A Gator fan’s J.T. Daniels evaluation, and what Bill Belichick taught me about COVID-19

Embed from Getty Images

We’ve all heard it at this point.

“We would have won if J.T. Daniels had been able to play.”

That refrain – which completely ignores the fact that Justin Fields was eligible and playing for Ohio State at the time – has become common this offseason as Georgia Bulldog fans try to justify the 44-28 butt whooping they received from Florida in 2020.

After Georgia rushed out to a 14-0 lead the game wasn’t even close, with Florida ripping off 41 of the next 48 points, and really all 48 if you consider Trask threw a TD pass, just to the wrong team. Florida fans have taken great glee in tweeting out various “wheel route” plays that completely flummoxed Kirby Smart’s defense. And unless Daniels was playing outside linebacker, there was no way he was going to help Georgia win that game.

But Daniels did show promise in limited action at the end of 2020, so the question is whether his ascension as the starting QB will make a different in 2021.

For Kirby Smart’s sake, he better.

J.T. Daniels’ Profile

Daniels was a 5-star prospect coming out of Mater Dei High School in California in the 2018 class. He earned that ranking when you look at his high school stats.

These aren’t quite Joe Burrow high school numbers, but these are numbers that worry me as a Florida fan. In his sophomore and junior seasons, Daniels showed he could push the ball down the field (over 11 yards per attempt) and complete a high percentage of his passes (both over 71.8%).

Daniels then skipped his senior year of high school and enrolled at USC, where he ended up starting as a true Freshman and played respectably (QB Rating of 128.6, 14 TD, 10 INT, 7.4 yards per attempt). Basically, he was a slightly above average QB as a true freshman who should have been playing in high school.

He showed his limitations in that season at USC as well though. Daniels was an absolutely statue, running the ball 45 times for -149 yards, indicating he will hold onto the ball and take a sack. In fact, that’s how he got injured in the opener of his sophomore season against Fresno State, as he backpedaled trying to get away from a blitz.

Looking at his 247Sports profile is interesting, as they rated Daniels at a “9” for intangibles, pocket presence, delivery and accuracy, but dinged him for his arm strength and size. At 6’2” tall, he is a little bit small to see over the offensive line consistently.

Daniels at Georgia

It’s the last four games of the 2020 season that is really starting up the hype train in Athens for Daniels.

Against Mississippi State, South Carolina, Missouri and Cincinnati, he completed 67% of his throws for a 10.3 yard per attempt average. That’s elite QB play against three SEC teams and a team some people wanted to see in the playoff. In fact, if we look at Kyle Trask and Daniels against their common opponents (South Carolina and Missouri), who would you take?

QB #1 is Daniels, who outpaced Trask in yards per attempt and edged him out in QB rating. I’m not saying Daniels is Trask, but the point is that if you’re going to call Trask a Heisman-worthy player throughout last season, you have to acknowledge that Daniels played equally well against the exact same teams as Trask.

So what are Daniels’ odds of continuing at his 2020 pace? Actually, pretty good.

I went back and looked at all 5-star transfers since 2010. Some of them had significant playing time at their first stop (Jeff Driskel, Florida to Louisiana Tech). Some of them didn’t play at all (Gunner Kiel, Notre Dame to Cincinnati).

That meant that I couldn’t compare their previous stop to their new destination, but that’s not really what I was interested in. Instead, I wanted to look at their first four games at the school they transferred to (many against sub-standard competition, the criticism leveled against Daniels) and see whether their performance in their first four games transferred throughout the rest of the season.

The answer is that it mostly does. Of the 11 QBs to fit this profile, they had a QB rating of 148.8 in their first four games and a QB rating over their entire careers at the new school of 143.3. That means that they put up almost the exact same performance over the remaining games they played as their first four.

The maximum increase in QB rating from 4 games to career was 33.7 by Max Browne at Pitt, but Browne only threw 135 passes in his career at Pitt so that’s an extremely small sample size. The maximum decrease in QB rating from 4 games to career was 28.6 by a player Georgia fans know well, Jacob Eason.

