TheMostViolentTeam presents the third installment of his analysis of Ben Roethlisberger's new contract, this time creating a metric that examines what kind of value, in terms of wins, that a number of teams have gotten and can expect in the future, from their franchise QBs. The QB of course is only one man on the field, but they certainly have a big hand in the outcome of games. As was the case in the first two submissions, TMVT has done all the research and writing, while I have merely helped present the data in a shiny, digestable format. Enjoy.

-Blitz-

In parts I and II we took an in-depth look at the contracts of the top QBs in the NFL, and how they compared to Ben's new contract. My theory was that while all the contracts may vary in terms of absolute total/guaranteed dollars and years, the per year numbers for the contract, as a function of cap allotment, would be very similar, and thus give us an appropriate barometer to measure Ben's contract by. What we found was that most of the top QB's took anywhere from 8-14% in terms of raw dollars of the cap allotment (NOT the cap hit), in the year the contract was signed, with most of them hovering around the 10% mark. Thus we judged that Roethlisberger's contract was on par with the other top signal-callers around the league. Lastly, and most importantly we'll take a look at the projected performance of all these top quarterbacks, and how it relates to how much money they are being paid.

To do this I created a statistic called Dollars Per Win. What this inteds to tell us is how much a team is paying their star QB for each win they can be expected to generate in the future based on past performance. It can be broken out to show how much total (i.e. total contract value divided by projected future wins), and guaranteed (guaranteed portion of contract divided by projected wins) money a team is spending on its QB. Also, since it is unlikely a quarterback earns every single dollar of the contract, and equally unlikely that he only earns the bare guaranteed minimum, we'll take the mean of the two numbers (total and guaranteed contract amounts) and divide that by projected future wins to come up with an average-type number based on the assumption the QB is more likely to earn somewhere in between his total and guaranteed contract figure. This statistic is fairly easy to come up with, all we need is the quarterbacks career win percentage (I used regular season only), number of years their contract is for, and number of games in a season.

For an example of how the formula works we will use Drew Brees:

To calculate his winning percentage I look at his entire career (which includes those games he's already played on his current contract). His W-L record is 39-35 for a winning percentage of 52.7% (.527). Now, because this number includes 2006 and 2007, which are played on his current contract, so you might say, "well we shouldn't include that in his winning percentage", except, his winning percentage without those two years is nearly identical, at %52.4 (.5238).  I checked this with some other QBs and found the same. Thus, I felt comfortable using their entire career winning percentage, THEN going back from the start of their main contract (if they've already played on it) and projecting how many wins they SHOULD (or should have) generate(d) from that point on.

Brees signed a 6 year deal, so we project for him (6 years * 16 games = 96 games * .527 win pct) = 51 wins. Over the course of his deal Brees should expect to win 51 games. Brees contract pays $60 million total, $20 guaranteed, and thus the average contract number we'll use is $40 million. So we calculate:

$60 million/ 51 wins = $1.18 million (total) dollars per win

$20 million/ 51 wins =  $.39 million (guaranteed) dollars per win

$40 million/ 51 wins =  $.78 million (mean) dollars per win

So for each win basically New Orleans is paying Brees $1.18 million dollars in total money, or $.39 million in guaranteed money, or $.78 million when we look at a "average" number that is more likely to be earned than the base guaranteed amount and the top total amount. Let's see how the other top QBs size up.

Projected Wins and Dollars Per Win Stats for Top-Tier QBs:

• Firstly, how does Ben Roethlisberger stack up from a value perspective?  Well, the numbers look pretty darn good. He has the 3rd best winning percentage; the 3rd best guaranteed dollars per win rating, and 4th best in total and mean dollars.

• As expected, Brady has the highest win percentage, and considering he didn't sign a gargantuan deal like Palmer or Manning, his Dollars Per Win numbers are some of the best. The surprise here is Hasselbeck, who despite only having the 5th highest winning percentage of QB's we looked at, looks to be an incredible bargain. He is #1 (or tied for it) in every way we look at Dollars Per Win. His total number ties Brady, and his guaranteed and mean number check in significantly below Brady's.

• While you would be hard-pressed to find anybody say Hasselbeck is a top 5 quarterback in the league, it's pretty clear that for the number of wins he produces, he is an incredible value.

• Moving down the list (we'll save a more thorough analysis of Ben for later) we see Romo comes in at third in most measures, although it should be noted that not only is his career win percentage based on a MUCH smaller sample size than everybody else on the list, but also that his guaranteed dollars per win is 5th. As we stated in Part II, that's largely due to Jerry Jones fat wallet and Romo's age (necessitating more guaranteed money with a shorter contract essentially).

• Brees and Bulger have nearly identical win percentage numbers but due to Brees having a smaller contract his value numbers look much better. Furthermore, New Orleans really got a steal because they were aided by concerns over the shoulder injury he suffered in his last season in San Diego, which helped keep his price down.

• Peyton Manning is a bit of a surprise coming down so low on this list. I would never say Peyton Manning is overpaid, in fact, I would treat him differently than every other QB since he not only makes the throws but is also an impromptu offensive coordinator on the field. While there is no way to factor this in, I still think Peyton was had for a decent value. I doubt any Colts fans would complain, considering they have been one of the winningest teams in the league with him at the helm.

• As stated in Part II, Carson Palmer is the big disappointment. With how dysfunctional that team has been it's hard to say if it's entirely his fault, but as the QB (fairly or unfairly) he should shoulder some of that responsibility. It's possible Cincy's front office was so anxious to get a deal done because they hadn't had a franchise type QB in ages, that they overvalued their own player a bit. If he could raise his winning percentage up to the mid 60s then he wouldn't be so far off the middle of this list, but given the division and conference he plays in, that would be a minor miracle.

Big Ben Without 2006:

I also calculated Ben's winning percentage and Dollars Per Win statistics if we removed the atrocious (and most likely statistically aberrational) 2006 season. It seems likely the Steelers front office didn't hold that season against him either, and when removed, his winning percentage jumps to 79.7% (.7969). Here is what his numbers look like after we adjust:

As you can see, he is now projected to win 102 games over the next 8 seasons. Furthermore, his win percentage is better than Brady's, and his guaranteed and dollars per win measures now move up to 2nd, behind only Hasselbeck. His total dollars is still firmly fourth, and as we see if Ben earns every penny of his contract, and delivers 102 wins, he'll earn 1 million dollars for every W.

Final Conclusions:

When we just compared Ben with the other QB's on a salary only basis, he seemed to have a decent contract, right in the middle of the pack. I specifically came up with Dollars Per Win because I wanted to see a way where we could relate not only contract size, but also contract size as a function of wins a quarterback produces. The results are just about what I expected, with the major surprises being Matt Hasselbeck and Peyton Manning.  Hasselbeck showing incredible value, and Manning's numbers SEEMING to indicate he is overpaid (although I clearly do not believe this to be the case). Based on the number of wins we might expect Big Ben to help generate for us in the future, we got an excellent value. Like any quantitative tool this doesn't tell the whole story, but I feel very confident the numbers back up what we see as football fans with our eyes (i.e. I think most people agree). Any feedback/input is greatly appreciated; also let me know if you want me to update these numbers as the season (and thus a QB winning percentage changes) goes on. I have some other ideas for measuring QBs as a function of their contract (such as dollars per touchdown thrown/ran for), so let me know if this kind of stuff interests the BTSC readers here.