The CardBoard
The best player in college football: The evidence - Printable Version

+- The CardBoard (https://thecardboard.org/board)
+-- Forum: C-House! (https://thecardboard.org/board/forum-4.html)
+--- Forum: The CardBoard (https://thecardboard.org/board/forum-5.html)
+--- Thread: The best player in college football: The evidence (/thread-11100.html)


The best player in college football: The evidence - Hulk01 - 12-16-2015

I get angrier about the Voldemort over time—and thought.

What started me thinking was kickoff returns.  I agree that adding every kickoff yard into a player’s total yards, and then comparing his total to players who do not return kicks, is not fair.  A merely average kickoff returner—a player who just easily could have been replaced by any of several others on his team—has no wins over replacement value, and shouldn’t get a bonus.

But stop there.  Let’s do that: Let’s assess yards gained per kickoff return over replacement value: the average NCAA kickoff return, and therefore the mythical average returner.

The average kickoff returner nets 21.09 yards.

And Christian?

28.94, or an added 7.85 yards per return, which experts say is worth 0.56 points per return.

Now take those bonus yards that Christian produces every year, to improve his teams’ field position and likelihood of scoring, and not being scored upon:
7.85 surplus yards per return x 36 returns :
273 yards
11 yards per game = 0.79 points per game.

Now, let’s add those 273 bonus yards into Christian’s running, receiving, and passing yards:
363 touches
2699 yards
7.7 yards per play

And here are Derrick Henry’s totals”
340 touches
2083 yards
6.1 yards per play.

Christian nets almost 50 yards more per game, and over 1.5 yards more per play, than the decisive winner, by a margin of over 40%, of the Voldemort Trophy.

Now, Henry scored more actual touchdowns.  First, he’s his team’s short yardage back.  But was Derrick Henry uniquely effective in short yardage situations?  Is Henry particularly valuable because, like Remound Wright and perhaps unlike Christian McCaffrey, he is near automatic on third and short? 

Did Derrick Henry deserve added credit because he is uniquely effective at converting on short yardage situations?

Well, let’s look:
Conversions of third and 1-3 yards, 2015
Derrick Henry: 13-22, 59%
Remound Wright:  19-27, 70.4%
Christian McCaffrey: 12-16, 75%

Was is those rugged SEC defenses that account for Henry’s 59% conversion rate?
If it was, how do you account for Tennessee’s Jalen Hurd, who was 14-20 on third and short, and Auburn’s Peyton Barber, who was 19-27?  Or, running behind Vanderbilt’s front line, Ralph Webb, who converted on 16 of 30?

Well, I’ll be darned: When defenses know Henry is coming, he’s pretty good.  But he’s not Remound Wright good or Christian McCaffrey good.  He’s only Derrick Henry good.

And that never should have been considered good enough to win the Voldemort.

—

But the story is that Henry won the Voldemort because as a pure running back only—forget all those other things Mac does—he was superior.

But was Henry a more effective runner, run by run, quarter by quarter, than McCaffrey?

Ist quarter rushing:
McCaffrey 603 yards 5.91 yards per carry
Henry 550 yards 5.91 yards per carry

2nd quarter rushing
McCaffrey 440 yards 6.38 yards per carry
Henry 433 yards 6.28 yards per carry

3rd quarter rushing
McCaffrey 519 yards 5.83 yards per carry
Henry 482 yards 5.24 yards per carry

After three quarters:
McCaffrey 260 carries 1562 yards 6.0 yards per carry
Henry 257 carries 1465 yards 5.7 yards per carry

Clear edge, McCaffrey

And then what happens?

Alabama leaves Henry in to pound on worn-down defensive lines, while Stanford often pulls McCaffrey off the field, to keep him rested and run out the clock,.

4th quarter:
McCaffrey 59 carries 285 yards
Henry 85 carries 521 yards

That’s it. Alabama kept Henry in, gave him 26 more carries, on which he averaged 6.13 yards per carry—almost half a yard more than he averaged over the first three quarters. Well, of course he did. It’s not as if he got stronger; the very idea, often offered by commentator is ridiculous; it's like suggesting that after 20 reps on a bench press, a great athlete gets a surge of energy.

Henry didn’t get stronger; the players trying to stop him, after over 50 snaps, got worn down, and Henry took advantage of that to roll up 226 more yards than Mac, to make it appear that might be a more effective runner than McCaffrey.

Which he was not—by any measure.

—
But let’s keep searching. 