Eason was an interesting case at Washington, as he lit up his FCS opponent (QB rating of 193.1), then struggled against Cal (98.7) but played well against Hawaii (199.6) and BYU (200.9). Not exactly a murderer’s row of defenses there. Eason also had very similar numbers to Daniels in his one year as starter (QB Rating = 120.2), but his peripheral stats (yds/att) were much worse.

So what does this mean for Daniels in 2021?

While I know this study is a small sample size, the takeaway is that even in a best-case scenario for Florida fans where he tanks as much as Browne did, Daniels is going to have a QB rating of around 150 (pretty close to Trask in 2019 and just slightly ahead of Kellen Mond in 2020). That’s enough to win a lot of games with the talent that exists at Georgia and is a major upgrade from Stetson Bennett (QB rating = 128.7).

If he approximates the average performance from the transfers I looked at, he’ll end up with a QB rating of 173, which would have ranked between Justin Fields and Spencer Rattler in 2020. That’s also in Jake Fromm territory (160.1 and 171.2 in 2017 and 2018), and we all saw how close Georgia was able to get with Fromm.

And if Daniels improves with a full offseason under his belt, you’re getting into Mac Jones/Kyler Murray territory and both of those guys made the playoffs.

I think it’s unlikely he improves significantly from last year for a few reasons. First, he apparently still hasn’t learned how to get rid of the ball and so injury is a big risk. At Georgia last season, Stetson Bennett – not exactly the most fleet-footed QB – had 24 rushes for 54 yards. Daniels had 10 for -71, indicating that he’s going to take a bunch of sacks.

Second, he excelled in the 11-20 yard throw range last season, completing 79% of his passes for an average gain of 14.8 yards. Considering his much lower percentage on shorter throws (68.9% on throws 0-10 yards from the line of scrimmage), I suspect that is an outlier. Even if that comes down 10% though, it’s still a lethal percentage at that range though, and so I’m not convinced he won’t make it up in other areas.

So let’s say Daniels is exactly the QB that he was over the last four games in 2020 (the most likely scenario). That means that Georgia will have a QB who is better than Jake Fromm was in either 2017 or 2018. That’s a scary proposition for an SEC East that is breaking in new QB’s (Emory Jones, Will Levis at UK) and new coaches (Heupel at UT and Beamer at USCe).

There isn’t any experience behind Daniels (redshirt freshman Carson Beck and true freshman Brock Vandagriff) and so if he keeps taking sacks that could derail the Bulldogs. But if Daniels can stay healthy, he is the main reason that the SEC Media just recently picked Georgia to win the division.

Pressure’s on Smart

So do I think Georgia wins the SEC East this year? I think that’s the most likely scenario.

But do I think Georgia finally gets over the hump and wins the National Championship this year? Well, that’s the question that Bulldog fans are asking themselves, and I’m not all that confident about that proposition.

Kirby Smart has an interesting reputation at this point. Because he has recruited at such a high level, and because he has been able to win an SEC title and reach the championship game, I don’t think it’s fair to label him as a poor on-field coach.

However, there is also no doubt that he has underperformed against teams with top-25 talent compared to some of his peers, and that he has been unable to beat the one person he needs to in order to fully scale the championship mountain: Nick Saban. Now, Saban has left plenty of decent coaches in his wake while at Alabama, but I do think it is telling that Dabo Swinney has been able to get him head-to-head a couple of times, even when an underdog.

The underperformance against talented teams usually isn’t an issue for Smart as Georgia normally has way more talent than its opposition, but when the talent level gets close, Smart has shown a proclivity for an ill-timed fake punt or going extremely conservative on the offensive side of the ball. Of course, Smart also hasn’t had a player who is as good as Daniels projects to be.

That’s means that 2021 is a huge opportunity for Georgia to finally take the throne. But it also means that there aren’t any more excuses if you’re Smart or a Bulldogs fan.

