Arkansas Post Game Thread | Page 7 | The Boneyard

Arkansas Post Game Thread

I just ran into Danny O at Stew Leonards. I said "great game last night." He replied "It got a little boring." I guess when sports is your living you can get a bid jaded :p
 
navery12 said:
That's just a summary, you have to subscribe to get the full breakdown. It's basically saying if you were to replay the game with every player taking the exact same shots that's what would be expected to happen with Arkansas scoring more than they did and us scoring less
Expand ...
From the (surface level) lack of credibility of their conclusions I would neither waste any time nor money on a subscription.

As far as your comment "they would ls score more, we would score less", yes, that may be true but to what level? If that game is replayed exactly as it unfolded 1,000 times they make make it a 10-12 point game once.

I also wonder if anything is factored in for the shot alterations Clingan caused (and later their unwillingness to even attempt something from in close when he was there). A layup attempt from 4 ft away may look like an automatic make in the computer but if your biggest concern is a giant swatting it away, rushing a bad shot could be the norm.
 
Waquoit said:
Fun With Numbers
Expand ...
This account must not take into account WHO is taking the shot. Arkansas had a lot of open 3s and long 2s, but UConn wanted them to take those because they are bad at shooting 3s.
 
navery12 said:
That's just a summary, you have to subscribe to get the full breakdown. It's basically saying if you were to replay the game with every player taking the exact same shots that's what would be expected to happen with Arkansas scoring more than they did and us scoring less
Expand ...

That reminds me of the Duke fans post-game March 29 1999 "If we played you ten times we would win nine of them"
 
FfldCntyFan said:
From the (surface level) lack of credibility of their conclusions I would neither waste any time nor money on a subscription.

As far as your comment "they would ls score more, we would score less", yes, that may be true but to what level? If that game is replayed exactly as it unfolded 1,000 times they make make it a 10-12 point game once.

I also wonder if anything is factored in for the shot alterations Clingan caused (and later their unwillingness to even attempt something from in close when he was there). A layup attempt from 4 ft away may look like an automatic make in the computer but if your biggest concern is a giant swatting it away, rushing a bad shot could be the norm.
Expand ...
It's a really good site and I find it very useful

According to their numbers, the adjustment would happen on both sides. So 13 more poinrs for Arkansas and 14 less for UConn. I think that's a bit aggressive for some of the reasons that don't get captured like you said with Clingan and Sanogo affecting shots in the paint and getting in the opponents heads
 
navery12 said:


This is... something
Expand ...


I know we used this guy's SQ analyses to make ourselves feel better in January but the output from his modeling is very very SUS at this point.

In no valid sim model do you plug the data from that game (shot locations, make probabilities, etc., etc.) and come out with an analysis that Arkansas wins 6-7 out of 10 games played with the stats data from the game.

LOL NERDS!!!!

Ellie Kemper Nerd GIF by The Office
 
.-.
navery12 said:
That's just a summary, you have to subscribe to get the full breakdown. It's basically saying if you were to replay the game with every player taking the exact same shots that's what would be expected to happen with Arkansas scoring more than they did and us scoring less
Expand ...

That's utter rubbish. This dude's sim model is seriously flawed.

I would love to see the data output for how many times the sim has Andre missing his easy dunk off the Sanogo feed. If it's any more than 1% of the time, he should throw his computer in the garbage!

office space nothing to see here motherfucker GIF
 
Yanksrock615 said:
This account must not take into account WHO is taking the shot. Arkansas had a lot of open 3s and long 2s, but UConn wanted them to take those because they are bad at shooting 3s.
Expand ...
It does take into account who is taking the shot, but one thing it doesn't do is take into account who is defending it, only the defender distance. Big difference having Clingan in the way vs. a Mitchell twin.

But yeah, the stat is called Shot Quality, but despite the revised score difference, it does say UConn had the better shots. But it was the turnovers and free throws that makes it think that Arkansas should have scored more points and that UConn shot better than usual and so combined effect changes the outcome.

