Writers of Pro Football Prospectus 2008

06 Apr 2009

ESPN INSIDER: What's a "Safe" Pick?

This Monday's ESPN Insider feature looks at which positions qualify as "safe" picks, and which offer the quickest path to the Pro Bowl.

Posted by: Bill Barnwell on 06 Apr 2009

16 comments, Last at 18 Apr 2009, 3:29am by Bill Barnwell


by Benn (not verified) :: Mon, 04/06/2009 - 5:29pm

Pro Bowl selections are a difficult metric to defend because they depend so heavily on reputation, and seem to heavily favor early draft picks. I wonder if there isn't an additional problem. Just as many tackles make the pro bowl each year as guards. But many more tackles are selected in the first round than guards, which means that each individual first-round tackle has many more players who match his talent level and reputation to compete against than each individual guard. Is Shawn Andrews likely to have been more valuable to his team than John Tait over the course of their careers? Perhaps. But the Pro-Bowl qualified depth at tackle is far superior to that at guard. It seems that in all sorts of ways Pro Bowl selection is a pretty glib metric to use for 'success,' and I'm a bit skeptical of any analysis that depends upon it and I don't really trust that taking a guard is a safer pick in round 1 than a tackle.

by Wait, what? (not verified) :: Mon, 04/06/2009 - 6:32pm

While I agree that the Pro Bowl is not always the best method of gauging success, I think the guard vs tackle debate specifically involves a question that most other positions don't- namely, are players being categorized by the position they played in college, or the position they played in the pros? Plenty of college tackles end up as pro guards. If Andre Smith makes the Pro Bowl at guard, is his team credited with drafting a successful guard, or the tackle that we currently regard him as? I assume the article looked at players by NFL position only, although that may not be the case (and the phenomenon of players switching positions may be less common than I think, although it also occurs among DB's and undersized DE's who become LB's).

by AlanSP :: Mon, 04/06/2009 - 10:24pm

I agree with Benn as far as Pro Bowls being a somewhat dubious measure of success, especially since we're talking about picks being "safe." When people talk about the relative safety of draft picks, they're really talking about the chances of failure more than the chances of success (i.e. the likelihood that the pick will be a bust). With tackles, it's pretty rare for early draft picks to just completely flame out like you see with some other positions. A few "busts" have become very useful players by shifting inside to guard.

Also there seems to be a pretty big difference between tackles taken in the first half of round one compared to the second half.

Here are the tackles picked from 1-16 in your time frame: Jamaal Brown, Robert Gallery, Jordan Gross, Mike Williams, Bryant McKinnie, Levi Jones, Leonard Davis, Kenyatta Walker, Chris Samuels, John Tait, Kyle Turley, Tra Thomas, Orlando Pace, Walter Jones, Jonathan Ogden, Willie Anderson, Tony Boselli, Bernard Williams, Wayne Gandy, Willie Roaf, Lincoln Kennedy, Brad Hopkins, Bob Whitfield, Ray Roberts, Leon Searcy, Charles McCrae, Antone Davis, Pat Harlow, Richmond Webb, Tony Mandarich, Andy Heck, Paul Gruber

There are very few outright busts in that group (Mike Williams, Walker, Bernard Williams, Roberts, McCrae, Davis, Harlow, Mandarich) and quite a few stars. Davis and Gallery were busts as tackles but have been successful as guards.

I won't go through the entire list of tackles from the second half of round one, but there are a lot more busts a very few star players (Tarik Glenn, Todd Steussie, and the late Korey Stringer being the only real success stories over the last 20 years). If we're talking about the Lions and the tackles near the top of the draft, it's sort of misleading to look at the boatload of busts from late in the 1st round.

by Newby (not verified) :: Wed, 04/08/2009 - 2:32pm

You named 32 "first half or first round" tackles, of which you named 8 "outright busts." Then you said that wasn't a big deal--but it's 25% failure rate! If you add in the two who switched to guard, that's 31% failure rate. I believe you about that being better than the "second half of first round" tackles, but isn't the expectation of the first 16 picks in the first round is that they will be more of a sure thing than a 75% success rate? I'd like to see a "you are inept" multiplier factor to handicap the numbers of teams (hello Lions) that usually choke on their first pick, versus others (Steelers, Colts, etc) that don't.

by Kibbles :: Thu, 04/09/2009 - 2:03pm

A 75% success rate might not seem that high for first rounders, but it's actually INCREDIBLE. Go back to any draft you like and count all of the first rounders that were outright "busts". Generally, you'll see that 50% of first round draft picks don't make the slightest impact at the NFL level (or worse- they make an impact, but it's a negative one). If one position is really half as likely to bust as another, then that's very "safe" by NFL standards, even if it's not safe by car manufacturer standards, or airline travel standards, or anti-terrorism standards.

by AlanSP :: Thu, 04/09/2009 - 4:13pm

For some perspective, here are the numbers for some other positions (obviously not very scientific since I don't have a clear definition of busts, but it's what I have time for at the moment.)
QBs 12/27 (not counting Leftwich or Dilfer)
WRs 16/41 (again being pretty charitable)

by Pat (filler) (not verified) :: Mon, 04/06/2009 - 11:33pm

I wouldn't use Pro Bowls because expecting a first-round pick to be a Pro Bowl player is way too high expectations. You just want them to be an above-average starter. You get them for ~10 years on average, and so if you get an above-average starter in the 1st round and an average starter in the second, you can basically maintain an above-average team.

