Grading on the curve: Wilton Speight
With another less than stellar performance, there have been numerous hot takes and much hand wringing on whether Wilton Speight is any good. Personally, I've contended that his performance has been good, if a little sporadic. But, I also have my reservations about his ability to win games for us if/when we come up against a team that can score a lot of points (Ohio).
So, I decided to take a look at how he stacks up against other power 5 QBs. Using ESPN's stats I created a spreadsheet where I could rank QBs based on several factors. To qualify, the a player must attempt, on average, 14 passes per number of games the team has played (to weed out backups, garbage time, etc). The factors I chose were:
- Completion percentage
- Yards per attempt
- TD per attempt
- INT per attempt
- INT/TD ratio
I graded based on percentiles, according to the following scale:
100% > A > 85% > B > 65% > C > 35% > D > 15% > F
Before we rank Speight against his peers, here is his stat line:
63%, 7.5 YPA (1094 yards on 146 attempts), 6.8% TD/ATT (10TDs), 1.4% INT/ATT (2 INTs), 20% INT/TD
Completion Percentage
Speight was 26th of 66th and in the 61% percentile, earning him a C+
This generally is in line with my gut feel. He is just above middling in his accuracy, offsetting good throws with the occasional air mailing or bad read.
The top QB in this category was Mitch Trubisky, UNC (76%).
Yards Per Attempt
At 7.5 yards per attempt, Speight was 37th of 66th, just below average at 46% percentile. He gets a C here.
This was a bit of a surprise. He's hit quite a few shots downfield including a shredding of UCF. The last few games must have really brought him down to earth.
The winner of this category was Patrick Mahomes II, Texas Tech (10.7yds/attempt).
TD per attempt
Speight placed 23rd in this category with 10TDs on 146 attempts, placing him in the 66% percentile and earning him a B-.
Again, no real surprise here. He runs the offense, he isn't the offense.
Bolstered by a recent torching of Stanford, Jake Browning, Washington, has 17TDs on just 116 attempts(!), taking this category for the Huskies.
INT per attempt
Speight's INT rate was 19th with 2 INTs on 146 attempts. This puts him at the 72% percentile and gives him a B-.
I know people will point to passes he's thrown that should have been picked, but every QB has those. Also his second actual pick needed a pretty (un)lucky bounce to end up in the defense's hands.
Winner here is Mitch Trubisky, UNC, who hasn't throw a pick on 175 attempts. Other QBs with no INTs are Sefo Liufau (Colorado) and Ryan Finley (NC State).
INT/TD ratio
Speight comes in again at 15th in this category, but percentile wise improves to 83%. There are some awful ratios towards to bottom, helping out with this. He gets a B+
Interestingly, Speight's INT/TD is quite good despite being pretty middling in the individual TD and INT categories. It suggests a steady game manager profile where he doesn't light up defenses, but limits mistakes as well. In the end he's made the plays to win and taken care of the football, which is in line with this stat.
Mitch Trubisky, UNC, again comes out on top with 0 INTs and 13 TDs.
Conclusions, etc
Comp % | YPA | TD/Att | INT/Att | INT/TD | |
Stat | 63% | 7.5 | 6.80% | 1.40% | 20% |
Rank (out of 66) | 26th | 37th | 23rd | 19th | 15th |
Grade | C+ | C | B- | B- | B+ |
The profile here is definitely in the dreaded game manager mold. Speight ends up slightly above average overall (B-, C+ borderline), doing well in the protect the ball categories and not as well in the generating offense categories. Paired with our defense, that's probably just fine for all our remaining games save one. Depending on how optimistic you are about Don Brown and his ability to stop elite spread teams, I still think he, and the entire offense, need to improve quite a bit to get over that last hump and beat Ohio.
EDIT: Hat tip to WhoopinStick, who pointed out Wilton should get some credit for being solid in all categories. despite not excelling at any. Instead of averaging his grades, I gave everyone points based on ranking in each category (1 for first, 2 for second, etc) and then totaled points in all categories. The QBs were then re-ranked on this total, lowest point total first. Speight comes out at 17th of 66, sharply better than with the other methodology. In this scheme, he gets a solid B.
Other notes:
Mitch Trubisky is quietly having a statistically impressive season at UNC. With Pitt, Georgia, and FSU on the schedule, this wasn't completely a pushover schedule either. He's 133 of 175 (76%) for 1711 yards, 13 TDs and 0 INTs.
Sefo Liufau did quite well for himself before getting hurt as well: 54 of 76 (71.1%) for 768 yards, 6 TDs and 0 INTs in just 2.75 games played. I know the back up has been lighting it up as well, but if they get Liufau back soon, Colorado will make a run at the PAC 12 (they might anyway).
Among BIG10 QBs, Speight is 2nd (behind Barrett, Ohio) in completion percentage, 2nd in INTs per attempt (behind Wes Lunt, Illinois), and 3rd (Barrett, Lunt) in INT/TD. He's middling plus in the other two departments relative to his BIG10 peers.