But yeah. Think it's wrong. Extended garbage time probably messes with the model.
 
Last edited:
The irony of the shot quality theory is that Dan Hurley/staff/our guys game planned and scouted their team..We gave Arkansas the shots-- certain shots --because our scout revealed that's how we beat these guys. And yet.. Shot Quality gurus say that if Arkansas made the high quality shots they were given.. They should have won the game..

We won..
 
.-.
Actually, seeing the dumb way they come up with that metric, it all makes sense now. During the January rut, SQ was probably saying AJ was expected to score 15 more points per game from his wide open 3 point attempts.
 
All algorithms have their flaws and last night exposed SQ's. No biggie, because it was a 20 point blowout

I'm usually a numbers nerd but I really haven't liked that Shot Quality page all year. When you win, it shows you that you should have lost. When you lose, it shows you that if you made your easy baskets you would have won. Lose/lose
 
FfldCntyFan said:
They really don't break anything down. They claim Sanogo should have scored fewer points and we should have scored less points on layups. Is this comparing overall season performance vs last night? If yes, it may work mathematically but the implication is that as the game was layed they should have won. That's a very tough sell
Expand ...
I’m confused as to why they weren’t the lower seed if they were expected to win.
 
.-.
navery12 said:
They break it down a bit in this tweet, though it doesn't really pass the test for me. Watching live it felt like we had way more quality shots and open shots than the tweet says

Expand ...

There are lies, damnable lies, and statistics.
 
Umami Zen said:
The irony of the shot quality theory is that Dan Hurley/staff/our guys game planned and scouted their team..We gave Arkansas the shots-- certain shots --because our scout revealed that's how we beat these guys. And yet.. Shot Quality gurus say that if Arkansas made the high quality shots they were given.. They should have won the game..

We won..
Expand ...
And the shots that Arkansas got were NOT good shots. They were almost all contested, and the closer to the basket, the more contested. Mid range jumpers are some of the toughest shots in the game. Ironic that the announcers were talking about the extremely high level of difficulty on the shots Arkansas was getting, yet SQ claims they were good shots Arkansas should have made. Its one or the other, it can't be both.
 
navery12 said:
This is... something
Expand ...


They had a follow up explaining it.

UConn had better quality shots.

Bama had more possessions.

Flawed as UConn had a number of TOS after the score had ballooned to nearly thirty. 8-4 Arkansas advantage there after UConn was up 29. So only 3 fewer tos.

Also though they had 9 more shots they outshot UConn 5-2 with the 10-15 players out there. So actually only 6 more shots.

Also 4-2 advantage with bench players in (last 3 mins). So only 4 extra FTs.

I’d imagine if you went back and looked at all those numbers from before UConn went up 29 UConn would be even more advantageous.
 
Last edited:
caw said:


They had a follow up explaining it.

UConn had better quality shots.

Bama had more possessions.

Flawed as UConn had a number of TOS after the score had ballooned to nearly thirty. 8-4 Arkansas advantage there after UConn was up 29. So only 3 fewer tos.

Also though they had 9 more shots they outshot UConn 5-2 with the 10-15 players out there. So actually only 6 more shots.

Also 4-2 advantage with bench players in (last 3 mins). So only 4 extra FTs.

I’d imagine if you went back and looked at all those numbers from before UConn went up 29 UConn would be even more advantageous.
Expand ...

You mean the players that were instructed to just run the clock?
 
  • Like
Reactions: caw
.-.
What a game - if these last few games don't prove Dan Hurley can plan and execute with this team I don't know what will. The guys were awesome, Arkansas tried hard to disrupt and it worked... for a minute.

Just watched the UCLA/Zags game and the Zags seem a bit like our team so should be an interesting game. !!!
 
You must log in or register to reply here.

Forum statistics

Threads
168,204
Messages
4,556,801
Members
10,442
Latest member
Virginiafan


Top Bottom