Pretty interesting read about hit rates.
http://www.arrowheadpride.com/2015/2/20/8072877/what-the-statistics-tell-us-about-the-draft-by-round
What the stats tell us about drafting positions by round
By
WesternChief on Feb 20, 2015, 9:20pm CST
42
Denny Medley-USA TODAY Sports
Tweet Share
Pin
From the FanPosts. Fun numbers. -Joel
As the Combine gets underway, I offer up some numbers to consider. I started this process in a previous post on the best round to draft WRs. This post attempts to provide a broader context from 10 years worth of draft data on all positions. The information was taken from
Pro Football Reference.
The debate is already raging over which direction Chiefs GM John Dorsey should go. Mix in feelings about free agency and the right or wrong direction the front office has already taken with the current roster and you find the hodgepodge of opinion that is
Chiefs' Kingdom.
Criteria
This post has a simple criteria: How many players were drafted by position and round over the last decade and how many went on to become a starter.
I did not distinguish superstars from regular starters. The determination of a starter comes from whether the player started at least half of their career. Obviously, this will run the gambit from below average to high performing starters.
The reality is that if you can start in this league for at least half of your playing career, you are better than most. If you would like to debate the merits of players at a particular position be my guest. However, I found that it would require a lot more work than I was willing to do to put together subjective criteria to determine various levels of starters. This also does not take into consideration undrafted free agent starters in the league.
The General Numbers
I broke the positions down as follows: QB, RB, WR, TE, OL, DL, LB, DB.
What we find is that from 2005 until 2014:
- 2,465 players (non-kickers) were drafted.
- 629 of those drafted became starters for at least half of their NFL careers
- Of the 2,465 players: 122 were QBs, 207 were RBs, 421 were OL, 143 were TE, 317 were WR, 442 were DL, 303 were LB, and 510 were DB.
The first thing we notice is that there is a very small percentage of QBs that have been drafted in the last decade. And you wonder why it's so hard to find a really good elite QB. Second, offense and defense were almost evenly split -- 1,210 offensive players and 1,255 defensive players.
Position Specific
When we get into looking at each position, the first round is king. Even though the first round rules, some positions fare a lot better than others. The disparity among the other rounds varies by position. However that cannot be overlooked when you look at the statistical probability of success by round.
Now let's look at little closer at the QBs.
- Of the 122 QBs drafted in the last 10 years only 25 have been starters for at least half of their career.
- The first round gives you a 63% chance of finding a starter.
- The second round gives you a 27% chance, the third a 17% chance, then it really plummets from there with 8% in the fourth and 6% in the 7th.
- In the last 10 years, 38 QBs have been drafted in the 5th and 6th rounds and not one has become a starter.
- On average three QBs are drafted in the 1st. This year looks to be a little below average with two probable first rounders.
With the
Alex Smith debate raging, no matter where you fall in the argument you will have to agree that unless you catch lightning in a bottle (Russell Wilson) draft success is limited to mostly the first round. Also, I would guess that a majority of those QB starters were high to mid first rounders. It also means that there is a high bust rate for any QB taken after the first round. The Tom Brady /
Russell Wilson's of the world are rare occurrences. For the Chiefs, this means that the likelihood of
Aaron Murray becoming more than a clipboard holder is about the same as my blind dog finding buried treasure in my backyard. So unless Mariota or Winston falls to No. 18, bank on Smith being the guy for a few more years.
Running Backs
The position has been devalued over the years but the statistics still show that it is better to draft early if you are looking for a starter.
- Of the 207 players drafted 33 have become starters for half their careers. This gives an indication that there is a lot of Running Back By Committee (RBBC).
- There is a very high bust rate for RBs. The first round gives you a 58% chance of finding a starter followed by 25% in the second, 16% in the third, 11% in the fourth, 9% in the fifth, 6% in the sixth and 0% in the 7th.
- If you rank the rounds by the total RBs drafted you find that the greatest number are drafted in the 7th, followed by the 4th, 6th, 2nd, 3rd, 1st, and 5th.
If you want a stud RB, they are likely to come from the first round. If you are looking for depth, the fourth round seems to be the place to go. This year Todd Gurley and Melvin Gordon have the first round grades based on statistics it seems likely that one of the two will be a bust. However, this could be like 2007 when AP and Beastmode went in the first round. For the Chiefs, since 2007 they have taken a RB every draft except 2010. Both Charles and Davis were the highest drafted both coming in the 3rd round.
