Anatomy of an Upset
September 13, 2012 by Kevin Beane • Print Story •
This past week in college football, we got a real rare vintage of one, where the significant underdog came back from a big deficit (as I noted in the second entry in the first link above, that is a vanishingly rare feat). Louisiana-Monroe, who has never had a winning season since joining what we then called (and still ought to call) Division 1-A, took out No. 8-ranked Arkansas 34-31 in overtime after trailing 28-7 in the second half.
I watched more of this game than you might expect from a Midwestern boy who is not a fan of either school, thanks to "ESPN Goal Line," a new channel from ESPN that jumps from game to game on the ESPN family of networks.
When Arkansas went up 28-7 in the third quarter, this upset-lover was bummed, because the game felt like it should have been a lot closer; ULM had two drives stall (once turning it over on downs at the Arkansas 2, once getting picked off) inside the Arkansas 15.
But the Warhawks didn't fold, and stayed aggressive. They gambled. That was half of the key to their upset. If I've heard it once, I've heard it a thousand times: some idiot pundit will say "if anyone is beating (dominant team in any sport this year), they're gonna have to beat them at their own game."
No! No! Playing a straight-up gameplan where you try to match Goliath blow-for-blow is exactly the formula for giving your team no chance to pull the upset. When was the last time you heard some jubilant coach interviewed on the sideline after knocking off a 25-point favorite say, "Well, we just played it safe, tried to be conventional as possible, made it a war of attrition, and it worked out for us." If you answered, "Never, and never will," you are correct.
If you coach a team like ULM and a play a team like Arkansas, it is imperative to realize that when it comes to talent, you are grossly outmatched. They are faster, stronger, and deeper than you by orders of magnitude (although those orders are thankfully shrinking, thanks to increasing parity and homogeneity in youth and high school training techniques). If you just throw your team out there to the wolves and only make cursory adjustments, the only way you will win is if Goliath makes a ton of mistakes.
And yet, the underdogs rarely get creative when trying to achieve the nigh-impossible. Maybe it's because their coaches are unimaginative, or in denial about the reality of the situation. Maybe it's a CYA approach, because if you gamble a lot and it doesn't pay off, you'll lose even worse than if you played a straight-up game plan.
Nevertheless, as much as Goliath might have a monopoly on talent, that never, ever means they have a monopoly on brains. This is pretty much the only area where David has any control over his fate. You can do it in complex ways by gameplanning to accentuate whatever strengths you might have to the hilt, and doing the most you can to semi-neutralize Goliath's strengths. It's hard work, and can't really be summarized in the column because it will change from game to game, but it's important.
The other thing you can do, should do, which is much simpler, is take chances. Gamble, Try to fool them. Trick plays. Misdirection. Fakes. Two-point conversions. Go for it on fourth down.
The latter is what ULM did. They went for it on fourth down a staggering seven times in the game, and converted on six. This included a pair of fourth down conversions of 10 or more in the third quarter, with the result still in doubt; that is to say, punting would have been conventionally "correct."
Although I'm primarily focusing on coaching for this piece, I'd be remiss not to mention ULM quarterback Kolton Browning, who is the Armanti Edwards of our time. He passed for 412 yards, rushed for 69 more, and was responsible for 4 of ULM's 5 touchdowns.
Back to coaching. Contrast ULM head coach Todd Berry's decision-making with Washington's Steve Sarkisian, who, down 20-3 to LSU, late in the second quarter, had a a 4th-and-12 at the LSU 46. Sarkisian sent in the punter. Because the way you come back from 17 down against LSU is to play a field position game, and you may not be able to stop them from going 55 yards if you go for it and miss, but you can damn sure stop them from going 90 yards with a good punt, and then come back and win. Totally.
The best summation I have read on Louisiana-Monroe's victory is this piece by Bill Connelly. Among his choicest comments:
In 2010, Berry did his best to do things a bit off-kilter — running when opponents expected the pass, passing when opponents expected the run, keeping things fast-paced (possibly not the best idea for an underdog), employing the underdog-friendly 3-3-5 defense, etc. — and the results were decent; despite low overall quality ... I am rooting for Berry to succeed for one simple reason: I want creativity and aggression to be rewarded. Nothing is more depressing to watch than a David trying to win games like Goliath would. Taking risks occasionally leads to calamity, but if they pay off enough, it might encourage other coaches to take similar risks.
Emphasis mine. Yes, he's preaching to the choir, but damn it, I've been saying this stuff not just in this column but in past ones, for years, and it's nice not to be the only voice in the college football desert. He even uses the same (granted, pretty easy and obvious) metaphor. If I could clip and mail that snippet to every school that is a 20+ point underdog in a game forevermore, I would.
I wrote earlier that you may not be able to outplay teams with superior talent, but you at least have a chance to out-coach them. That task becomes easier when the favorite goes full dunce, and boy, did Arkansas ever oblige.
It hasn't been an easy offseason for the Arkansas football program. They had to get rid of head coach Bobby Petrino after he got into that motorcycle accident with his student mistress. They awarded his successor, former Michigan State coach John L. Smith, with perhaps the least-confident contract of all time: a 10-month offer. He surely won't make it past that, nor perhaps to that, now.
Arkansas marched out to that perhaps-undeserved hefty 28-7 lead in the third quarter. They were down to their backup quarterback, but that's okay, you've got two potential NFL running backs in Knile Davis and Dennis Johnson, to grind out the clock and the game against a Sun Belt defense.
But bafflingly, that's not what Arkansas did. Instead, Smith and offensive coordinator Paul Petrino passed and passed and passed, with their backup quarterback. Between the time when they went up 28-7 and when ULM tied the game, Arkansas ran 6 running plays and 11 passing plays. They completed three of those passes.
That play selection is straight-up lunacy for a big favorite trying to protect a lead with two good rushers and a backup quarterback. It's as if they needed throw ULM off-balance instead of the other way around.
Connelly wrote that it depresses him when David tries to win games like Goliath. Not only did ULM not do that, but Arkansas tried to win like David. Up is down. If Smith isn't doing this again, he should be.