There's also the opportunity cost to think about, too: it might be "safe" to pick a fullback in the first round, but you'd never want to do it because no fullback would be worth the cost of not acquiring an above-average tackle, for instance.

by J (not verified) :: Tue, 04/07/2009 - 10:17am

I’d have to see the details and raw data you used to create the study.

When I pulled the data from 1st round ’88 through ’05 QB’s I got 13 out of 37 as Pro Bowlers, a 35% success rate. Of course, all 1st round picks have higher success rates at all positions and many teams seek sub-1st round picks to fill the need so I feel this overstates the general QB success rate.

Also, I disagree with casting out players who succeed with other teams even though this might explain our data difference. A draft pick is made in a point in time, T(0); if a player fulfills his potential by becoming a Pro Bowl player at some point in time that validates the point in time selection and should not be invalidated by personnel decisions made after T(0). In other words, don’t blame the drafters by what the coaches and GM’s do after the draft (even if they are the same people).

Also, when I pull the data from every round over the same time period I get 31/235 Pro Bowl QB’s selected, 13.19%, which is much more in line with what you see from other time frames. In a league where draft position, talent available during certain years and multiple position needs control the process I think this rate of success is a much truer reflection.

by crack (not verified) :: Tue, 04/07/2009 - 10:31am

Yeah this article is of questionable value. Safe doesn't imply likely to produce a pro-bowler. That's almost the opposite of safe, high risk-high reward. Safe is concerned with downside not upside. The possibility for a below average NFL T to still be a serviceable or better NFL G is the biggest reason Ts are considered safe picks. Not because people expect highly rated Ts to make the pro-bowl.

by johonny (not verified) :: Tue, 04/07/2009 - 7:35pm

It always seems to me the hardest part of these type of draft evaluations is the handling of players that don't produce as drafted due to catastrophic injury early in their career.

by Chris (not verified) :: Wed, 04/08/2009 - 10:54am

I'd agree that a lot of being a high pick, getting the TV time, the hype contributes to reputation/pro bowls.

One thing I always wonder about is the whole nature/nurture thing as far as busts are concerned. Is Johnny Quarterback a bust because of the poor coaching/talent around him that help ruin him, or is he just garbage ( Ryan Leaf).

I am sure most people would agree that if Bill Bellicheck was a linebackers coach, the same exact player has a better chance of success than if Ted Cottrell were his linebackers head coach. To what extent does that all play into a pick being a bust or not?

Pat, I always wonder about the "sustainability" or to what level of talent a team must draft to "sustain" or build a great team. What is the average, or baseline, or par draft a team needs to succeed. You say an above average 1st rounder, an average 2nd rounder ( and is that it)?

You plan on getting 2 long term (decent) starters ( everything else is just gravey?) every single year when you have 24 starters.. with some being more important than others...

It would also depend on what you already have. If you are San Diego or some other team with a pretty stacked roster, you would ideally wouldn't need to bring in as many starters as say Detroit or Cincinatti....

I guess I was just wonder what the average/sustainability a team would need/desire.

by tuluse :: Wed, 04/08/2009 - 12:00pm

Well, if you accept his assertion that the player will play for 10 years (which I think is optimistic to say the least) that gives you 20 average or above average players. I think most teams can live with 2 below average starters. In your count of 24, I'm assuming that you're counting a punter and a kicker, those are easy to find, and no one drafts one in the top 2 rounds (except Herm Edwards).

by Chris (not verified) :: Wed, 04/08/2009 - 12:49pm

You are right, assuming your guys are going to be there 10 years is generous, but you also need to know that even bad teams ( like the Lions) don't start with nothing...

I did count a punter/kicker as having bad ones can really kill your team... Not all positions are of equal value and not all positions are even of equal value to different teams. I'd argue that a pro bowl punter is worth more to the Ravens or some other defensive team that will play in a game with more possessions ( and punts) than a team like the 07' Patriots. Having a horrible kicker can easily cost your team close wins.

by Bigmaq (not verified) :: Fri, 04/10/2009 - 12:27am

Safe Pick:

You can pick your friends and you can pick your nose, but, you can't pick your friends nose.

More "baseball stats" from the kings of baseball stats.

What dweebs.

by Bigmaq (not verified) :: Fri, 04/10/2009 - 12:29am

Safe Pick:

You can pick your friends and you can pick your nose, but, you can't pick your friends nose.

More "baseball stats" from the kings of baseball stats.

What dweebs.

Increasingly the people at "Cold Hard Football Facts" seem to have the edge.

by Bill Barnwell :: Sat, 04/18/2009 - 3:29am

Unspeakably funny.