Friday, August 17, 2007

Does it Matter Who Starts in the Opener?

Evan Sharpley, Demetrius Jones, Jimmy Clausen. These are the three candidates to be Notre Dame's starting quarterback for the first game of the 2007 season against Georgia Tech. Who will Coach Charlie Weis choose? Who gives the Irish the best chance to win the game?

Does it matter?

Whichever of these three players is named the starter, it will be their first game as Irish starter. Which got me thinking: How do Notre Dame quarterbacks generally do in their first start? How about when that first start comes in the season opener? So I did a little digging into the archives.

The first surprise was how hard it was to find the data. Win-loss records and scores for every game back to the beginning of time are readily available. But individual game statistics for older games are hard to find on-line, and I was unable to find any reliable database of "Notre Dame starting quarterbacks." Once I went back prior to the early 1990s I had to rely on snippets from news stories on-line to try to piece together who started for the Irish, and when. I found a few articles that claimed to list first-time starters for certain periods of time that failed to mention players like Gary Godsey (2000) or Mike Courey (1980). Thus, my confidence level in the data included below is only abut 90%. I have probably left somebody out, or listed the wrong first start for one or more players. If you spot an error, please let me know so I can fix it!

The data below concerns first starts for the listed quarterbacks. In many cases the listed quarterbacks had significant playing time in back-up roles prior to getting their first start. Being the "starter" can be a little arbitrary. It's possible to come off the bench and get a majority of the meaningful snaps. It's also possible to "start" and play only one series until the regular starter comes in (See "Paul Failla, 1991"). But the question at hand concerns the "first start" for either Sharpley, Jones, or Clausen, so "first starts" is what we'll examine.

How far back in time should we look? It would be fun to go all the way back to the days of Rockne. But I have a day job, and the data would be very hard to extract. Plus, it's hard to argue that the data is very relevant once you go back beyond a certain point. So I have chosen to look at Notre Dame starting quarterbacks in my lifetime. John Huarte is first. He made his first start in 1963, and was still the starting QB when I was born in 1964. Listed below in tabular form (via an embedded spreadsheet, I hope) is the year of the first start, the QB's name, an indication of whether it was a season opening game or not, the opponent, the game result, and stats and notes, for every starting Irish QB from 1963 to the present.

Observations: There have been 34 different starting quarterbacks for the Irish in my lifetime. If you had asked me before today to list all those that I knew about or could remember, I honestly would have been lucky to get half the names. In my memory are Theisman, Tom Clements (my childhood idol), Montana, Rusty Lisch, Blair Kiel, Beurlein, Andrysiak, Rice, Mirer, McDougal, Powlus, Jarius Jackson, Eric Chappelle (ugly USC loss I saw in the Coliseum), Arnaz Battle, LoVecchio, Holiday, Dillingham and Quinn. I only recall Lisch, Andrysiak, Chappelle, and Dillingham after looking at the list. Amazing how you can block things out.

34 seems like a big number. 34 QBs in 43 seasons works out to just over one new quarterback starting every 1.25 seasons, or a new QB every 15 games or so (12-game modern schedule) or every 12 games or so (10-game older schedule). Since we know that an established starting quarterback will stay in the line-up for two, three, or even four years, that means a lot of short stints for the tweeners, the guys that fill in the gaps between QB eras. The 1971 season saw three different players make their first start at QB. They all won. Same thing in 2000 - three first time starters, three wins.

Overall, of the 34 different quarterbacks making their first starts in the "OC Domer Era", there were 27 wins against 7 losses. That's pretty impressive.

Even more impressive is this: Notre Dame quarterbacks have made their first start in the first game of a new season twelve times. Notre Dame is 12-0 in those games. Whether it has been Bill Zloch or Joe Theisman, Mike Courey or Rick Mirer, Notre Dame has not lost an opening game with a first-timer starter in my lifetime. Twelve wins and zero losses. That's surprising, but also encouraging. Whoever Weis taps, I like our chances.

By the way - we open at home.

First time starters are just 15-7 when the start doesn't come in a season opener. Not bad, but not great either. But it makes sense. These guys are starting after (1) they failed to win the job at the start of the season, and (2) the #1 QB either got hurt or struggled

Injuries can force a change at any time, but quarterback changes are much more likely in the first half of a season than in the second half. New quarterbacks got their first starts in games 2 through 5 of the season fifteen times. Games 6 through 11 had first-time starters only eight times. Weeks 4 and 5 are the most popular times for a mid-season QB change. It seems that by the time the mid-point of the season is reached, you're dancing with who brung ya.

So what does all this mean? Well, past success is no guarantee of future performance and all that. But you have to like our chance in the opener with new quarterback. If history is any guide, the opening day starter has a better than even chance of being the starter for the whole year. If he stays healthy, his chances of keeping the job for a full season are significantly better than 50-50. There is a chance that a change at QB could be made, even barring injury, although it is much more likely to happen in the first five weeks of the season if it's going to happen at all.

All of which merely reinforces how important it is for these guys to win the competition in camp before the opening kickoff. History shows that it is much easier to keep the starting QB job than to dislodge an established starter, absent an injury.


Anonymous said...

GREAT Information OC, thanks for dropping it at Domer Domain. I read where OC Domer is #2 behind BGS. (You try harder like we do at DD) Again a great job, that I know took some time trying to find the information.

Anonymous said...

Great post OC Domer. I have great respect for bloggers like yourself who take the time to research and write a post like this.

IrishGlory said...

Great job OC Domer. Thanks for taking the time to do the research and provide us with a top notch article!!!

Anonymous said...

Great acticle! Test on Tom Krug (after having his bell rung) showed a pre-existing abnormality of his neck/spine.

Anonymous said...

Actually, Theismann's first game as a starter was the 1968 USC game (21-21 tie at the Coliseum -- Irish defense held O.J. to 55 yards in 21 carries).

OC Domer said...


Thanks for the correction. I think you're right. Thus, first-time starters in season openers are actually 11-0. First-time starters overall are 26-7-1. And first time starters in non-openers are 14-7-1.

I will try to edit the post, but Blogger gave me a lot of trouble when I originally published it, and its acting funny when I try to edit (i.e., when I click on "edit" for the post I just get a blank page). I don't want to risk losing the whole thing. I think it's the embedded spreadsheet.