October 3rd, 2016 at 7:25 PM ^
MIddling, 32 of 66
October 3rd, 2016 at 6:42 PM ^
http://www.cfbstats.com/2016/leader/national/player/split01/category02/…
Speight's ranked 50th nationally in QB rating. Rudock was in the 90's for the first half of the season last year before he rocketted up the list. Connor O'Tyler from STAEE is ranked 56th. I'd much rather have Speight than O'Tyler.
October 3rd, 2016 at 8:46 PM ^
October 3rd, 2016 at 6:44 PM ^
On the heels of one of the most remarkable seasons in regards to quarterback progression; I think we might have unreal expectations for our new starting quarterback.
October 3rd, 2016 at 6:50 PM ^
WRs and TEs did a poor job separating vs. Wiscy DBs all game. Wilton has guys in his face all game as well.
October 3rd, 2016 at 6:52 PM ^
For the effort involved, I still don't know if your work produced anything meaningful. Based on your stats - and qbing goes far beyond stats - the best qb in the nation resides in Chapel Hill. Because there are so many factors involved and so many different types of qbs, I don't know how you can gain meaningful information by compiling data respective to only one aspect pof the position; passing.
I watched two of the top four qbs in the nation on Saturday evening and I didn't need a spread sheet to tell me that. When the game was on the line, they both stepped up big time. I also know the best qb in the BIG resides in Columbus. I am certain this is a consideration in recruitment. it speaks to leadership and athleticism, but might not be found on a spreadsheet. There is no mention of extending plays, caliber of receivers, OL protection, classification on dbs dls and overall defenses they've faced, etc.
Another major consideration is offensive philosophy and style. The spread teams in the Big 12 pile up huge numbers as to YPA, TDss, et. al., but they don't even pretend to have a balanced offense. 63% might be perfect - and it is damn good - for the offense J.H.and staff desire. Given that he's played all of five games and one drive last season, the most important stat: Ws vs Ls., I believe are higher for Wilton than for the other leaders on your sheet.
Hell, none of Bo's qbs won more than Dennis Franklin, but Leach, although a decent passer by Bo's definition and 'ABSOLUTE BUCKEYE KILLER," will be forever considered the better qb. Probably a fun exercise but attempting to grade any position on a handful of measureables is bound to produce questionable results, unless your goal was merely to find the most efficient passer. All areas of the offense will have to improve in order to beat OSU and their all-time passing and scoring leader. That's a given and if you've watched closely you'll see the parts coming together and improving each week.
We' may or may not be ready this season, but we'll give them a much better battle than last season.
October 4th, 2016 at 8:22 AM ^
October 3rd, 2016 at 8:06 PM ^
October 3rd, 2016 at 8:44 PM ^
October 3rd, 2016 at 10:06 PM ^
October 3rd, 2016 at 6:53 PM ^
Mobility, arm strength, a few ill-advised throws. However, I have complete confidence he is the best available option if Harbaugh says he is. Just hope he continues to improve and continues to be put in position to succeed.
October 3rd, 2016 at 7:11 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 3rd, 2016 at 7:28 PM ^
Offense had some plush starting field position, though.
osu/wisky will tell us plenty
October 3rd, 2016 at 7:51 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 3rd, 2016 at 7:11 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 3rd, 2016 at 7:51 PM ^
Especially when you know that line is paired with an elite D and running backs that looked pretty competent last game.
October 4th, 2016 at 12:02 AM ^
October 3rd, 2016 at 7:13 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 3rd, 2016 at 7:24 PM ^
October 3rd, 2016 at 7:52 PM ^
October 3rd, 2016 at 10:12 PM ^
A high complete percentage is not really a good measure without knowing the nature of the passes. As you point out 80% of 0-5 yard passes really isn't that impressive. But 65-70% of throws going 15-25 yards is a pretty good passing attack.
The term Game Manager is very perjorative. Some teams have enough weapons that the QB does not have to have a great game to win it - it is the best strategy therefore to avoid INTs and focus on moving the ball and sustaining drives.
October 3rd, 2016 at 7:52 PM ^
I watched the BTN replay of the MSU game and the two QBs seem pretty similar to me in a lot of ways. O'Connor is obviously more mobile (he should scramble more than he has), both can throw a decent ball, but both can hold on to the ball for too long and throw into double or triple coverage sometimes. Speight has just been luckier so far to only have 2 picks.
Yet MSU fans are calling for a QB switch as though O'Connor's the sole reason they've lost twice. It's all a matter of perspective and mediocre QBs on a 5-0 team are given a lot more slack than mediocre QBs on a 2-2 team.
October 3rd, 2016 at 8:40 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 3rd, 2016 at 8:27 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 3rd, 2016 at 8:54 PM ^
October 4th, 2016 at 6:37 AM ^
October 4th, 2016 at 10:40 AM ^
A "Game Manager" is a Quarterback that wins a lot but the author still thinks he stinks.
