Last week in the game day thread I caught a little flack for a couple of things I posted. Since it’s hard to fully explain myself in a fast-moving GDT while a little drunk and highly pissed off at the on-field results, I thought I would do it in a fan post instead.
The two things I said that I caught the most flack for were (1) this is the pretty much the highest level of performance you can expect from a true freshman quarterback in an air raid system, and (2) Baker and Webb will improve over time and it’s too early call either of them busts. I’ll break down my reasoning behind each in turn.
First, I’ll explain why I think this is the best, or close to the best, you can expect a true freshman QB to perform. Let me briefly explain why I chose to only compare air raid QB’s, rather than all true freshmen starters: I think any comparisons between pro system and spread system QB’s would be apples to carrots. The big difference to me is, in pro style offenses, it’s a lot easier to dumb down an offense. The coaches can work in one and two-read passing plays or have the QB only read one side of the field. In a spread system, QB’s have to read the entire field and make multiple reads by necessity on most plays. Likewise, most pro-style offenses use the run to set up the pass, which makes a qb’s job potentially easier. Obviously, some air raid QB comparisons (i.e. RGIII to Webb) will be apples to oranges, but at least they’re both fruits. For comparisons sake, here is a chart with all starting true freshmen QB’s in BCS conferences since 2003, http://media.247sports.com/Uploads/Boards/805/805/297424.jpg, note that both of our QB’s are on track to top this chart in terms of completion percentage and be in the upper third of quarterback efficiency.
The one disadvantage of my approach is it yields a really small sample size: I’ll be comparing our two starters to Baylor’s Robert Griffin, Houston’s Kevin Kolb, and Cal’s current starting QB Jared Goff (if anyone can think of any others, let me know in the comments).
What I’m primarily arguing is that when you start a true freshman who hasn’t spent enough time in a system to fully learn it (see: Malcolm Gladwell on the importance of practice), the QB is going to struggle with making reads and with the timing of their throws. This results in inaccuracy and a low completion percentage/passer efficiency rating. RGIII and Kolb both ended up being pretty successful, record breaking air raid QB’s. Both of them had incredibly low completion percentages as true freshmen. Kevin Kolb only completed 61.1% of his passes, RGIII only completed 59.9%, and Jared Goff at Cal is only completing 61.4% of his. Baker Mayfield is completing 65.7% (albeit mostly against cupcakes and in garbage time against KSU) and Webb is completing 61.7%. Would the offense work better if either of them could complete more of their passes? For sure, as has been discussed, for an air raid system to be successful, a QB needs to complete nearly 70% of his passes. Is it reasonable to expect either of our true freshmen to complete more? No, not really. No other true freshman QB has ever done it, or really anything close.
There are compelling arguments for discounting comparisons between our QB’s and RGIII. For instance, the quality of our receiving corps is much better than what RGIII had at Baylor. This is true; it’s a lot easier to make reads when all of your receivers are open and you have a beast like Amaro who is open even when he is covered. Undoubtedly our QB’s completion percentages would be lower if every pass thrown to Amaro when he was covered resulted in an incompletion. If we had Baylor’s caliber of receivers in 2008, our qb’s completion percentage might be closer to what RGIII’s was, but, even if it were, being as good as a Heisman trophy winner was when he was a freshman is an awfully high bar to set (RGIII’s Baylor also had over a thousand yards rushing independent of RGIII’s own 800 rushing yards). For me, the things that would offset or explain away RGIII’s low completion percentage (tougher schedule, less quality in his receiving corps) are offset by other factors our quarterbacks face, like our complete lack of a competent running game and our patchwork offensive line, so for me, those things are a wash, and the comparison is a useful one if not a perfect one.
