Early odds and 538 projections

[soonerbms]

On paper and on the eye test, we look like a better team than Villanova. Not to mention the fact that we decimated this very team four months ago in Hawai'i. Obviously, no one expects to win by 20+ again, but one would think we'd be the favorites. However, the oddsmakers have installed Villanova as an early 2 point favorite and 538 gives them a 54% chance of winning this game. In fact, 538 currently has Villanova as the favorite to win the whole damn thing ahead of Carolina.

http://projects.fivethirtyeight.com/2016-march-madness-predictions/

What do the oddsmakers and metrics analysts know that we don't? Maybe they think Buddy's scoring or outside shooting is unsustainable or that Villanova's D will stifle us the way it did Kansas.

 

[OUAllDay28]

We swooned in the late season and they played really well (Big East tournament notwithstanding). They're not the same team they were in early season. I think this game is as close to 50/50 as possible.

 

[stormspencer]

Can't say I'm surprised. Nova has improved steadily over the course of the season....all of the metrics and analytics support it. That doesn't necessarily translate into outcomes on the court, but they showed last night that they can beat the best.

 

[MNsooner]

Villanova have been playing great over the last few months. They're a different team than the one we played in Hawaii. I hope we're ready for a tough game.

 

[TINY]

Can't say I'm surprised. Nova has improved steadily over the course of the season....all of the metrics and analytics support it. That doesn't necessarily translate into outcomes on the court, but they showed last night that they can beat the best.

Ous metrics would be through the roof in the big least too.

 

[chuckb]

On paper and on the eye test, we look like a better team than Villanova. Not to mention the fact that we decimated this very team four months ago in Hawai'i. Obviously, no one expects to win by 20+ again, but one would think we'd be the favorites. However, the oddsmakers have installed Villanova as an early 2 point favorite and 538 gives them a 54% chance of winning this game. In fact, 538 currently has Villanova as the favorite to win the whole damn thing ahead of Carolina.

http://projects.fivethirtyeight.com/2016-march-madness-predictions/

What do the oddsmakers and metrics analysts know that we don't? Maybe they think Buddy's scoring or outside shooting is unsustainable or that Villanova's D will stifle us the way it did Kansas.

Carolina might lose tonight. They have 1 more game to play in order to win the tournament so it's not surprise that Villanova's chances are better than theirs.

 

[soonerbms]

Carolina might lose tonight. They have 1 more game to play in order to win the tournament so it's not surprise that Villanova's chances are better than theirs.

Carolina's going to crush ND tonight.

 

[Big Old Booger]

On paper and on the eye test, we look like a better team than Villanova. Not to mention the fact that we decimated this very team four months ago in Hawai'i. Obviously, no one expects to win by 20+ again, but one would think we'd be the favorites. However, the oddsmakers have installed Villanova as an early 2 point favorite and 538 gives them a 54% chance of winning this game. In fact, 538 currently has Villanova as the favorite to win the whole damn thing ahead of Carolina.

http://projects.fivethirtyeight.com/2016-march-madness-predictions/

What do the oddsmakers and metrics analysts know that we don't? Maybe they think Buddy's scoring or outside shooting is unsustainable or that Villanova's D will stifle us the way it did Kansas.

I have mad respect for Nate Silver (538), that dude is very smart. However, I have no clue how a statistician can predict percent probability of an athletic event so I don't put much weight into his prediction.

it's one thing to predict elections based on statistics of red/blue voters in each area and what not but too much grey area in sports to accurately predict anything.

 

[califca]

You can throw that in the toilet. Means nothing . Syracuse defeats Virginia today. What were the percentages of that happening. Slim and None. Players play.

 

[SoopaSooner]

On paper and on the eye test, we look like a better team than Villanova. Not to mention the fact that we decimated this very team four months ago in Hawai'i. Obviously, no one expects to win by 20+ again, but one would think we'd be the favorites. However, the oddsmakers have installed Villanova as an early 2 point favorite and 538 gives them a 54% chance of winning this game. In fact, 538 currently has Villanova as the favorite to win the whole damn thing ahead of Carolina.

http://projects.fivethirtyeight.com/2016-march-madness-predictions/

What do the oddsmakers and metrics analysts know that we don't? Maybe they think Buddy's scoring or outside shooting is unsustainable or that Villanova's D will stifle us the way it did Kansas.

So if OU wins they'll be favorites against say UNC right? Nope. For whatever reason the analytics side doesn't think OU is that impressive. Those analytics have looked like crap all around the past few days.

I expect a battle in all threes games in Houston. Hopefully OU comes out on top.

 

[SoopaSooner]

I have mad respect for Nate Silver (538), that dude is very smart. However, I have no clue how a statistician can predict percent probability of an athletic event so I don't put much weight into his prediction.

it's one thing to predict elections based on statistics of red/blue voters in each area and what not but too much grey area in sports to accurately predict anything.

Yeah like the ball going into the hoop lol

 

[opsooner]

I have mad respect for Nate Silver (538), that dude is very smart. However, I have no clue how a statistician can predict percent probability of an athletic event so I don't put much weight into his prediction.

it's one thing to predict elections based on statistics of red/blue voters in each area and what not but too much grey area in sports to accurately predict anything.

He spoke at a conference I attended once and his predictions ended up to be near perfect. However, I agree. Gathering trends from prior elections and polls makes a lot more sense to me than trying to use game statistics. The increase in the number of variables is exponential when it comes to predicting sports. Even then, it is only one game and anything can happen.

 

[stormspencer]

So if OU wins they'll be favorites against say UNC right? Nope. For whatever reason the analytics side doesn't think OU is that impressive. Those analytics have looked like crap all around the past few days.

I expect a battle in all threes games in Houston. Hopefully OU comes out on top.

Actually, almost any respected advanced metric rating or analytic measure has OU no worse than a top 5 or 6 team. That is a fairly reasonable assessment and accurate as well. But one other thing to remember is that these measures take a sample size of over 30 games into account and this can create some lagging indicators (i.e. teams may be playing better or worse than they were two months ago and the metric/measure has not fully caught up with their true standing as of now).

No advanced metric or analytic system is perfect. But 3 of the 4 teams in the Final Four are well within the bell curve as "belonging" in the final four. The only outlier of course would be Syracuse. But if UVA had not choked, most of these metric/analytic systems would have been dead on accurate for the Final Four.

The final issue to understand is that these "measures" are probabilities....not definitive, etched in stone laws. Just like every year, we have had some "upsets" in the tournament. These upset occurrences happen because of "one game sample sizes". If each matchup in the NCAA tournament was a "best of 3 series" instead of a one game, winner take all.....then you would see far fewer upsets and the advances analytics would look even better/more accurate.