Offensive Line
Let's look at one of our biggest areas to fix. Offensive line is typically one of the safest positions to draft because the success rate is greater than almost all the other positions.
- Of the 421 players drafted, 147 wound up as starters for at least half their career.
- The first round has an 83% success rate. The second round is almost as good with 70%. Even the third and fourth aren't too shabby in comparison to success rates of other positions in the same rounds. (3rd - 40%, 4th - 29%).
- The later round success rates hold up well (5th and 6th - 16%, 7th - 9%) but the numbers are driven down due to the higher numbers selected later.
- On average, 5 linemen are drafted in the first round. The first and second rounds, on average, draft the fewest offensive linemen.
- If you were ranking rounds by the numbers drafted it would run 7th, 6th, 5th, 4th, 3rd, 1st, 2nd
For all of you pro offensive line with the first pick, the low "bust" factor plays to your advantage. Although it can be countered that there is a pretty high success rate later on in comparison with other positions. This would support the argument that you can wait some and still have a good chance of success. For the Chiefs, the last three years have seen the Chiefs select two OL each draft. Dorsey so far has drafted four OL, with three coming in the sixth round.
Tight End
With this year light on quality tight ends, it has been floated that maybe the Chiefs should consider Maxx Williams at No. 18 depending on how the draft falls. Overall, the last decade has shown us that it really doesn't make a lot of difference which round you select a tight end.
- Tight ends are rarely drafted in the first round. Only nine tight ends were drafted in the first round in the last decade. Of those nine, six have had success.
- The success rate for tight ends is as follows: 1st - 67%, 2nd - 50%, 3rd - 39%, 4th - 33%, 5th - 32%, 6th - 26%, 7th - 0%. In other words, selecting a tight end in the 3rd through 6th doesn't make a lot of difference.
- The highest number of tight ends has been selected in the 7th round (31) with the least amount of success (0%)
- The third and fifth round seem to be good bets for picking a tight end. Since if you were ranking based on number drafted it would go 7th, 3rd, 5th, 4th, 6th, 2nd and 1st.
These numbers lead you to believe that you are better off looking for a draft "gem" among your later round TEs due to the fact that the success rates are similar across the board. The Chiefs have selected six TEs over the last decade with half being third rounders (
Brad Cottam,
Tony Moeaki, and
Travis Kelce). With potential for comp picks in the 3rd, 4th or 5th this year, those might be the best times to consider another TE rather than picking one early.
Wide Receiver
Again, another position of need. The crazy thing is that unlike the other skill positions (QB, RB, TE) they don't have a goose egg in one of the rounds. WR has at least had success in every round. That said, the numbers are really that great and the bust factor in the first round is tied for the worst overall.
- Of 317 selected, only 74 have become starters for at least half their careers.
- The first round success rate is 58% and the second round is almost as good at 49%.
- The third round has the second highest number of receivers drafted with 52 but only a 25% success rate.
- If you are ranking the rounds by numbers drafted, you would go 7th, 3rd, 4th, 6th, 2nd, 1st and 5th.
- The fifth round actually has a higher success rate (16%) than the fourth round (12%). The sixth is at 9% and the seventh is 5%.
- On average, four WRs are taken in the first round and then you see an average of five per round for the 2nd, 3rd, 4th, and 6th rounds.
The name of the first receiver taken in this stretch pretty much sums up the success that KC has had drafting receivers during that time frame -- Craphonso. Dorsey really hasn't attempted to draft a receiver yet (much to the consternation of many) but Pioli tried and failed five times (unless you consider McCluster and Hemingway successes). With this again being a deep WR draft, you shouldn't be disappointed if Dorsey waits until the second round to go for a WR because statistically he'll have about the same chance of success.
Defensive Line
Now, we move on to the highest drafted position for the first round and second highest overall. This year shouldn't be much different. The numbers may be slightly skewed when you consider that several college listed DEs convert to OLB. However, for every
Dee Ford that isn't considered a success (currently), there is a
Tamba Hali that does count.
- Defensive line ties with running backs and wide receivers for the lowest first round success rate at 58%.
