clock menu more-arrow no yes mobile

Filed under:

Shooting Stars and Falling Stars, Part 1

Looking at the relationship between star rating and performance for elite QB recruits

Michael Penix and his golden left arm
AP

The merry-go-round of how much recruiting stars matter is never-ending, and we’ve had great examples of the good and bad side of that right here at UW. Sam Huard was a 5-star golden boy, and he was an absolute non-factor on the field before transferring down to FCS Cal Poly, while Michael Penix Jr. was a middling 3-star, the 66th-ranked QB in his class, sandwiched between such legends as Kenyon Oblad (to UNLV) and Coran Taylor (to Illinois), and... well, you know how he’s played the past season and a half. In between, of course, we have everyone’s favorite whipping boy, Dylan Morris, a solid 4-star get rated the #5 pro-style QB in the Class of 2019, played a pretty solid shortened COVID season as a true freshman and a not-so-good 2020 under the blasphemous reign of JonDon. Maybe he’s good now. Maybe he sucks. Maybe he’s just OK.

Is recruiting status totally irrelevant, then? Or nearly so? Player development, coaching, and system obviously have a lot to do with how a QB comes out. So do injury luck and supporting cast, or even whether you end up stuck behind a superstar and you never get your shot. But how much does raw talent (to the extent that it’s reflected by recruiting rankings) really matter? That’s what I got curious about and that’s what I’m exploring in this article.

THE METHOD

The QBs in this analysis cover a span of 5 recruiting classes, 2017-2021, using 247sports.com. This includes 13 5-star QBs and 63 4-stars in those 5 classes. Because there are so many 3-stars and I only have so much free time, and since part of the point is to compare UW’s QBs to their similarly ranked peers, I also included the 53 3-star QBs with the same recruiting ranking as Michael Penix (85, dead center in the 3-star range) for a total N = 129.

I then looked at those QBs’ stats over their time in college, using sports-reference.com (including games for this season up to and including week 4, so games from the weekend of 9/30 are not included) and devised a formula to measure their total accomplishments as QBs. Notably, I did not include wins and losses as part of this formula; yes, I understand the QB has an outsized impact on team success compared to any one other position, but I don’t think more than every other position. I think there’s way too much noise to assign W/L to the QB. Instead, I’m focusing on their individual stats (with the obvious caveat that, yes, this runs the opposite direction to my statement above about wins and losses, since someone needs to catch every pass they throw).

I focus on total accumulated stats because they reflect a lot more about the player’s total accomplishments over the course of time on that team. To use a UW RB analogy, Corey Dillon had an amazing 1996 season at UW, but if you’re ranking all-time greatest Husky RBs he’s a long way down the list behind guys like Gaskin, Sankey, Polk, Lewis, Kaufman, and McElhenny. That year was great, but the fact that he went on to NFL success later instead of toting the rock for the Huskies didn’t help the Huskies. Big ups for his individual career, but he only gets to count his college time when measuring his college success. Same with the QBs here. Racking up numbers across multiple years helps your college team more than having a big season and leaving for the NFL or sitting as a backup for multiple years and then coming in for one big final year. Consistency matters and volume matters to show what you’ve gotten done between the lines for your alma mater.

THE FORMULA

I considered passing yards, TDS, interceptions, rushing yards and TDs (fumble data was tricky to find for everyone so I skipped it; ideally, I’d like to include it) and some impact from passer rating, though efficiency is less important than volume for this purpose. I also included a conference modifier for players outside the P5, based on how bad their conference is relative to typical P5, and had a bonus award at the end for players who earned AP All-America status, won the Heisman or Davey O’Brien, and one last bonus based on their NFL draft status (again, yes, that’s after their college career, but it’s also sortakinda the last thing they do as a player that’s from their school - one last day when he’s “Trevor Lawrence from Clemson,” before he becomes “Trevor Lawrence of the Jacksonville Jaguars.” If you want to play along at home, here’s the formula. Quibble if you like, but it at least gives us a stable baseline for comparison:

(Passing Yards/100) + (Rushing yards/100) + Adjusted TDs (Passing TDs + Rushing TDs - INTs x 1.5) + (Passer Rating/10) = BIG SCORE

Adjustments:

  • Conference Modifier: P5 = 1, MWC, AAC = 0.8; Conference USA = 0.7; MAC, Sun Belt = 0.6; FCS = 0.4
  • All-American/Awards: +20 Heisman or +15 1st-team, +10 2nd-team, +5 3rd-team; +5 Davey O’Brien (unsurprisingly, the Heisman winner is usually the 1st-team All-American, but not always)
  • NFL Draft Status: Top 5 pick +10, R1 (after pick 5) +8, R2 +6, R3 +4, R4-7 +2

Each QB’s BIG+ SCORE is determined by multiplying their base BIG SCORE times their conference modifier and then adding any bonus points after that for awards and draft status.

