I think that I’ve finally got enough done to post at least a first version of this and put it out there. If you haven’t been following my series on looking at Brad’s draft history to try and pin-point guys the Lions might target in this year’s draft, you can catch up here:
Part 1: Who is a 'Brad Holmes' draft pick?: Diving into Brad's pick history
Part 2: Who is a 'Brad Holmes draft pick'? Pt. 2: PFF Score Eligible Players
A brief overview: I’ve been looking at PFF and RAS scores to find trends and commonalities in Brad’s drafting history. Again, if you’re not a fan of PFF, that’s fine, but they’re the grades I’m using for this dataset. In the last topic, I had combed through the trends to eliminate some players to narrow it down, just based on PFF scores. Now, I’ll be doing the same with RAS, in combination with PFF scores, to score the class based on Brad’s drafting history.
Introducing: The Knee Cap Score
What is the KCS? I won’t bore you with all the details, but it’s a weighted score blended on PFF Scores (final year and previous 2 years, with more weight applied to the final year score, which also takes snap counts into effect) and RAS scores. It is a score out of 10, with 10 being the highest and 0 being the lowest. I’m not taking character/football IQ/positional weights, etc into account. A lot of that stuff we’ll never know on how a player interviews, how much Brad drafts on ‘need’, etc, so I don’t think it’s worth trying to chart. You guys can make those distinctions on your own. This is purely a performance/athleticism dataset.
I honestly think it’s a good indicator of the type of player Brad likes or has been drafting. At least for the first 3 rounds - after that, it’s pretty unpredictable. For the sake of brevity, let’s take a look at just the Lions Day 1 & 2 draft picks over the last two years and their WRAS (Weighted RAS), WGRD (Wighted Grade) and KCS (Knee Cap Score):
| PICK | POS | PLAYER | WRAS | WGRD | KCS |
|---|---|---|---|---|---|
| R1.2 | ED | HUTCHINSON | 4.940 | 4.725 | 9.665 |
| R1.7 | OT | SEWELL | 4.495 | 4.790 | 9.285 |
| R1.12 | WR | WILLIAMS | N/A | 4.080 | 4.080 |
| R2.41 | DI | ONWUZURIKE | 4.365 | 4.125 | 8.490 |
| R2.46 | DL | PASCHAL | 4.850 | 4.500 | 9.350 |
| R3.72 | DI | MCNEILL | 4.255 | 4.535 | 8.790 |
| R3.97 | S | JOSEPH | N/A | 4.520 | 4.520 |
| R3.101 | CB | MELIFONWU | 4.845 | 3.940 | 8.785 |
Let’s look at some observations and baseline trends. The average KCS for each round (not taking the N/A RAS scores into account, but still counting the WGRD):
- Round 1: 9.25
- Round 2: 8.92
- Round 3: 8.88
- (R4: 7.12, R5: 3.07, R6: 8.75, R7: 6.28)
As you can see, at least for rounds 3, the KCS shows that players scoring into the high 8s and 9s are the guys that Brad has been taking in the first 3 rounds. In the cases of Jamo and Kerby, who didn’t have RAS scores, they had at least Weighted Grade Scores of 4.080.
The lowest KCS for any day 1 or 2 pick was Levi at 8.490, so that will serve as the cutoff for my list of players I think Brad will target for the first 3 rounds. I will also include any players that have a WGRD of 3.940 (Iffy’s WGRD) as a floor for those without a RAS. (Jamo would’ve needed an 8.84 RAS and Kerby a 7.96 for them to reach the 8.490 KCS - I think that is easily done for both of them. I will be doing these calculations for players without a RAS score, who meet the WGRD standard).
Please don’t complain if your ‘guy’ isn’t on this list. There are tons of guys I like for the Lions who didn’t make this list. This is more just to see if after the draft, if the guys the Lions select fit based on the criteria they’ve shown in the past. There is always the chance that they do something different than they have in the past.
Here is the Top 50 list as it sits now, I have the rankings sorted by their KCS ranking (those without RAS scores will not show on this list. I will make another post after this for a breakdown of each position:
Hope you enjoyed this. Feel free to ask question or for specific player’s grades on this. Curious to see how the real draft falls in line with this data.