- Of 442 players selected, only 114 have become starters for at least half their careers.
- Drafting a defensive lineman in the 4th round has the highest success rate of all positions at 37%.
- If you were ranking rounds by number drafted it would be 7th, 1st, 3rd, 4th, 6th, 2nd and 5th.
- The success rates are as follows: 1st - 58%, 2nd - 26%, 3rd - 27%, 4th - 37%, 5th - 13%, 6th - 13%, 7th - 3%
- On average, 8 linemen are taken in the 1st, 5 in the 2nd, 6 in the 3rd and 4th, 5 in the fifth and sixth and 9 in the 7th.
It's interesting to think that just last year,
Jadeveon Clowney was the next big, sure thing. His injury and microfracture surgery has cast a pale over his career. Statistically, he has a high bust potential. Yet, this little fact is not mentioned around combine or draft time. For the Chiefs, the results have been mixed when it comes to the line. While you could say that
Tyson Jackson and
Glenn Dorsey were technically successes due to being starters, their overall results were not necessarily considered as such. Poe has been the shining star but the jury is still out on Bailey. The Chiefs later round d-line picks have been rather anonymous.
Linebacker
Again, we come to another position of need. The numbers regarding the linebackers surprised me. I will say again that these might be a little skewed because of the lack of converted DEs on the list.
- Of the 303 players drafted, 72 became starters for at least half their careers.
- Almost half of the linebackers drafted came in rounds 5-7 but only 5 total starters came from those rounds.
- If you select a linebacker in the first round, there is a relatively low bust rate. As a matter of fact, Linebackers have the second highest first round success rate behind only the offensive line. 70% of first round linebackers were starters for at least half of their careers.
- There isn't much drop off in the second round either, with a success rate of 55%.
- The third round isn't that bad at 34% but after that the drop is steep with the 4th round plummeting to 16%, the 5th round to 4%, 5% in the 6th and only 2% success in the 7th.
For those debating the merits of Perryman, Kendricks, McKinney, Dawson, etc. with an early round pick, the numbers say they could be a good bet. However, those wanting to wait until the 4th round or later to find DJ's replacement will be really trying to find the needle in the haystack as
Nico Johnson proved. For the Chiefs, we have been fortunate, in that, we have only drafted five LBs in the last 10 years and two of those have turned into All-Pros (Johnson and Houston). The statistics say that a LB in the first three rounds gives us an opportunity to keep the success rolling.
Defensive Backs
Finally, we come to the largest drafted group in the last decade. Nearly, 21% of all players drafted in the last decade are defensive backs.
- Of the 510 defensive backs drafted, 121 became starters.
- You have the same success rate drafting a DB in the 7th round as you do drafting one in the 4th (11%)
- On average, you have the following number of DBs taken per round: 1st - 6, 2nd - 6, 3rd - 7, 4th - 8, 5th - 8, 6th - 7, 7th - 9
- After you get past the success rates of the first two rounds (64% in the first, 46% in the second), there isn't a huge difference in success (24% - 3rd, 11% - 4th, 17% - 5th, 8% - 6th, 11% - 7th)
- DBs provide the best success potential in the 7th round versus other positions in that round.
KC obviously needs to pick up a few new faces for the secondary, however it isn't necessarily a position that you need to reach for early. Guys like Nick Marshall, Eric Rowe, and Craig Mager who project into the 5th round have the potential to fit into the 17% success category while also fitting Dorsey's preference for larger corners.
Notables from the numbers
A couple quick things to consider as you think through who you think the Chiefs should or might take.
- If you want a safe first round pick, OL (83%), LB (70%) and TE (67%) have the lowest "bust" rates.
- TEs have a pretty reasonable chance of turning out in most rounds.
- You are just as likely to have the same amount of success selecting a WR in the first or second round.
- QBs seem to be first round or bust.
- O-line in the first four rounds is a pretty safe bet.
- Never take a TE, RB or QB in the 7th round if you hold out any hope of them being a starter.
- Taking a defensive lineman in the 4th round has a higher success probability than a 2nd or 3rd round pick.
- RBs are a dime a dozen and so you might as well have a committee.
Obviously, there are other factors that go into building a draft board. However, these numbers should give you pause when considering the potential in a given position and whether it's better to go after the position early or late. Looking at the outline for success, you can see why OL is considered the safe choice.
