Blake Bortles did what good second-year quarterbacks do.

He showed improvement.

Bortles saw his Adjusted Net Yards per Attempt increase by 2.28 from 2014 to 2015, his Quarterback Rating jump from 69.5 to 88.2, and his Tom Brady threw more touchdowns than Bortles, who tied Cam Newton, Carson Palmer, and Eli Manning for second. But only five quarterbacks threw more times than Bortles did, including Brady and Manning. More passing opportunities, obviously, means more chances to score touchdowns. Right? Right.

That's simple enough, but things get deeper than touchdown rates, or touchdowns divided by the number of passes thrown.

And that's because not all opportunity is created equal, either. What if 90% of Bortles' throws were on the Jaguars' side of the field? What if Brady threw 30 touchdowns from within his opponents' 10-yard line? Clearly a 10-yard touchdown is easier to score than a 60-yard one.

Well, this is all part of why the number of touchdowns Bortles threw in 2015 is being a bit overblown. 

This past season, Bortles led the NFL in red zone attempts with 96. Aaron Rodgers was second with 94, while Manning (Eli) and Brady were at 91 and 87, respectively.

Those are prime opportunities for Bortles to score touchdowns.

Within the 15-, 10- and 5-yard lines, things weren't any different. 

Quite simply, no quarterback in the NFL had better scoring opportunities in 2015 than Bortles did. And, as you can see in the chart above, his opportunities were a lot greater in terms of volume than even the 10th-ranked passers within the field position parameters.

Opportunity is big for passing touchdowns, and Bortles had plenty of it in 2015.

Regression Coming

You often hear the phrase "regress to the mean" in numbers-driven analysis, and the idea states that if a variable is extreme during its first measurement, it will tend to be closer to the average during its second measurement.

In other words, Bortles' touchdown number was a little extreme this season (only 20 quarterbacks have ever thrown 35 or more touchdowns in a season), so we'd naturally expect it to move closer to the average next season.

This isn't something that's only credible with Bortles, either. At a quick glance, we should probably expect players like Eli Manning or Cam Newton to regress a bit in the touchdown column in 2016.

And based on history, that's what's generally happened.

Since 1984 -- which was Dan Marino's historic 48-touchdown season -- we've had 31 instances where a quarterback has thrown 35 or more touchdowns. (I mentioned earlier that Bortles is one of 20 to ever hit the mark, but that's because quarterbacks have done it multiple times.)

Of the 27 seasons where we have next-season data -- four quarterbacks threw 35 touchdowns this year -- only three were able to throw more touchdowns the following year: Peyton Manning from 2012 to 2013 (18 more), Tom Brady from 2010 to 2011 (3 more), and Brett Favre from 1995 to 1996 (1 more).

Among these 27 occurrences, the average drop off from one year to the next among players who were healthy and playing the following season is over 10 touchdowns.

Now, because the limit set here is 35 touchdowns, we shouldn't automatically assume Bortles will see that big of a drop off in 2016. When you throw, say, 45 touchdowns in a season, you're bound to throw a lot less than someone who tossed 35 -- it's easier to throw fewer touchdowns when you threw for so many the year prior.

But there's even more to all this regression talk. 

Since 2011, the average touchdown rate (touchdowns per attempt) has been 4.55%. In 2015, Bortles' rate was much higher, at 5.78%.

We've seen much higher rates than that in NFL history, but Bortles is definitely flirting with the top tier with such a rate. Look at it this way: in NFL history, we've seen 557 seasons where a quarterback had 200 or more pass attempts, and just 63 of these passers had a touchdown rate that was better than Bortles' 2015. Hell, only 140 of these passers saw a touchdown rate above 5.00%.

What's unsurprising to math nerds is that, of the quarterbacks with high touchdown rates in one season, the vast majority of them see lower touchdown rates the following year.

First off, the reason the "instances" number doesn't equate to 140 in the chart above is because not every quarterback played or threw a touchdown the year after throwing 200 or more times with a 5.00% touchdown rate. And, of course, there are passers who hit the mark in 2015 and haven't played yet in 2016. That left us with 113 quarterback seasons.

Naturally, you'd expect the "6.01% and Above" group to see massive drop-offs from one year to the next in touchdown rate. That's partially because it includes hugely anomalous seasons, but also because reaching the 6.01% touchdown rate mark is tough to do to begin with.

Bortles' group -- the 5.76% to 6.00% range -- has seen just 4 of the 20 quarterbacks hit a better rate the following year. In other words, regardless of volume, we would expect Bortles to throw fewer touchdowns per pass attempt next season. Because of regression.

Now, something I've yet to even mention is that Bortles wasn't exactly a fantastic passer in 2015. Did he show improvement? Definitely. Was he a top-10 quarterback? Absolutely not.

There are plenty of numbers that suggest Bortles was an average NFL quarterback this season. I don't even think Jaguars' fans would disagree. But that's also just another reason to think his touchdown numbers will regress in 2016.

Take a look at the quarterbacks over the last five years who have thrown at least 35 touchdown passes in a season, and compare their Passing Net Expected Points per drop back averages to Bortles'.

In the rank column above, we're looking at the quarterbacks' Passing NEP per drop back ranking among all 200-plus attempt passers in that given season. In other words, Bortles ranked 23rd in per drop back efficiency in 2015, which isn't very good.

And while it may be a tad unfair to group Bortles with quarterbacks who threw 40, 45 and 55 touchdown passes, even if you compare his efficiency numbers to quarterbacks who threw a similar number of touchdowns, it's really not close.

Clearly a passer who's not very efficient will have a hard time repeating strong touchdown numbers, no?

Our Net Expected Points metric isn't the only statistic that says Bortles' season was a bit crazy, either. According to Rich Hribar's study over at rotoViz.com that compares yards per attempt to passing touchdown percentage, Bortles' numbers dictated a touchdown finish in the mid-to-upper 20s, rather than the 35 touchdown passes he actually threw.

Looking Ahead to 2016

It's possible that Bortles gets to 35 touchdowns again next year, but it would call for just as pass-heavy of an approach close to the goal line for Jacksonville, a lack of improvement in the team's running game, a defense that continues to call for shootouts and, of course, a little luck.

But even if Bortles doesn't come close to the mark, it's important to remember that it doesn't mean a lack of improvement. Plenty of Bortles' opportunities came because he was inefficient to begin with -- he put his offense in poor situations early in the game, he turned the ball over, or he took a careless sack. That combined with one of the worst defenses in the league creates a need to throw the ball, giving a quarterback more overall opportunities to throw touchdowns.

Better efficiency may lead to fewer chances, but fewer chances may also mean a better Jacksonville Jaguars team.

Bortles ended up as fantasy football's fourth-best quarterback last season, all while capturing more top-12, QB1 performances than everyone not named Tom Brady. And it was, for all intents and purposes, because of his touchdown passes -- touchdown passes that will almost certainly not be there in 2016.

So when you see Bortles sitting there in the late-fifth or early-sixth round in August -- which is what current average draft position dictates -- don't forget about regression. 

Don't forget that improvement doesn't simply come in the touchdown column.