Joe Starkey has no use for football stats, logic

I am in the progress of writing a QB-related post that I will have for you in a few days. But first, I can’t resist going all FJM-style (but at about 1/3 the
entertainment value) on this Joe Starkey column from Thursday. I actually think he’s baiting the “stats crowd”, but that’s a different topic.

Hit the jump for a little point-counterpoint on Starkey’s head-scratching column.

(Note: Starkey is bolded. My thoughts are below…)

 

Starkey: Ben’s Unique Talent

Oh boy, I can’t wait to find out what kind of talent Ben has that not even Peyton Manning, or Tom Brady have!

What’s he do on his worst days? That’s what I want to know about a quarterback

Ben Roethlisberger’s worst day is much better than, say, Alex Smith’s. But Roethlisberger is a much better quarterback than Smith. A good QB’s good days and bad days are better than those of a bad QB.

Ben Roethlisberger often wins on his worst days…

Because the Steelers’ defense consistently ranks in the top ten according to almost every stat imaginable, including Roger Goodell fines.

He also has the best definition of quarterback toughness I’ve heard…

Being able to come up with definitions of previously undefined terms helped him immensely on that game-winning drive against the Cardinals in SB XLIII.

…one sure to send…

I think I know what’s coming next!

…stats geeks running to their hard drives to see if it fits into one of their unfathomable formulas.

Leaving aside the “hard drives” crack, if a formula is truly unfathomable, not even the stats geeks would understand. So, who finds them unfathomable? I’m guessing people who read columns by Starkey.

We broach that topic because it’s that time of year – time to rank the quarterbacks.

Who says it’s time to rank the quarterbacks? A Google search for “QB rankings” yields on the first page six fantasy football articles, a John Clayton article, a USA Today article, and this article that I might just cite in my next post. So, fantasy football players say it’s time to rank the quarterbacks. Those guys use lots of numbers, Mr. Starkey. Are you sure you want to associate yourself with them?

I don’t know if Roethlisberger deserves to rank first, but I know this: he belongs in the conversation.

I agree with him on this. In one of the simplest passing efficiency stats out there, yards per pass attempt, Ben is tied for first among active QBs, higher than both Peyton Manning and Tom Brady.

Numbers freaks are flummoxed on how to measure a quarterback because the statistics lie like winos.

I am flummoxed by this simile. A Google for the phrase “lie like winos” yields only the article (and this post will likely join it when it’s published), which means it’s a simile of Starkey’s creation. A wino is a homeless person who consumes alcohol. What about that indicates that such a person is also prone to lying? From observing other people, I’m pretty sure drinking alcohol makes you more honest, rather than less.

Super Bowl titles? If that’s your primary measurement, you must believe Mark Rypien is better than Dan Marino.

Yes, anyone using only Super Bowl titles to evaluate QBs is lacking common sense. What else?

Passer Rating? Tony Romo tops Tom Brady and Joe Montana.

You know who else tops Joe Montana? Ben Roethlisberger! So what were you saying about Roethlisberger being an elite QB? Granted, using only the traditional passer rating probably isn’t the best way to evaluate a quarterback, and…dammit, why couldn’t you have waited until I finished my other post?

Accuracy? Chad Pennington is the most accurate passer of all time.

Passer rating uses accuracy (completion percentage), along with several other stats, so using just completion percentage is (pardon the pun) less accurate.

Won-loss record? Getting warmer.

Really? We are?

Roethlisberger is 69-29 in his career.

Hall-of-Famer Warren Moon was 102-101. He stunk! Jim McMahon went 67-30 (.691). Why isn’t he in the Hall of Fame, if his winning percentage is so good? Is Jim McMahon really better than Manning, Terry Bradshaw, Young, Unitas, Elway, etc.? You can make the same type of arguments against winning percentage that Starkey made against all the other stats he listed. So why doesn’t he? Oh right, he needs some kind of transition into the crux of his article, the toughness thing.

Which brings us to his definition of toughness…”Toughness is playing the worst game of your life but not backing down…Staying in that game, keeping your head up, trying to drive your team when everything is going wrong – that’s the kind of toughness I want in my quarterback.”

I think of last year’s game at Baltimore, when Roethlisberger – having a miserable night with a banged-up foot and a broken nose – was presented with an opportunity to win…

…by Troy Polamalu’s great defensive play…

…and snatched it…Is there a stat for that?…

Yes, actually, there is! It’s called Number of Scores Executed, or NOSE. Neat acronym, huh?

Seriously, though, it’s called a fourth quarter comeback, and Roethlisberger actually does rank highly in this stat – tied for fifth among active players and first in just the first seven seasons of a QBs career. So, you can use a stat to measure a QB’s effectiveness while down one score late in the game, and it supports the toughness quote from earlier? Man, I guess stats are good for something, right Mr. Starkey?

I’m beginning to loathe the statistical revolution that has engulfed the sports world. When did everybody become Billy Beane?

Billy Beane? What is this, 2005? Sure, there’s that Moneyball movie coming out, but the most successful GMs of today (Theo Epstein, Andrew Friedman, etc.) use both statistics and human-based scouting, which renders the rest of this article pointless.

Don’t get me wrong. I love the numbers game…

Could have fooled me, with all the glowing reviews of different statistics you wrote earlier.

The problem is that statistical devotees have forgotten the human element. They disregard the notion that an athlete…can separate himself from his peers by consistently summoning the right stuff at just the right time.

“From his peers” mean what, exactly? Is Roethlisberger a better QB than Joe Flacco because he has more fourth quarter comebacks, or because he’s a more efficient QB? Better QBs are more likely to lead a scoring drive at any time. Therefore, the best QBs will also have the most game-winning drives. It won’t be exact, due to things like luck and opportunities for game-winning drives, but that’s why you use several criteria.

Winning time. Sports is still about winning, right? Right?

Yes, it is about winning, but using “just winning” to evaluate a QB isn’t right, because it’s not just about the quarterback! There’s an entire portion of the team that is on the field when the other team is on offense, and the quarterback doesn’t stay on the field for those plays. I mentioned Troy Polamalu above. If he doesn’t make that strip sack of Flacco, Roethlisberger might not have a chance at that comeback. Football is a team game, not just a QB game.

Quantcast