Ellis Quarterback Formula: Winston vs. Mariota

Going into the 2012 draft I heard that if Brandon Weeden was 23 instead of 28 he would have been the third pick. I also heard that if Russell Wilson was four inches taller he would have been the third pick. I bought into the hype and actually thought Weeden could have been an answer, because this was what I was told.

It was near the end of the 2013 NFL season. I had just finished watching Russell Wilson excel and spent a year watching Brandon Weeden flounder and thought it shouldn’t be this hard to find a solid quarterback. So I turned to my computer to find answers.

It was then I decided that maybe scouting wasn't perfect and to try and develop a QB success formula. So I started with every quarterback who was a starting in the NFL outside of Joe Flacco, because he was a division two starter, and I analyzed all of their college numbers. I found what appeared to be the most important data.

I then added in every quarterback taken in the top three rounds of the last decade. Now, of course, it’s up to every regular starter from the previous year and the last 12 years of top three round picks.

I started to weigh the values of each stat and then decided to make the average value for it 100. I adjusted the weighed numbers every year. More data means more information. It actually has two steps. The first one is a winnowing point for finding failure and this might be the most successful point. It has found who won’t succeed at about a 90% clip so far.

The thing is, I know who are the successful quarterbacks and who aren’t. So I know that the answer is just about finding a means to get there. While my numbers aren’t perfect, there are several I would like to use that I don’t have the ability to get. The amount of stats out there in one place on former college football players is very limited. I have still seen the formula grow and become a growing way I can evaluate prospect quarterbacks.

You might ask who am I to make a math formula and make it relevant and useful. This is fair and I had the same issue. So I went to my father in law. He has a master’s in mathematics and a PHD in engineering. He spent years working as a mathematician in the private and governmental sectors. He went over the formula ran it through a computer program and declared based on what I hoped to find what I was using were relevant stats to find success.

Editor's Note: Jeff Ellis is one of the most respected baseball draft gurus in the nation. Last year, while I was working at both Indians Baseball Insider and Cleveland Sports Insider as Senior Editors to both, Jeff shared his draft formula for the first time at CSI. It should be worth watching over the coming years to see the formula, as he hones it in. Put your trust in Jeff.

Last year, by the numbers, this was how it rated the top quarterbacks in the draft. It had Derek Carr solidly at the top. Then Teddy Bridgewater and Johnny Manziel were nearly identical, but a tier down from Carr. Last was Blake Bortles ,who earned a grade which was below the hundred point average score.

Now that the preamble is out of the way, I plugged in five players into the formula this year even though I know there are only two players people care about at the top of this year's draft. I will focus on the big two, and then give notes on the other three at the end.

Jameis Winston, QB, FSU

I don’t think there is a more polarizing player this year, but he might not even be the most polarizing quarterback. Scouts love Winston based on tape. He has it all and can lead a team in success. He makes the right reads and is a true field general. He has the off-the-field issues, which make him polarizing, but on-the-field, the consensus is that he is a franchise quarterback.

Yet, my formula disagrees very strongly. I don’t think it would surprise anyone to find interceptions are part of the formula. The reason is simple big numbers in college tend to balloon in the pros, as the amount of dropped interceptions in college is much, much smaller compared to the pros. I could go into the other numbers, but I guard my formula pretty closely. Needless to say, this is one area my formula shows for warning. It should not be too hard to understand interceptions are bad.

Now in the history of my formula, a negative grade in either section puts a player on the don’t draft list. The biggest success with a negative score is Michael Vick. The problem with Vick is his second season, he only played ten injury plagued games. He was drafted more on his first year performance, when he nearly became the first freshman to win the Heisman. Vick is the most successful quarterback in my system with a negative score. Vick’s data I nearly didn’t use because of the injury issues, which lead to performance issues.  The next most successful would be Rex Grossman. I don’t think any comment is needed.

Winston’s numbers in general aren’t awe inspiring. He came out with a score 104 in the second section, after a negative in the first section. The score of 104 makes his nearest statistical comparisons Ryan Mallet and Colt McCoy.

I am not going to go into red flags and off-the-field issues. I can’t make those numbers and plug them in. The numbers I have found say that Winston is a player you don’t draft in the first two rounds at least. I realize this might make me sound crazy and he could be the exception that breaks the rule. Yet I am going on record as saying if I ran a team, I would not draft Winston early the numbers have yet to fail me and they say he is a likely bust.

Marcus Mariota QB Oregon

In many ways I think Mariota is a more polarizing quarterback. Almost everyone is so sure of Winston, yet since the last time Mariota played, it has been mostly talk about what he can’t do. People seem to forget that after all the hype, when the two quarterbacks met, it was Mariota who was the victor in spite of having the less talented squad, at least in draft terms.

By the first measure, Mariota is solidly above average. He doesn't quite hit the peaks that guys like Andrew Luck, Russell Wilson, and Derek Carr hit but the first number is more about winnowing out players than true judgement.

The second number is where Mariota shines. Numerically, when one looks at every component that has been tied to future success, Mariota is scoring in the top 10-15. Overall he has the highest point total my system has ever produced at a 167.322. This is a full 21 points ahead of Sam Bradford, who is number two.

The numbers can be wrong, and it’s hard to predict success and failure, but so far there has been one bust in the top ten quarterbacks, and that was Geno Smith.

For me, it’s not even a choice, and I know to others who are trying to do similar things, predicting quarterback success rates it isn't either. Mariota has top ten potential, while Winston screams bust.

Other QB’s

Bryce Petty, QB, Baylor – He has a negative score in my system as well, though his secondary score is very high. I would not be shocked to see him become a solid backup. A guy who had a similar negative score than high secondary score last year was David Fales.

Brett Hundley, QB, UCLA – He scores three points higher than Winston, and is a better athlete. I know there are issues, but I would rather have Hundley than Winston based on numbers. He tried to change what he was as a quarterback, and was not as effective this year.

Garrett Grayson, QB, Colorado State – Below average in the system numbers wise, he matches up nearest to Kellen Clemons and Zach Mettenberger. He is nearly identical to Mettenberger. Then you consider the level at which he played, and I have issues with Grayson as a top three-round quarterback.

I don’t have any other projected top three round quarterbacks. So here it is the yearly look at how this year’s crop grades out using the EQF, Ellis quarterback formula.
Share on Google Plus

About Jeff Ellis

Under Construction