Article - Who passes the Natty Test ? | The Boneyard

Article - Who passes the Natty Test ?

Interesting article. My eyeball test suggests Texas has a slightly more credible shot than UCLA or SC — they beat both of them in the regular season after all. But Tulin prefers TCU to them. And to be honest, I find it hard to take TCU seriously. They play in a weak conference, BE level weak, and don’t appear to know what defense is half the time. They have the stats to pass these tests. But their conference leads me to doubt the credibility of their defensive stats.
 
He didn't provide any actual numbers, so I ran the statistics for Connecticut. The formula for estimated possessions gives about 2496, which gives them a turnover rate of 17.4%, 13th in the country. (LSU, by the way, is at 18.95%, almost at his elimination threshold).
Connecticut barely clears the rebound threshold at .542. I find the whole premise of eliminating teams based on only five factors a little shaky, although leading the country in defense (both by points and FG%, as well as steals) and being second in scoring (and 1st in assists, A/TO, FG%, and 3pt%) should allow them to overcome a comparative weakness in rebounding.
 
fomalhaut said:
"I will always believe automatic bids should be exempt from the First Four."
I couldn't agree with him more. Make the last eight at-large teams play-in.
Expand ...
Even though I agree here I wish that the number of conferences would be cut down. There are to many with 9, 10, 11 teams. With expansion destroying the look of college sports maybe one day the NCAA will just say every has to have so many teams. If that’s legal. They currently have a minimum I believe.
 
.-.
This is just repackaging Dean Oliver’s Four Factors, but with a twist:
  • Dean Oliver’s paradigm is based in the aggregate: the team with a better Four Factors dashboard in the aggregate wins.
Texas has a reputation (other people’s words, not mine). So, it’s not as if TX is not contesting shots, but the author says they aren’t as successful as you would have expected.
  • This happened in its game with UCLA. Both UCLA and TX are good Four Factors teams (see TX v UCLA line), with TX just being a little bit &1 better on forcing turnovers (which resulted in a FGA differential which was the difference in the game).
  • &1 UCL 65, TEX 76 - FINAL
I would add an intangible — the ability to do what it takes when things aren’t going as they normally do.
  • TCU has demonstrated it doesn’t have it (its loss to West Virginia when Miles got into foul trouble which she occasionally does).
 
I just checked today that no team in women's NCAA history has ever led Division 1 in steals while also leading in either scoring defense or field goal percentage defense (much less both).
 
Bone Dog said:
Interesting article. My eyeball test suggests Texas has a slightly more credible shot than UCLA or SC — they beat both of them in the regular season after all. But Tulin prefers TCU to them. And to be honest, I find it hard to take TCU seriously. They play in a weak conference, BE level weak, and don’t appear to know what defense is half the time. They have the stats to pass these tests. But their conference leads me to doubt the credibility of their defensive stats.
Expand ...
How do you figure B12 is BE level weak?
 
This is a kind of unique way at looking at things though. It's a nice addition to the other stat models. Doesn't supersede anything, but the TCU outlier is a little odd.
 
.-.
fomalhaut said:
He didn't provide any actual numbers, so I ran the statistics for Connecticut. The formula for estimated possessions gives about 2496, which gives them a turnover rate of 17.4%, 13th in the country. (LSU, by the way, is at 18.95%, almost at his elimination threshold).
Connecticut barely clears the rebound threshold at .542. I find the whole premise of eliminating teams based on only five factors a little shaky, although leading the country in defense (both by points and FG%, as well as steals) and being second in scoring (and 1st in assists, A/TO, FG%, and 3pt%) should allow them to overcome a comparative weakness in rebounding.
Expand ...
Once the factors were presented it was pretty obvious the relative weak link for Uconn was rebounding, and if we were excluded because of one that would be it. Are we an above average rebounding team compared to the entire Division 1 universe? Of course, but the vast majority of true contenders are better than us on the boards.

That is largely by choice., The small ball that Uconn frequently plays to extreme helps on other measures but clearly hurts on this one. Every minute Sarah is at the 5, Blanca at the 4, or Ash is at the 3 contributes to it. If we have Serah or Jana at the 5, Sarah at the 4 and Blanca at the 3, their natural positions, then of course that relative weakness is largely eliminated, with expected tradeoffs in other categories.

We don't play natural 5's for anywhere close to 40 minutes a game, basically don't use a backup 4 at all, and start a 2 at the 3. When we are not using a natural 5 the solution is replace those minutes with a 2 at the 3, and move a 3 to the 4 (Blanca) and a 4 to the 5 (Sarah). That is three players used at one position bigger than their best one, with the expected consequences when it comes to rebounding.

I suspect nothing will change in the early rounds of the tournament, but against some teams using the small ball strategy may have to be tweaked a bit IMO.
 
You must log in or register to reply here.

Forum statistics

Threads
167,685
Messages
4,534,975
Members
10,411
Latest member
RussellSage


Top Bottom