB and WR.