Let’s see if we can somehow, somewhere, find a case for Derrick Henry being a more outstanding and valuable football player than Christian McCaffrey.

And for our next step, we go to what a growing number of pro and college coaches believe is, with turnovers, one of the two key statistics in football: Explosive plays, meaning runs from scrimmage over 10 yards and receptions over 15 yards. 

Who is the more explosive of these two players?

Explosive plays, touches % of touches resulting in explosive plays:
Derrick Henry    57    339 touches  16.8%
C’n McCaffrey    61    322 touches  18.9%

And no, let’s go back to a totally different measure—short yardage—and see if Henry’s ability to pound out a first down on a third and short (1-3 yards) compensates for the fact that McCaffrey made an explosive play on ever 5.3 touches, while Henry managed one ever 5.9 touches:

[b]Conversions of third and 1-3 yards:[/b]
Derrick Henry: 13-22, 59%
Remound Wright:  19-27, 70.4%
Christian McCaffrey: 12-16, 75%[font=times new roman]

Gosh, and Henry just looks like a lethal weapon on short yardage, doesn’t he?  He just so passes the eyeball test for a running back, it appears, that no one thought to conduct any other test to see just how effective he is—or isn’t.

And the answer is, Henry is very good—and by no reasonable measure as valuable, explosive, or effective a football player as Christian McCaffrey—even if you don’t factor in McCaffrey’s punt returns (an added 5.2 yards per game) and two touchdown passes (and added 3.0 yards per game).

Do Derrick Henry was not as explosive as McCaffrey.
He was not as effective in short yardage.
He averaged fewer yards per touch and per game.
He was dramatically less effective catching passes.
He didn't return on kickoff for a touchdown, as McCaffrey did, or return another 96 yards for a near-touchdown, and he didn't force any teams to concede Stanford the 30 yard line, or even the 35, to avoid even kicking the ball to him.
He didn't throw a touchdown pass, much less the two that McCaffrey threw.
He didn't return punts at all, much less for almost 60 yards on the season.

And yet he won the Voldemort Trophy--decisively.  Can anyone think of a good reason why?

Re: The best player in college football: The evidence - OutsiderFan - 12-16-2015

You forget to throw in negative plays - runs for loss.  I believe Alabama was very poor on this count.  Not sure about Stanford.

Alabama was 12-1.
Stanford didn't start promoting CMac soon enough. 
And more importantly, Stanford didn't present the data as you, Hulk.

No way is Stanford going to put out numbers that compare one of its players to another.  They would never try to denigrate another player from another school.  It's up to an unbiased 3rd party to do the figuring and make it public.  Perhaps there is room for a disgruntled Stanford fan to start a new website called "heismantrophyvotingfordummies" that distills all the data and comparisons between players for voters too lazy to do the work themselves?

There is also this narrative that CMac isn't going to be a RB in the NFL. 

But one question I have is if Henry averages about .1 YPC more than CMac per NCAA stats, how is it Henry never had a quarter in which he rushed for a greater average than CMac?


Re: The best player in college football: The evidence - CTcard - 12-16-2015

(12-16-2015, 05:19 AM)Hulk01 link Wrote:4th quarter:
McCaffrey 59 carries 285 yards
Henry 85 carries 521 yards

Henry didn’t get stronger; the players trying to stop him, after over 50 snaps, got worn down, and Henry took advantage of that to roll up 226 more yards than Mac, to make it appear that might be a more effective runner than McCaffrey.

OR, Henry was the reason those players trying to stop him got worn down.

4th quarter per carry:
McCaffrey: 4.83 yds/carry
Henry: 6.13 yds/carry

-------------------------------
Look, McCaffrey had better stats overall than Henry, had more "wow!" explosive plays, and did more for his team. There's is no reason to delve into minutiae of stats to find that.

The pro-Henry argument is simple: he gained the most rushing yards in the country against the most difficult schedule in the country.
[well#4 SOS by Sagarin, but toughest of the contenders]
[true that is overall SOS, but I don't know a straight ranking of offensive SOS (opponents' defenses) that I trust].

It's not an argument that should win out - but it's simple.

Re: The best player in college football: The evidence - stupac2 - 12-16-2015

(12-16-2015, 06:26 AM)OutsiderFan link Wrote:Perhaps there is room for a disgruntled Stanford fan to start a new website called "heismantrophyvotingfordummies" that distills all the data and comparisons between players for voters too lazy to do the work themselves?

That's actually not a bad idea. As far as I can tell there's not really anyone out there who's trying to make statistical cases for all the front-runners. And like what I found with the yardage over expected analysis, the stats speak for themselves.

