Interesting statistical question

Printable View

02-11-2019, 07:47 AM
rtnorthrup

Interesting statistical question

Duke has already played two conference games against teams ranked 1st or tied for 1st this season: winning both games against UVa. Theoretically, Duke could play two more times against teams ranked 1st or tied for 1st in the conference (UNC). Has any team ever won 4 conference games against teams ranked 1st or tied for 1st in the league?
02-11-2019, 10:59 AM
Phredd3

Quote:

Originally Posted by rtnorthrup

Duke has already played two conference games against teams ranked 1st or tied for 1st this season: winning both games against UVa. Theoretically, Duke could play two more times against teams ranked 1st or tied for 1st in the conference (UNC). Has any team ever won 4 conference games against teams ranked 1st or tied for 1st in the league?

Trivially, the answer to this is almost certainly yes. Even if you discount that everyone begins tied for the #1 spot, after just one game, half of them are. Co-leaders are common through much of the early schedule. In order to have a meaningful metric, you'd have to specify some kind of limiting criteria to make it meaningful (say, alone or tied for the league lead after the halfway point [Duke would not qualify, since the first UVA win was only 5 games into the ACC season], or at least tied for the ACC lead and ranked in the DI top 10 [Duke might qualify, if we win and UNCheat doesn't drop in national or ACC ranking before we get to them]). Then of course you'd have to select criteria that aren't just specific to the team you are interested in because you want a good result to "prove" your point. What was the stat about Haley Gorecki I was hearing last week? I think it was that she was the only player in the ACC to be averaging 18 points, 6 rebounds, 3 assists, and 3 steals (or something like that - not sure about the actual numbers). When you get a criteria list long and that specific enough, it tends to just prove that all players (or teams) are different, rather than that player X (team X) is awesome.

Statistically proving awesomeness is actually kind of hard.
02-11-2019, 11:02 AM
HereBeforeCoachK

Quote:

Originally Posted by Phredd3

Trivially, the answer to this is almost certainly yes. Even if you discount that everyone begins tied for the #1 spot, after just one game, half of them are. Co-leaders are common through much of the early schedule. In order to have a meaningful metric, you'd have to specify some kind of limiting criteria to make it meaningful (say, alone or tied for the league lead after the halfway point [Duke would not qualify, since the first UVA win was only 5 games into the ACC season], or at least tied for the ACC lead and ranked in the DI top 10 [Duke might qualify, if we win and UNCheat doesn't drop in national or ACC ranking before we get to them]). Then of course you'd have to select criteria that aren't just specific to the team you are interested in because you want a good result to "prove" your point. What was the stat about Haley Gorecki I was hearing last week? I think it was that she was the only player in the ACC to be averaging 18 points, 6 rebounds, 3 assists, and 3 steals (or something like that - not sure about the actual numbers). When you get a criteria list long and that specific enough, it tends to just prove that all players (or teams) are different, rather than that player X (team X) is awesome.

Statistically proving awesomeness is actually kind of hard.

True...lotta games every season between 1-0 and 2-0 teams and so forth...
02-11-2019, 11:25 AM
CameronBornAndBred

Quote:

Originally Posted by HereBeforeCoachK

True...lotta games every season between 1-0 and 2-0 teams and so forth...

I think the OP might be on to something, though, because the number of teams sharing the top spot begins to quickly fall off. So to reach having played and beaten 4 teams that are at least tied for first gets pretty tricky, unless you are in a season with an incredible amount of parity. It takes both having a really good team, and a schedule that is lucky enough to make it work out. Teams that only play either unc or UVA only once don't get that chance. In fact, neither the Heels nor the Cavaliers get that chance this season because they only play each other once.
02-11-2019, 11:43 AM
NSDukeFan

Quote:

Originally Posted by Phredd3

Trivially, the answer to this is almost certainly yes. Even if you discount that everyone begins tied for the #1 spot, after just one game, half of them are. Co-leaders are common through much of the early schedule. In order to have a meaningful metric, you'd have to specify some kind of limiting criteria to make it meaningful (say, alone or tied for the league lead after the halfway point [Duke would not qualify, since the first UVA win was only 5 games into the ACC season], or at least tied for the ACC lead and ranked in the DI top 10 [Duke might qualify, if we win and UNCheat doesn't drop in national or ACC ranking before we get to them]). Then of course you'd have to select criteria that aren't just specific to the team you are interested in because you want a good result to "prove" your point. What was the stat about Haley Gorecki I was hearing last week? I think it was that she was the only player in the ACC to be averaging 18 points, 6 rebounds, 3 assists, and 3 steals (or something like that - not sure about the actual numbers). When you get a criteria list long and that specific enough, it tends to just prove that all players (or teams) are different, rather than that player X (team X) is awesome.

Statistically proving awesomeness is actually kind of hard.

You might want to take a peak at the Zion thread.
02-11-2019, 11:47 AM
CDu

Quote:

Originally Posted by CameronBornAndBred

I think the OP might be on to something, though, because the number of teams sharing the top spot begins to quickly fall off. So to reach having played and beaten 4 teams that are at least tied for first gets pretty tricky, unless you are in a season with an incredible amount of parity. It takes both having a really good team, and a schedule that is lucky enough to make it work out. Teams that only play either unc or UVA only once don't get that chance. In fact, neither the Heels nor the Cavaliers get that chance this season because they only play each other once.

UNC has a chance to do the same thing: they beat NC State and Va Tech when each was tied for first, and will get Duke twice.

