This post will discuss what close games can ultimately tell us about college basketball teams. Whether past clutch performances can predict future play in tight situations. Lastly, what teams in the NCAA we should trust going into the bracket.
The Clutch Performance Fallacy
Lets take a quick look back at that IU game I referenced. The Hoosiers were down 4 with 1:26 left in the game. I would venture to guess teams in that situation lose at least 90% of the time, probably more. Then Christian Watford drew and and-1 off balance push shot from 12 feet away and converted the free throw. That play in and of itself was not lucky, but IU was fortunate to pull out the game. ESPN analysts will rave about Indiana's late game execution, as they should, but thinking this will be a trend would be a mistake.
Basically the theory boils down to this, in a close game there are only a certain amount of possessions to determine the winner. If a game is tied with 2 minutes left, there is only 6 or so possessions to dictate who the "better" team is. In 6 possessions, anything can happen and it would be foolish to judge who is the better team just based on the winner.
One of college basketball's most respected bloggers, Ken Pomeroy, recently collected a bunch of data on this. He found what I expected, regardless of how good a team is, close games (decided within 3 points) are basically 50-50 on who ends up winning. He even looked at if a team won its first 4 close games and would it predict how many close games they won in the future. It turns out teams who won their first 4 close games won 54% of the rest of their close games, while teams that lost their first four won 46% of their remaining close games.
It might shock people to hear this, but most coaches would prefer to not be in close games for this very reason. To quote coach Jack Reilly from Mighty Ducks 1, "it’s not worth winning if you can’t win big". Billy Donovan's Florida Gators have provided us with a perfect example of a team that is being criticized by the media for not being "battle tested". Florida has blown out SEC teams by an average of around 15 points a game. So the media would prefer that Florida play close games against sub par SEC competition and believe that somehow prepares them for March? 1) average margin of victory is one of the best indicators for how good a team is. 2) Lets say Forida ends up playing more close games, but inevitably loses some of them. Then Florida gets reamed for losing to a crappy SEC team. It is a no win situation for Donovan. So while it is nice to see your favorite team execute and win in the close games, please don't fall into the trap of thinking that will be the norm.
Consistency is King
So now that we are well into March, which teams should you trust heading into the tournament. Below is a table, courtesy of @RealGM, that tells us how many bad performances a team has had as of 2/26/2013. A "bad performance" is any game where the team’s game-specific Pythagorean Rating would be less than 0.9000. (Pythagorean Rating adjusts for opponent and venue as on kenpom.com.) In my opinion this is probably one of the best ways to see who is the cream of the crop in the NCAA and definitely a better indication than wins and losses.
Some things that jump out right away, again would be Florida. The Gators had 5 bad showings as off 2/26 and have played 5 total close games if you count yesterday's game against Kentucky. They actually have lost 4 of the 5 close games and will be penalized accordingly in next weeks rankings. We will probably see Florida as a 3 seed come March, but I would be buying their stock right now. This is one of the top 5 teams in the nation.
Teams near the bottom of the table but top of the polls: Michigan State, Kansas and Miami. Most of the Spartans bad performances came in the non conference. As of Big Ten play they have been much better, so I wouldn't worry too much about them. Kansas and Miami on the other hand are teams to be wary of. Kansas has had some clunkers, TCU comes to mind and last night against Baylor was not pretty. Bill Self is top 3 in game coach right now so I would hate to doubt the Jayhawks, but their ranking could be somewhat inflated come tournament time.
Then there is Miami. This is a team I want to love so bad. A cast of seniors from Kenny Kadji (who is older than Kevin Durant), to Reggie Johnson and Durand Scott who are led by coach Larranaga and sophmore point guard Shane Larkin. They are the media darling. Coming out of nowhere and looking like they could run the table in the Durham dominated ACC. Most of their bad performances came in February against ACC schools that aren't exactly power houses. Now they have lost two more in March and were tied going into half time last night at home against Clemson. I think this team is good and well coached, but I'm on the fence if they are a team I can trust.
If there is one takeaway from this post it is that close games can be decided by improbability That is exactly why we love March Madness. The double digit seed upsets, teenagers being put in pressure situations that we over examine and everyone thinking they can predict what is going to happen. There is no favorite and that is exactly how I like it. Good luck in your tournament pools and I will have more about tournament match-ups as the end of March is on the horizon.
Bonus Video
Random video that popped up on my timeline the other day. Team is up 1 with 3.7 seconds left. All they have to do is hold onto the ball or get fouled. In bounding from the back court and a player loses focus for a split second. To quote the camera guy, "He stupid". The kid scored a buzzer beater for the opposing team. This isn't exactly what I had in mind when saying close games are impossible to predict. It's truly unbelievable.
No comments:
Post a Comment