WHAT - is this really a stat one or the so called metrics? Paralysis by analysis
There is almost the same gap between 4 and 5 as 4 and 1.
was just going to post that but figured I'd wait and let him yell at someone else
My eyes bulged out when I saw that, been wondering what it was. Sounds crazy to me also. Can someone explain the significance, and maybe exemplify how it is used?
The goal post is about to move. Just wait.
www.cbssports.com
Check out the other metrics.
So Bethune Cookman will determine Uconn's seeding?I had posted about WAB a few weeks ago because I didn't focus on it before and didn't really know what it is. In this linked article is a good explanation of how it's calculated.
The NCAA's seeding committee will use Wins Above Bubble (dubbed WAB) for the second year in a row. WAB is all about the resume. What have you done on the court? WAB measures what the average bubble team would have done, and each team is given a boost or a reduction based on the outcome. Missouri can beat Bethune-Cookman by eight or 80; the final score does not matter.
How is this calculated? Simple math. The average bubble team is given a 95% chance of beating Bethune-Cookman at home. WAB is calculated from the gap between the projected win percentage and the actual win number of 1.0. Since Missouri won, it was given 0.05 WAB for that victory. If it had lost, it would have been a -0.95 WAB in the ledger.
Trotter's Trends: Auburn's Keyshawn Hall might be best transfer scorer ever; WAB lessons
Which teams have built the best resumes through two months according to the NCAA's new Wins Above Bubble metric
Here is a website that tracks WAB if you ever want to see where teams are:
Very little, honestly not sure we'll have anything more than 1 more loss little. We simply have no way to make up for it. And I don't think we can lose that StJ game. Getting swept by another team is a pretty strong suggestion.
It doesn’t seem like you’re trying to understand the metric. The basic idea is that it assesses the quality of a team’s wins, rather than the quality of a team’s performance, which is what predictive metrics like Ken Pomeroy’s rankings do. The baseline is what a a hypothetical average bubble team would do. I think we all understand that a road win against Kansas means a whole more than a home win against New Haven; this is just one way to quantify it. According to this, an average bubble team would have had a 13% chance to win at Kansas. Our win there counts as 0.87 WAB. An average bubble team would have had a 99% chance at beating New Haven, so our win was 0.01 WAB.
So how is the % determined? Is it a bunch of monkeys at a keyboard?
I think the key word is Hypothetical.
I honestly don’t know the exact details. But any number of sites give win probabilities for individual matchups. Even if the baseline for comparison differed, this metric should be similar in relative ranking. It doesn’t have to be perfect to provide useful information about a team, as long as nobody uses it as the only measure. Since it is impossible to watch every team play every game, mixing in WAB, along with something like KenPom, paints a more complete picture of a team‘s performance and overall talent level than W-L record or a subjective AP poll ranking.
Last year for our schedule it was 20.7 wins expected for the bubble team. But for Auburn in the record breaking SEC+Maui, it was only 15.3 wins expected for a bubble team against that schedule.
The win percentage for each game is set by the NET ratings of the opponents on your schedule and the strength of the hypothetical baseline (which I believe for the NCAA's version is set by the average strength of the historical 45th team in the NET), which get plugged into a common win probability formula tuned to college basketball.
Thx...so it's all based on the NET. So if the Net is a flawed metric, everything that's based on it is also flawed.