Hindsight is always 20/20 but these numbers really opened my eyes as to why Dorsey as a first year GM would draft Fisher instead of Dion Jordan, Sheldon Richardson, Barkevious Mingo and the like.
Historic Success Chart
The numbers show us the following outline for finding consistent starters:
1st Round - OL (83%) LB (70%) TE (67%) DB (64%) QB (63%) WR (58%) RB (58%) DL (58%)
2nd Round - OL (70%) LB (55%) TE (50%) WR (49%) DB (46%) QB (27%) DL (26%) RB (25%)
3rd Round - OL (40%) TE (39%) LB (34%) DL (27%) WR (25%) DB (24%) QB (17%) RB (16%)
4th Round - DL (37%) TE (33%) OL (29%) LB (16%) WR(12%) DB (11%) RB (11%) QB (8%)
5th Round - TE (32%) DB (17%) WR (16%) OL (16%) DL (13%) RB (9%) LB (4%) QB (0%)
6th Round - TE (26%) OL (16%) DL (13%) WR (9%) DB (8%) RB (6%) LB (5%) QB (0%)
7th Round - DB (11%) OL (9%) QB (6%) WR (5%) DL (3%) LB (2%) RB (0%) TE (0%)
Dorsey's Drafts vs the Numbers
So what have John Dorsey's two drafts with KC looked like?
2013 - 2014
Eric Fisher OL (83%) - Dee Ford DE (58%)
Traded - Traded
Travis Kelce TE (39%) Knile Davis RB (16%) - Phillip Gaines DB (17%)
Nico Johnson LB (16%) - D'Anthony Thomas RB (11%)
Sanders Commings DB (17%) - Aaron Murray QB (0%)
Eric Kush OL (16%) Braden Wilson RB (6%) - Zach Fulton OL (16%) Laurent Duvernay-Tardiff OL (16%)
Mike Catapano DL (3%)
What we notice is that the first year, Dorsey was very safe in his picks. This seems consistent with a GM who isn't real sure what he has inherited. Fisher, Kelce, Commings and Kush were all relatively safe picks based on historic numbers. Even Davis and Johnson weren't bad. Wilson and Catapano were / are unlikely to be more than roster filler. You could even say that after Dorsey secured a relatively safe pick in Kelce in the 3rd, he used the comp pick to take a little flier on Davis.
Compare that with his 2014 draft and we see that he was willing to roll the dice a little more.
In 2014, Dorsey was coming off a team with 11 wins and he knew that he could look a little further down the road with this draft. Ford, Gaines, Thomas and Murray by position and round are in the lower half of the historic success rate. Fulton and LDT rank better based on round. As a matter of fact, Fulton became a full time starter, while Gaines and Thomas could be considered part-time players. Thomas will probably not become anything more than a 10-15 play per game guy plus PR.
Moving forward to this year, it will be interesting to see what Dorsey does especially since the team has a few positions that need immediate upgrading. This year seems to have a sense that we need to climb back a little and the future can wait.
What should we expect?
Here are a few options based on the numbers, of what might provide a good draft
If Dorsey is satisfied with his LBs (DJ's recovery, his FA pick-ups), the following draft would be ideal:
1st - OL (83%) 2nd - WR (49%) 3rd TE (39%) 4th DL (37%) 5th DB (17%) 6th OL (16%) 7th DB (11%) Any comp picks would be used on RB, WR, DB
If Dorsey is satisfied with OL (Allen / Stephenson's return, LDT development plus FA pick-ups):
1st - LB (70%) 2nd - WR (49%) 3rd TE (39%) 4th DL (37%) 5th DB (17%) 6th DL (13%) 7th DB (11%) Comp picks used on RB, OL, WR
If Dorsey is satisfied with WR (Major Free Agent pick ups):
1st - DB (64%) 2nd - LB (55%) 3rd OL (40%) 4th DL (37%) 5th WR (16%) 6th TE (26%) 7th DB (11%) Comp picks used on RB, OL, DL
Conclusion
While there is a lot more that goes into building a draft board, these numbers give us something to think about and help bring some structure to an inexact science. I know I'll be watching the Combine and the draft buildup a little differently knowing the success / starter rates.