October 4th, 2016 at 11:00 AM ^
Show me where I say he stinks. I use the "game manager" term pretty tongue in cheek, apologies if that doesn't come across in text. For the record, I think he's solid and done a nice job protecting the ball. I still think he needs to be a little more productive in games where we need to find points against a good defense.
October 4th, 2016 at 3:20 PM ^
I was talking to Brandon, answering his question about Nappa's post. Nobody was talking to or about you.
But since you're here, I think Speight has done a great job. He's got a natural ability to avoid the pass rush and extend plays. He's missed throws and forced some iffy passes, but he doesn't look panicked. His bad games have come against increasingly stiff defenses in Colorado and Wisconsin. Now we're going to watch him (weather permitting) torch Rutgers and Illinois and everyone will be going ga-ga about the Harbaugh Effect heading into MSU.
Dude has Sophomore eligibility and 5 career starts. The game is only going to slow down and he's only going to get better.
While I'm at it, it's funny how people just gloss over Jabrill Peppers blowing deep safety responsibility and giving up a TD to Colorado. No problem JP! Love you! Heisman!
And in general, nobody holds it against Mike McCray or Dymonte Thomas for giving up a UCF 90 yard TD. Or Gedeon for getting beat on the wheel - "Ah! Well, take the good with the bad. Ben's been great!"
Newsome blew a blitz pick-up on the first play of the game yesterday and let a guy in clean as a whistle, untouched, and only Speight's awareness and ability to evade a free rusher one-on-one turned that from a loss of 7 into a gain of 2. --- nobody mentions that.
But watch him underthrow a 'who-cares' ball in a PSU blowout or take an ill-advised shot to the corner of the endzone against UW that falls harmlessly incomplete and all-of-a-sudden he's a poop, 3-star, limited game manager with no future.
I f***ing love it.
Not that you said any of this, reshp1. You didn't. You just got me going and I'm taking the opportunity to vent against all of the nonsense.
I take exception to "Game Manager", because what people generally mean by that is "He stinks, but we'll win with defense".
October 4th, 2016 at 6:47 AM ^
otherwise wouldn't he be starting John instead of Wilton? Obviously, through all the spring, summer and fall practices, Harbaugh doesn't see it. And although Peters would appear to have the highest ceiling of the three, he no doubt needs more seasoning in the leadership and decision-making areas, hence the redshirt. Next season's QB competition will be most interesting, indeed.
October 3rd, 2016 at 10:41 PM ^
October 3rd, 2016 at 10:50 PM ^
looks like a 3 star Jr QB. Rudock had more experience but was playing worse at the same point in the season. I just hope Jim can do to Wilton that he did to Rudock. Though, I'm starting to think Rudock was a more accurate QB but just was hampered since he had to take time to learn the play book. Until Wilton improves on his down field accuracy he should throw more out passes to Butt or look for some slant plays with 8 to 12 yard passes.
Our defense is great but I find it a stretch to hold OSU under 24 points, so we will need some offense when playing in Columbus, and much of that will fall on Wilton's arm! Overall, we are kind of out of phase again. We have an elite defense but a so-so offense and I'm not sure we will have the talent on the OL or at QB to beat an opposing elite team. But eventually Harbaugh will get us there in a few more years when we will be elite on both sides of the ball.
October 4th, 2016 at 7:53 AM ^
I like this thread a lot. I hope you can replicate it on a week to week basis. I'm sure Wilton would have graded much higher after week 3. Right now, after two sub-par games, he's not looking as good. Bottom line is that a greater sample size will tell us more, but I like your methodology and analysis.
Please call them Ohio State though.
October 4th, 2016 at 11:06 AM ^
How about THE Ohio State University?
October 4th, 2016 at 8:02 AM ^
Sent from MGoBlog HD for iPhone & iPad
October 4th, 2016 at 10:21 AM ^
Great post. I love creative use of stats.
October 4th, 2016 at 11:02 AM ^
Speight has the following 12-game stat projection: 2,625 yards, 24 touchdowns and 5 interceptions -- with a completion rate of 63 percent.
October 4th, 2016 at 11:13 AM ^
and they don't always tell the story. I too have some concerns about Speight, but the team is winning and when it was needed he lead them to a game winning TD on Saturday. So for now he's doing a good job and I believe the team has confidence in him and that too is just as important.
I hope he continues to improve as the year goes on and is clicking on all cylinders when we play OSU.
October 4th, 2016 at 1:09 PM ^
Certainly there is room for improvement.
Consider the last 4 oppoents UCF, Colorado, Penn State, and Wisconsin to date have better to far better (Colorado and Wisconsin and even UCF) Defensive statistics against the pass than average. From the NCAA rankings based on Pass defense efficiency rankings Colorado is 3rd Nationally, Wisconsin is 24th, UCF is 27th and Penn State is 41st. For reference, Ohio State is 1st, Michigan is 7th, Alabama is 25th, and Michigan State is 73rd.
Should the question be more about how well does Speight do compared to other QBs who have played those teams? Certainly better than Tyler O'Connor in his matchup vs. Wisconsin.