So how do they stack up?

Note on Means and Medians: As in previous articles, I include here both means (just the numerical average of all scores) and medians (the midpoint value, where half of the values are above and below), because means are easily skewed by a huge number at the top that might distort the information.

THE 5-STAR QBS

5-Star QBs by BIG+

Stars Rating Year Name School Transfer? Yards TDs INT Rtg Scrim Yds TDs Total Off Adj TD BIG SCORE Conf Mod BIG+ SCORE NFL Draft All-Am/Heisman Other Honors
Stars Rating Year Name School Transfer? Yards TDs INT Rtg Scrim Yds TDs Total Off Adj TD BIG SCORE Conf Mod BIG+ SCORE NFL Draft All-Am/Heisman Other Honors
5 Median N = 13 7442 65 11 157.6 340 9 7782 39 132.58 147.09
5 Mean 6420 57 13 156 429 10 6849 46 130.49 139
5 101 2018 Trevor Lawrence Clemson 10098 90 17 164.3 939 18 11037 82.5 209.30 1 224.30 10 5
5 99 2021 Caleb Williams Oklahoma 7649 78 9 175.6 893 19 8542 83.5 186.48 1 221.48 10 20 5
5 101 2020 Bryce Young Alabama 8356 80 12 165 162 7 8518 69 170.68 1 178.68 8
5 98 2017 Tua Tagovailoa Alabama 7442 87 11 199.4 340 9 7782 79.5 177.26 1 177.26
5 98 2020 CJ Stroud Ohio St. 8123 85 12 182.4 136 1 8259 68 168.83 1 168.83
5 101 2018 Justin Fields Georgia Ohio St. 5701 67 9 178.8 1123 19 6824 72.5 158.62 1 178.62 20
5 99 2019 Spencer Rattler Oklahoma South Carolina 8863 65 26 156.1 385 13 9248 39 147.09 1 147.09
5 101 2020 DJ Uiagaleilei Clemson Oregon St. 6509 43 20 127 1012 20 7521 33 120.91 1 120.91
5 98 2021 Drake Maye North Carolina 5597 44 11 157.6 879 10 6476 37.5 118.02 1 128.02 10
5 99 2018 JT Daniels USC Georgia, West Virginia, Rice 8184 56 28 139.9 -331 1 7853 15 107.52 1 122.52 10 5
5 100 2021 Quinn Ewers Ohio St. Texas 3210 24 6 142.4 -15 4 3195 19 65.19 1 65.19
5 98 2017 Davis Mills Stanford 3464 18 8 141.9 94 4 3558 10 59.77 1 63.77 4
5 98 2021 Sam Huard Washington 265 1 4 94.5 -45 0 220 -5 6.65 1 6.65

Here is where we begin to see just how far off the wagon our beautiful baby boy Sam Huard fell in his time at UW. It’s not just that he did so little, but that the only other 5-stars with a BIG+ of less than 100 are Davis Mills (who didn’t start a ton but showed enough to get drafted in the 3rd round, and Quinn Ewers, who’s only 16 games into his career. Generally speaking, a 5-star QB is pretty much a sure thing to be at least solid if not amazing. So far, he’s orders of magnitude worse than anyone in this group. (Spoiler: It Gets Worse).

I’m not doing a full analysis, but I’m doing a quick scan back through other 5-star QBs before this time range:

2016: Shea Patterson, Jacob Eason (both solid, neither incredible)

2015: Jarrett Stidham (good), Josh Rosen (crap in the NFL but good in college), Kyler Murray (Heisman), Blake Barnett (bust at Alabama and Arizona, but did put up 16 pedestrian games at South Florida)

2014: Kyle Allen (two decent years at Texas A&M)

2013: Christian Hackenberg (good), Max Browne (sadly for Washington HS QB fans, this might be our closest contender - he’s still a clear step above Huard, but was a mediocre-at-best backup at USC and Pitt)

2012: Gunner Kiel (good)

2011: Jeff Driskel, Braxton Miller (both good)

2010: no 5-star QBs

That’s as far back as I can find data in a quick search, so over the course of the 25 5-star QBs over the course of 12 recruiting classes, Huard is by far the last successful and it isn’t close. Browne and probably Barnett are in the next category of never-weres, a few more like Mills who were in the “solid but not amazing” bin, and the rest all somewhere between good, very good, exceptional, and Heisman/Top 5 pick land.

The takeaway: The bust rate on 5-star QBs is really low. You’ve got about a 50% chance of getting an absolute stud and about a 90% chance to get at least a solid, serviceable QB. Looking back, only about a 10% chance to get a bust. Bummer for us that two of them came from the Seattle area. :(

Stay tuned for next time to bring in the 4-stars and the 3-star Team Penix types!