Georgia fans have heard the comparisons. They’ve seen the jokes and the memes. Kirby Smart in his first five seasons is 52-14 with one SEC Championship while Mark Richt went 52-13 and with one SEC title as well. The excuse has always been that the QB hasn’t been good enough to finish the job.

Smart is getting one year of J.T. Daniels before he goes off to the NFL. Daniels is likely going to put up big numbers. So if Georgia doesn’t beat Florida, that’s on Smart. If Georgia doesn’t beat ‘Bama, that’s on Smart. And if Georgia doesn’t win the playoff with good QB play, that’s on Smart too.

Smart gets paid a lot of money and has been given a ton of money to build the Georgia recruiting apparatus. Both the Gators and the Tide have brand new QBs. On the broader national stage, Ohio State and Clemson have new QBs as well.

Smart’s calling card is his defense and he’s built his reputation on it. If Daniels ends up being fantastic, there’s nowhere else to hide.

Experts and Mask Mandates

It may seem odd, but one of the hinge moments in my life was a late coaching decision in a November regular season NFL game.

The call – going for it on fourth-and-2 in his own territory – worked out terribly for Bill Belichick and the Patriots as Peyton Manning immediately drove the Colts down the field for the win. It seems obvious that you punt there, right? After the game, Tony Dungy and Rodney Harrison both skewered Belichick for his decision. These were people who played and coached the game at a high level and so the fact that they were incredulous that he didn’t punt seemed authoritative. After all, they were the experts.

Except, it seemed weird that Belichick – someone well known for his coaching prowess – would make a decision like that off-the-cuff. So I started looking into why he might have done it and came across a whole new world of football analytics. These people used statistical analyses to make their case, and their analysis of this decision made it very clear that Belichick was actually correct.

I was thinking about that when the CDC announced that they are recommending masks be worn by K-12 kids in schools this fall. These are experts – people that we should trust – making a recommendation that seems like common sense, but just seems a little bit off to me.

It’s not that I’ve been anti-mask. I’ve worn them throughout the pandemic. It’s also not that I’m a COVID-19 denier. I got it back in March and it is something I wouldn’t wish on my worst enemy. I also wrote about the coming pandemic in March of 2020 and looked at the possibility of wide-spread disease and a crisis (or surprising buoying) of faith that accompanied it.

But that nagging feeling that I felt after Belichick got criticized for going for it on fourth-and-2 was sitting there in the back of my mind that the experts are missing something with these new mandates. That was particularly true because when I started looking at the data I saw three things:

1. Florida – which famously has had limited lockdowns and limited mask mandates – has a much older population than California, which has had large-scale lockdowns and mask mandates. (1,2)
2. Florida has had a much higher percentage of its COVID-19 cases (14%) diagnosed in people 65+ than California has (10.4%). (3,4,5)
3. The death rates between Florida (1,811 per million as of 7/28/2021) is really close to California (1,629 per million) when just based on that age data, you would expect Florida’s death rate to be much higher than California’s. (6)

So what happened? I’m not an expert, and I can’t prove my theory, but I do have a hypothesis.

If we look at the percent of COVID-19 cases for California and Florida plotted against age (these are not perfect comparisons because the states reports different age brackets), California has a significantly higher portion of its younger population that has tested positive for COVID-19. The chart also says that California has been extraordinarily efficient at keeping older people from getting the disease.

Again, you would think that is a good thing for California, as younger people tend to recover at much higher rates. That would also make you think that the death rates in Florida should be way higher. Instead, what you see is that in total numbers and percentages, California has a similar death toll than Florida and a much higher death toll for people aged 50-64 than Florida. And it isn’t particularly close.

As of July 28, 2011, California had seen 12,685 COVID-19 deaths in that age bracket (50-64). Florida had seen 4,190 deaths between the ages of 55-64 and 6,317 if you extend that to 45-64. That means that even if you extend Florida’s age ranges down an extra five years, deaths in that age range account for 17% of the total deaths in the state compared to almost 20% for California.

