Tuesday, December 29, 2009

What College Basketball does better than the NFL

Does anyone really watch college basketball between the ACC Challenge and the start of conference play? OK, sure I remember a few things from this month. I remember John Wall’s slashing dunk against Indiana. I remember Stanford’s ferocious comeback against Oklahoma St. in the Pac-10 / Big 12 shootout that came up just short.* And I remember Seton Hall’s Jeremy Hazell hitting a pair of three pointers to tie the game against West Virginia and send the game to OT before Seton Hall ultimately lost. But I, like most college basketball fans consider this to be the “ticket exchange” season. This is the time of the year when you end up spending more time trying to give your tickets away then actually enjoying the games.

*In case you didn’t notice, Stanford is the unluckiest BCS program as of today. Sometimes Pomeroy’s “luck” is a statistical fabrication where teams win big and lose by five or six, but Stanford has several legitimate heartbreaking losses. The Cardinal lost to Kentucky in OT, Oklahoma St. by 1, and Oral Roberts by 2.

And although I’ve paged through the numbers and amused myself with a few facts, there hasn’t been enough to hold my attention. Here’s one topic: Louisville, Syracuse, and West Virginia all entered the season with serious ball distribution questions and the fate of each team has been linked to the addition/return of a turnover plagued point guard.

For Syracuse, it was Scoop Jardine. He missed 2009 due to injury and had huge shoes to fill in the place of Johnny Flynn. For West Virginia, it was Darryl “Truck” Bryant who was suspended this summer after a hit-and-run and was questionable to play and start early this season. And for Louisville, it was McDonald’s All-American freshman Peyton Siva whose ball handling was needed with point forward Terrence Williams graduating in the off-season.

So far, the results have matched the team standings.
-Syracuse’s Scoop Jardine sports an assist rate of 38.2% which is good for 15th in the nation. His ability to set up his teammates has Syracuse in Pomeroy’s Top 5.
-West Virginia’s Darry “Truck” Bryant sports a 26.5% assist rate which given the Mountaineer’s depth has been good enough for a Top 5 Pomeroy Rank.
-But Louisville’s Peyton Siva has turned the ball over 34.3% of the time which has been good enough for a spot on the bench. This has meant no real challenge for playing time for the inconsistent Edgar Sosa, and an inconsistent start for Louisville.

By the way, I’ve always been puzzled by Pomeroy’s characterization of Major Contributors and Nearly Invisible based on possessions used. But what does it mean for a player like Siva? Siva only takes 17.4% of the team’s shots but is classified as a major contributor because he turns it over so much he takes 25.4% of the team’s possessions. Amusing.

But none of that was the point of this post. The point of this post was that watching other sports reminds me why college basketball has a near perfect system for selecting the post-season.

The Play to Win Principle

This week, the Colts rested their starters because the games were meaningless. This robbed the season ticket holders of an entertaining game. And it provided an unfair advantage in playoff positioning. Whereas Houston had to play Indianapolis twice when Indy was at full strength, the Jets got to play Indy for 2.5 quarters, and a JV squad for 1.5 quarters.

But one beautiful feature of college basketball is that this rarely happens. Because a committee evaluates and seeds teams, teams need to keep playing hard at the end of the year. No one can wrap up a 1 seed on Feb. 15th, and play the bench the rest of the year. The finishing stretch is a vital component of seeding, and the good teams try to play their best at the end of the year.

Imagine if the NFL was like college basketball this year. Would a 13-2 Saints team really be a “lock” for a 1 seed after a bad loss to Tampa Bay? Or would streaking San Diego be moved out of their “region” and given a 1 seed in the NFC? And what about a team like Carolina which sits at 7-8 after crushing wins over Minnesota and the New York Giants. How big would their game against New Orleans be this weekend if New Orleans was fighting to hold a 1 seed and if Carolina was a hot bubble team? The property that team’s have to play hard at the end of the regular season is something most sports leagues should want, but is sadly missing in most pro leagues in the US.

I note that the “play-to-win late in the year” principle isn’t unique to college basketball. The World Cup implicitly includes it. That’s because soccer teams are seeded in the draw for World Cup slots. Thus even if England clinches a World Cup Berth in their Euro Qualifying Pod, they still have to try to win the remaining games to maintain their FIFA ranking and earn a protected seed in future tournaments.

Keeping this in mind, there are several solutions the NFL could implement to get more incentive compatibility. First, like the FIFA rankings, the Colts wins this year could have some minimal impact on future seeding. I.e., this year’s record could be a tie-breaker for next year’s playoff spot.

Or for something wacky, the NFL could do the reverse of the NBA draft lottery. What if playoff seeding was determined through a draw where the more wins a team had, the better odds they have of getting a favorable slot? Nothing is set until the final draw, so every game counts. Maybe this wouldn’t be ideal, but it would be incentive compatible. And it is an issue the NFL should think about, particularly if the NFL goes to 18 games in the future.