The Vikings drafted offensive tackle Brian O’Neill with an eye towards how he’ll help the offensive line a few years down the line, without much concern about his short-term prospects as an immediate contributor.
While this doesn’t resolve the problems they might have in 2018, it could mean a more stable future down the road.
That can only really be the case, however, if O’Neill actually develops. Touted as an incredibly athletic but very raw offensive lineman from Pitt – like another former Vikings draft pick – there’s a risk that the second-round prospect could flame out like so many other linemen set to develop into full-time starters.
How often do linemen who sit end up becoming quality contributors in the NFL?
In order to look at the question, at least as far as it pertains to O’Neill, I investigated every second or third-round pick drafted between 2006 and 2015. I divided them into two groups: those who played fewer than 500 snaps their rookie year and those that played over 500 snaps.
To create a baseline of comparison, I grabbed their Pro Football Focus grades for their third-through-fifth seasons and added a bonus for every snap played during those seasons. Essentially, linemen with better PFF scores per snap did better and so did linemen that played more snaps. Those that were cut early in their careers, therefore, don’t benefit from having a smaller snap count.
The final score distribution was normalized to act like PFF grades in any individual year. Linemen that played fewer than 500 snaps in their rookie year had an average veteran score of -22, which is pretty bad; think Menelik Watson, John Miller, Jake Fisher and Dallas Thomas. Those with more than 500 snaps did much better – they had a score of +8 on average. Scores in that range include players like Rob Havenstein, Ali Marpet and Trai Turner.
The best overall scores belonged to players like Marshal Yanda, Kelechi Osemele and Mitchell Schwartz, while the worst belonged to Billy Turner, Chaz Green and Marcus Martin.
Looking at simple success rate seems to tell the same story; 72 percent of linemen with significant rookie playtime ended up scoring above 0.0 as veterans, while only 42 percent of the rookies without that playtime did the same thing.
That might seem like a pretty definitive statement; that players who play a lot of snaps in their rookie year turn out to be better than their similarly-drafted peers without that rookie exposure. But the graph above can tell you that the relationship is not actually all that strong. The correlation from rookie snaps to veteran performance is week overall and there’s a wide distribution of outcomes.
The absolute worst performers played very little early on, but so did some of the best performers. Eric Winston, Ryan Kalil and Brandon Brooks are pretty good evidence that it is certainly possible to miss time as a rookie and still craft a top-level career.
As another way to look at comparable data, we can check to see if rookie starts – which shouldn’t be all that different from snaps played – could tell us the same story.
Unsurprisingly, the data looks pretty similar. The odds are pretty good for rookies who start 13 or more games, and the cluster certainly seems tighter, but other than that, the distribution seems pretty random.
What this might be able to tell us is not that missing rookie time is a bad thing – only that grabbing rookie time is a good thing. There aren’t red flags to be found in rookie playing time for offensive linemen, mostly green lights. Even so, you could end up with Will Rackley or Eben Britton, a hardly enticing outcome.
For a less scientific but perhaps more precise approach to finding offensive linemen who are “developmental,” I went through the scouting reports from NFL.com for every lineman drafted in Day 2 to look for signs that a player was raw. That search included keywords like “raw,” of course, as well as “developmental,” “project,” “takes time,” etc.
When it wasn’t clear if a prospect was explicitly “developmental,” I would look for alternate scouting reports to see if they came to a consensus conclusion.
For example, the impressively large Rob Havenstein was described as having “the tools to be a starting right tackle in the league,” which seemed ambiguous enough to check with other sources. One described him as “technically sound,” while the other opined that there aren’t many reasons “to assume Rob Havenstein won’t be the starting right tackle on day one.”
That tells us that Havenstein wasn’t a “developmental” prospect in the same way that someone like Vlad Ducasse was, described as “extremely raw.”
It isn’t always the case that a big athlete with pass protection issues is labeled as a developmental project. For example, Phil Loadholt had a lot to learn before becoming a high-level right tackle but was tagged as a prospect that should see the offensive line rotation sooner rather than later.
In any case, tagging a player as “developmental” in a scouting report didn’t seem to have an impact on their long-term prospects.
The average score for a “developmental” prospect was -3.9, while the average score for a non-developmental prospect was 0.2 – not every lineman was easy to find clear reports for, so only data for 75 of the 105 linemen drafted in that timespan were used.
The success rates were similar as well, with a 47.4 percent success rate for developmental lineman and a 56.8 percent success rate for immediate starters.
It turns out that many rookies that were tagged as developmental ended up starting for their franchises pretty early in their career; that includes Donovan Smith (who had a “good shot of getting drafted and fighting for a backup spot”), Kelechi Osemele (who “may make some teams wonder if he is worth the time and early-round investment as a project when they have an immediate need to fill”) and Mitchell Schwartz (“A nice later round developmental type draft pick who could eventually fight for playing time”).
Conversely, some linemen expected to provide an immediate impact, like Josh LeRibeus (who “has the size and technique inside to play early at the next level”), Jake Fisher (who can “compete for a starting position right away”) and Rodney Hudson (who “should be an immediate starter at guard or center and could make a few Pro Bowls down the line”), had to wait their turn.
What this tells us is that the Vikings aren’t necessarily wrong to find a developmental lineman in the second round, as those can be huge successes. Finding the next Terron Armstead – who some compare Brian O’Neill too – or Sebastian Vollmer would be a wonderful thing.
Instead, what might give us a clue as to how O’Neill is doing as a potential tackle-of-the-future is the context in which he sees – or misses – time on the field. Should an adequate right guard push Mike Remmers to the right tackle position while Riley Reiff mans the left, there likely won’t be any cause for concern. But if Rashod Hill or another tackle is forced to start on the right side and struggles while O’Neill rides the pine, then we can at least raise some eyebrows.
Even then, the data suggests that O’Neill wouldn’t be doomed to irrelevancy. Not only would that follow the path of Armstead and Brooks, it would also match fellow developmental tackles Morgan Moses and Joe Barksdale.
The possibility space is pretty large for O’Neill, even as a developmental tackle. Though the Vikings have had issues developing offensive linemen in their recent history, that’s no reason to think that “developmental tackles” are themselves a problem.