These are younger people being infected who should be recovering at higher rates. The overall infection rates in Florida (120,929 per million) are considerably higher than California (99,850 per million). Yet Californians are dying at higher rates than Floridians. So what is going on?

There is a really interesting book I’ve been reading recently called Clumsy Solutions for a Complex World, which is a collection of essays from researchers in various fields who propose counter-intuitive solutions to world problems.

One chapter took a look at data following the implementation of mandatory seatbelt laws in various countries compared to those that did not pass those laws and came to the following conclusion: “Collectively, the group of countries that had not passed seat belt laws experienced a greater decrease than the group that had passed laws.”

But how can this be? Seatbelts work, right?

Of course they do. But that wasn’t what was happening here. Instead, the authors postulated that what was happening was that humans have an internal risk calculus that they are constantly adjusting and seatbelt usage shifted that calculus.

Just as you take smaller steps when up on a cliff, and you’re much more apt to take more aggressive steps if tethered securely, drivers as a whole decide to take more risks when they have seatbelts to protect them. The result is more death because the severity of the crashes increases as well as the number.

So what does this have to do with California? Well, I suspect the higher death rate in the younger population has to do with masks. It’s not that masks don’t work (though I think we can debate cloth vs. surgical vs. N95 masks). Instead, it’s that they induce people to take more risks when they have the mask to protect them.

The result is that a 50-year old woman in Florida with pre-existing health conditions stays away from people to protect herself while a 50-year old woman with the same health issues in California may be more apt to rely on the mask for protection. After all, the government wouldn’t mandate them if they didn’t work, right?

That would be fine if masks were perfect, but they are not. Just as a seatbelt can’t protect you from an extremely energetic collision, a mask can’t protect you from exposure to someone with COVID-19 for an extended time. That’s why the guidelines from the CDC have always included 6-feet of separation and time limits for exposure.

My objection to mask mandates is not the conventional ones that I’ve heard from people who are against the mandates (restriction of individual freedoms, it’s not a child’s job to protect people who choose not to vaccinate, etc.), though admittedly, I am sympathetic to those arguments as well.

Instead, my objection is that masks may be objectively harmful within a broader group, not because they don’t provide some level of protection, but because the level of protection they provide is not nearly the level of protection that people perceive that they provide.

The argument from the people pushing for mandates is essentially “what harm can it do?” The problem is that after looking at the data I think the answer may be “a lot.”

I’m not telling you masks don’t work. But I am telling you is that it is important that you teach your loved ones who are wearing masks that the mask is not a substitute for actually taking care of yourself based on your individual risk factors, vaccination status, ability to distance and ability to practice good hygiene.

I’m also not telling you to ignore the experts. But I am telling you that you need to take the time to discern whether experts have used data-based decision making to come up with their recommendations or are just leaning into their credentials as reasons for you to believe them.

Perhaps my conclusions are wrong. I’m open to a different interpretation of the data or that some factor I haven’t considered is actually responsible for the difference I’m seeing in the numbers. Single-factor analysis is always dangerous, but that’s actually why I think my argument makes sense.

The reality is though that the difference in the numbers is there and there’s only one way it’s going to go away: vaccinations. It certainly isn’t going to go away with a mandatory mask mandate.

The experts in California already tried that.

 

Sources:

1. https://www.infoplease.com/us/census/florida/demographic-statistics
2. https://www.infoplease.com/us/census/california/demographic-statistics
3. http://ww11.doh.state.fl.us/comm/_partners/covid19_report_archive/cases-monitoring-and-pui-information/county report/county_reports_latest.pdf
4. https://www.cdph.ca.gov/Programs/CID/DCDC/Pages/COVID-19/COVID-19-Cases-by-Age-Group.aspx
5. https://covid19.ca.gov/state-dashboard/
6. https://www.worldometers.info/coronavirus/country/us/