Though it's worth noting that (1) UNC has to beat UVa for this to be relevant for us, and (2) if we beat UNC this weekend, it's less likely that UNC will be in or tied for first the next time we play them.
02-11-2019, 01:02 PM
Phredd3

Quote:

Originally Posted by NSDukeFan

You might want to take a peak at the Zion thread.

I said "hard". But sometimes the awesomeness is so awesome, it just kinda falls off the page. :cool:
02-11-2019, 05:41 PM
sagegrouse

I was gonna post a question, and this thread is on a similar track.

"How many times has a Final Four match-up been the FOURTH meeting between two teams?"

I found one, and it wasn't what I expected. I expected Georgetown-Villanova in 1985 -- but they only played three times. One case was Georgetown-St. John's in 1985, where Georgetown won the national semis beating St. John's at Rupp Arena, after beating St. John's in the Big East finals at MSG. The two teams split in the regular season, each winning on the other's home court (well, Capital Centre and MSG).

I also thought there might have been four games between Virginia and UNC in 1981, when they met in the Final Four -- but the two teams did not play at the ACC's.

Also, there are many examples, I expect, back in the days of the Dixie Classic and the Big Four tournament, because the Big Four could play three times in the regular season and again in the ACC's -- which is why I specified "had to meet in the Final Four." I can't remember when the Big Four stopped, but Duke and UNC played four times in 1978-1979, splitting the games.
02-11-2019, 05:43 PM
CameronBornAndBred

Quote:

Originally Posted by CDu

UNC has a chance to do the same thing: they beat NC State and Va Tech when each was tied for first, and will get Duke twice.

Though it's worth noting that (1) UNC has to beat UVa for this to be relevant for us, and (2) ̶i̶f̶ when we beat UNC this weekend, it's less likely that UNC will be in or tied for first the next time we play them.

FIFY
02-11-2019, 05:47 PM
uh_no

Quote:

Originally Posted by sagegrouse

I was gonna post a question, and this thread is on a similar track.

"How many times has a Final Four match-up been the FOURTH meeting between two teams?"

I found one, and it wasn't what I expected. I expected Georgetown-Villanova in 1985 -- but they only played three times. One case was Georgetown-St. John's in 1985, where Georgetown won the national semis beating St. John's at Rupp Arena, after beating St. John's in the Big East finals at MSG. The two teams split in the regular season, each winning on the other's home court (well, Capital Centre and MSG).

I also thought there might have been four games between Virginia and UNC in 1981, when they met in the Final Four -- but the two teams did not play at the ACC's.

Also, there are many examples, I expect, back in the days of the Dixie Classic and the Big Four tournament, because the Big Four could play three times in the regular season and again in the ACC's -- which is why I specified "had to meet in the Final Four." I can't remember when the Big Four stopped, but Duke and UNC played four times in 1978-1979, splitting the games.

I'd like our chances over UNC in the final four this year :)
02-11-2019, 05:48 PM
BLPOG

Quote:

Originally Posted by sagegrouse

I was gonna post a question, and this thread is on a similar track.

"How many times has a Final Four match-up been the FOURTH meeting between two teams?"

I found one, and it wasn't what I expected. I expected Georgetown-Villanova in 1985 -- but they only played three times. One case was Georgetown-St. John's in 1985, where Georgetown won the national semis beating St. John's at Rupp Arena, after beating St. John's in the Big East finals at MSG. The two teams split in the regular season, each winning on the other's home court (well, Capital Centre and MSG).

I also thought there might have been four games between Virginia and UNC in 1981, when they met in the Final Four -- but the two teams did not play at the ACC's.

Also, there are many examples, I expect, back in the days of the Dixie Classic and the Big Four tournament, because the Big Four could play three times in the regular season and again in the ACC's -- which is why I specified "had to meet in the Final Four." I can't remember when the Big Four stopped, but Duke and UNC played four times in 1978-1979, splitting the games.

Well, the first that comes to mind would be:

Saturday, January 27, 2001 in College Park
#2 Duke v. #8 Maryland - W, 98-96

Tuesday, February 27, 2001 in Durham
#2 Duke v. #16 Maryland - L, 80-91

Saturday, March 10, 2001 in Atlanta (ACC semifinal)
#3 Duke v. #11 Maryland - W, 84-82

Saturday, March 31, 2001 in Minneapolis (National semifinal)
#1 Duke v. #11 Maryland - W, 95-84
02-11-2019, 05:50 PM
tbyers11

Quote:

Originally Posted by sagegrouse

I was gonna post a question, and this thread is on a similar track.

"How many times has a Final Four match-up been the FOURTH meeting between two teams?"

I found one, and it wasn't what I expected. I expected Georgetown-Villanova in 1985 -- but they only played three times. One case was Georgetown-St. John's in 1985, where Georgetown won the national semis beating St. John's at Rupp Arena, after beating St. John's in the Big East finals at MSG. The two teams split in the regular season, each winning on the other's home court (well, Capital Centre and MSG).

I also thought there might have been four games between Virginia and UNC in 1981, when they met in the Final Four -- but the two teams did not play at the ACC's.

Also, there are many examples, I expect, back in the days of the Dixie Classic and the Big Four tournament, because the Big Four could play three times in the regular season and again in the ACC's -- which is why I specified "had to meet in the Final Four." I can't remember when the Big Four stopped, but Duke and UNC played four times in 1978-1979, splitting the games.

Apologies if I missed this but your search for Final Four matchups where it was the 4th meeting between teams was exclusive of Duke-Maryland in 2001 because that was assumed, right?

EDIT: credit to BLPOG for faster fingers