Regarding the second point, that either Mayfield or Webb is a bust, this line of thought seems to come from two things. The first, I think, is that as Tech fans, we simply aren’t used to evaluating true freshman (or even redshirt freshmen) quarterbacks. Because of this, we evaluate them against what we’re used to seeing from our redshirt sophomore or junior quarterbacks because that is our only basis of comparison. According to that standard, yeah, they would be huge busts and should be benched immediately. However, as I explain above, when compared to other true freshmen, they’re about where you would expect them to be. The second idea, which I will address below, seems to come from the aphorism "you can’t coach accuracy," which would doom both of their careers to sub optimum accuracy and subject us to bad quarterback play for the next four years.
I can’t say definitively that you can teach accuracy but the evidence seems to suggest that you can. Every Tech quarterback in the Air Raid era that has started at least two seasons has seen a substantial jump in their completion percentage in their second year. Graham Harrell completed 66.9% of his passes his first year and increased that to 71.8% his second year. Seth Doege completed 68.5% his first year as a starter and increased it to 70.2% as a senior (as a redshirt freshman he completed 62.3% of his passes). Even the much-maligned Taylor Potts increased his completion percentage from 65.7% his first year to 67% his second (also under a new coordinator in a new system).
Looking at other air raid quarterbacks, I noticed a similar trend. RGIII increased his completion percentage year by year from 59.9% to 65.2 (injury cut short his season), to 67%, and then finally to 72.4% his senior year. Case Keenum threw for a completion percentage of 68.5% as a redshirt freshman in limited snaps. In his first year as a full time starter he regressed to 67.4%, attempting about twice as many passes. In his third and fifth year (injury cut short his fourth year) he completed 70.3% and then 71% of his passes. Johnny Manziel completed an astounding 68% of his passes as a redshirt freshman and has managed to increase that number to 72% this year. From this data, it seems that not only does completion percentage increase with experience, the younger the quarterback is in his first year as a starter, the higher the increase in completion percentage will be the following year. (The one outlier here is Kevin Kolb, who had bad sophomore and junior seasons, but then had a much better senior year)
Even the poster boy for the claim that accuracy is determined by nature not nurture, Garrett Gilbert, went from completing 53% of his passes in his first year at SMU to 64.1% this year. As a side note, I do think Gilbert makes an interesting case for some QB’s just not having "it" at all, and I do think there is a line from below which a quarterback will not be able to improve to the point of being a quality, division one quarterback. That line, however, for a freshman seems to be in the 40 or low 50 percentage range rather than the 60’s where our qb’s are now.
Now looking at Webb, he makes an interesting case. If you remove the outlier, Texas State, where he completed a mere 44.2% of his passes on 43 attempts, he’s completed 62.5%, 72%, 62.3%, 63.4% and 65% of his passes in his five starts. Not great, I agree, but roughly on par with what Deoge did as a freshman (rs). This is interesting because many posters have said that in high school he had a 55% completion percentage. If it’s true, as some have said, that he looked extremely under-coached in high school, I think it’s extremely promising that he’s increased his completion percentage by 6%, against better competition, with only a few months of exposure to quality coaching. There isn’t any reason to think he wouldn’t similarly be able to increase his completion percentage to near 70% by his redshirt sophomore or true junior year. Regarding Mayfield, his completion percentage is already pretty salty, usually in the 65 to 66% range (77.3% against KSU!!). His main problem has been interceptions and the lack of passing touchdowns since week three. Reassuringly, touchdown to interception ratio tends to improve over time as well.
To summarize, I’d like to point out what I’m not arguing. I’m not arguing that either Mayfield or Webb should be the starter over the other. I’m also not arguing that either should start over Brewer, as I believe my arguments above pretty strongly indicate that Brewer, as a redshirt sophomore gives us a much better chance to win now than either of the freshmen. I’m also not arguing that either Mayfield or Webb are going to be great or even good quarterbacks. My only points are that both quarterbacks are performing as well as you can expect for true freshmen, they’ll both improve over time, and it’s too early to evaluate either one as a bust or a savior.
Thanks for reading my first fan post. Please feel free to quibble with, eviscerate, or totally debunk any or all of my points.