Game Thread [2022-23] Oral Roberts (NCAA Tournament R64)

[DurhamSon]

@childress22 is there a way to calculate Deff for the non Lively mins? Is Scheyer actually an elite defensive coach or is this just the Anthony Davis effect?

 

[aiw]

From 24-->17 in kenpom. D up to 17 after that

 

[aiw]

We are also now "white" instead of "red" in 3pt% at 33.7% (189)

 

[aiw]

 

[rome8180]

@childress22 is there a way to calculate Deff for the non Lively mins? Is Scheyer actually an elite defensive coach or is this just the Anthony Davis effect?

To be fair, that number could tell you just as much about Young's limitations as Lively's prowess.

 

[best4heisman]

 

[rome8180]

It's hard to believe that isn't parody.

 

[Pantone287]

Well that was awesome. Dominant, way-above-expectations win with our three potential returnees combining for 21 on 9-29. Perfect game.

 

[DurhamSon]

I don't think this is modeling the impact of Lively around the rim as much as it should be, but it matches my game chat intuition that we scored around as much as we would have been expected given our shot profile, and were benefitting, especially to start the game, that ORU couldn't convert on some very good looks from three, which SQ says should have produced 9 more points. Still win if it's something like 40-36 going into halftime, but maybe the 2nd half is a lot more dicey depending on whether that 15-2 run they had was based on us getting complacent going up 27 or not.


If we make it to the 2nd weekend, our offense (Flip) has to get better and fast

 

[LastHearth]

Honestly this just tells me that analytical system needs a lot more work. There is no way. If you gave everyone the same shots again last night, ORU would win 46% of the time.

 

[physicsfactor]

Yeah, I’m not buying that. I’m as much a stats/math geek as most on here, but I thought it was pretty clear that Duke was simply the much better team. Don’t need to attempt to quantify it more than that.

 

[DurhamSon]

I don't agree with the literal statistic as spit out, as I do think for models like Pomeroy, the shittiness of ORU's schedule all year and lack of data from just about any opponent not garbage-tier makes it very difficult if not impossible to get accurate parameter estimations for whatever goes into their model, but the general direction of it I think is useful (i.e with the exact same shot profile repeated 10,000 times last night, median final score is probably something along the lines of Duke 74, ORU 62 or w/e). Open shots with no close-outs are open shots whether they are against a D3 team or the Milwaukee Bucks. For that reason I tend to put a lot more stock into their assessment of perimeter shooting and FTs than anything else

 

[LastHearth]

The general direction might be useful, but the impossibility of gathering the data nullifies any possibility of usefulness.

ORU only played a few quality opponents, you would needs hundreds of seasons of NCAA season and tournament data to have a useful profile of teams that look similar, and really we had had maybe 5-10.

 

[Grit]

Yeah, I'd need to see evidence of shot quality's quality to trust it, because it often doesn't align with my impressions of the shot quality

 

[rome8180]

I watched the first half again without my emotions tied up in it. ORU missed a LOT of open threes and layups. They had like five shots rim out in the first five minutes. Not all of their layups were bothered by Lively either. I 100% buy this, and I think you guys are letting rooting interest influence your impressions here.

 

[rome8180]

I don't think this is modeling the impact of Lively around the rim as much as it should be, but it matches my game chat intuition that we scored around as much as we would have been expected given our shot profile, and were benefitting, especially to start the game, that ORU couldn't convert on some very good looks from three, which SQ says should have produced 9 more points. Still win if it's something like 40-36 going into halftime, but maybe the 2nd half is a lot more dicey depending on whether that 15-2 run they had was based on us getting complacent going up 27 or not.


If we make it to the 2nd weekend, our offense (Flip) has to get better and fast

I guess I'm the only one who saw the game the same way you did.

The saving grace is that if we got ACC Tournament Flip, we would have won by 25 anyway.

 

[dukebluedevils05]

Eh, I watched it last night after knowing the final score and not having read a single thing here or anywhere about the game and it felt to me like we were clearly the better team. I’m surprised the SQ score had it that close. Certainly they missed some open ones, but not enough to have it a single digit game. We had them totally sped up due to our size, length, aggressiveness against their ball screen action, and how hard we were running them off the line. Kinda like how we had Virginia sped up, as Bennett acknowledged after the game, and playing out of their comfort zone in the ACCCG. I don’t think that’s something a tool like SQ can accurately account for.

 

[rome8180]

I feel like I'm taking crazy pills. I'm willing to concede some allowance for fear of Lively's shot blocking. I don't think it was really a one-point game, but I also don't think it was even close to a 23-point game.

But not only does your guys version not match an objective measurement, it also doesn't match my subjective one in the slightest. If it was possible, I would screenshot every single missed shot in the first five minutes. Unfortunately, YTTV has some technology that blocks screenshots. Found that out recently.

I'm okay with a couple people disagreeing with me on this one, but the fact that it's the majority of you is really disappointing. I thought this board was a little less susceptible to clear bias.

 

[rome8180]

Also, "Eh" before you disagree with something is really fucking annoying.

 

[SeanMayTriedToEatMe]

If they’re appropriately factoring in what % the individual player typically shoots on each type of shot from each location with a hand within X feet of the ball at its release point or whatever, then I don’t think there is any eye test to explain away the numbers.

I would question whether this site/company actually does all of that, though, or if it’s just very basic programming, like Whitehead shoots Y% from 3 this season, so he should’ve scored (3PA * Y%) points on his 3PA last night. It must be a really advanced operation to collect all that data and spit it out for every game so quickly. I’m skeptical because I’m pretty engrossed in this stuff as a degenerate gambler and I’ve never heard of anyone paying for a subscription to this thing.