Re: The best player in college football: The evidence - Rally - 12-16-2015

(12-16-2015, 09:19 AM)stupac2 link Wrote:[quote author=OutsiderFan link=topic=14084.msg148579#msg148579 date=1450272395]Perhaps there is room for a disgruntled Stanford fan to start a new website called "heismantrophyvotingfordummies" that distills all the data and comparisons between players for voters too lazy to do the work themselves?

That's actually not a bad idea. As far as I can tell there's not really anyone out there who's trying to make statistical cases for all the front-runners. And like what I found with the yardage over expected analysis, the stats speak for themselves.
[/quote]

This could be very successful if unbiased.  If it appears to be a Stanford site there will be no impact, but an unbiased site could grow into something significant, even a revenue generator.  There are people on this board who could do a fantastic job with this.  Not me for obvious reasons. 

Re: The best player in college football: The evidence - Hulk01 - 12-16-2015

There is a reason to delve: the voters assumed Henry was the nation's best big back, a guy who got Alabama yards whenever they needed them. 

And four times in ten, on third and short, Henry failed.  He failed to do what the voters assumed he did better than anyone.,

And the reason to delve is that while not all kickoff yards have value, the extra ones do; A 34 yard kickoff return, statistically, is worth a point.  Until voters realize this, they'll treat special teams as "just plays that happened to happen," as if good returns are simple reflections of randomness and not worth valuing.



Re: The best player in college football: The evidence - fullmetal - 12-16-2015

(12-16-2015, 09:19 AM)stupac2 link Wrote:[quote author=OutsiderFan link=topic=14084.msg148579#msg148579 date=1450272395]Perhaps there is room for a disgruntled Stanford fan to start a new website called "heismantrophyvotingfordummies" that distills all the data and comparisons between players for voters too lazy to do the work themselves?

That's actually not a bad idea. As far as I can tell there's not really anyone out there who's trying to make statistical cases for all the front-runners. And like what I found with the yardage over expected analysis, the stats speak for themselves.
[/quote]

You could give the user fields to populate for player A and player B and determine which would have more statistical value than the other.

Re: The best player in college football: The evidence - BostonCard - 12-16-2015

Here's an assessment from ProFootballFocus, who has McCaffrey as the top running back.

The pro-Henry case:

Quote:Against Power-5 competition, Henry and Stanford’s Christian McCaffrey are essentially neck-and-neck at the top in attempts and yards, but Henry’s 978 yards after contact leads the country, (well in front of second place Royce Freeman of Oregon) as do his 16 rushing touchdowns (just ahead of Ezekiel Elliot’s 15 for Ohio State).

Henry’s 21 breakaway runs (of 15-plus yards) and the 646 yards gained on them are both second to Freeman, while his impressive total of 56 missed tackles forced trails only the 58 that Freeman and UCLA’s Paul Perkins have produced.

Much of the above, of course, reflects the sheer volume of work Henry has received and some rate numbers may be more informative. His 3.81 yards after contact average, for example, has him tied for ninth in a category that Florida State’s Dalvin Cook leads with a 4.5 mark.

The case for McCaffrey:

Quote:Against FBS competition this year, Henry’s rushing grade of +23.9 is fifth, trailing McCaffrey, LSU’s Leonard Fournette, Utah’s Devontae Booker and Perkins. If only looking at games against Power-5 schools, Henry moves up to second, but still checks in behind McCaffrey’s comfortable lead.

Obviously depends on how you view the PFF grades.  It's subjective, but they appear to invest a lot of time into their grading.

https://www.profootballfocus.com/about/how-we-grade/

BC

Re: The best player in college football: The evidence - winflop - 12-16-2015

(12-16-2015, 05:19 AM)Hulk01 link Wrote:I get angrier about the F Stanford award over time—and thought.

I thought too - and I've decided that the system is inherently rigged against West Coast football and that I will no longer care about any individual awards unless a Stanford player wins (which I will never expect) and I won't care about the CFB Playoff until a Pac-12 team other than Low APR U is in it.  This has probably dropped my blood pressure by 20 points.

Re: The best player in college football: The evidence - BostonCard - 12-16-2015

Oregon was in it last year.

BC

Re: The best player in college football: The evidence - CornFed - 12-16-2015

(12-16-2015, 05:05 PM)Boston Card link Wrote:Oregon was in it last year.

BC

I would presume he's speaking prospectively, don't you suppose?