| CollegeHighway.com Login |
| Don't have an account yet? You can create one. As registered user you have some advantages like theme manager, comments configuration and post comments with your name. |
| Trippin? |
 |
| Check Yourself |
 |
| Ephemerids |
One Day like Today...
|
| Welcome |
| You are Anonymous user. You can register for free by clicking here. |
 |
|
| Home / College Guide / Re: Union @ Cornell |
| Posted on Wednesday, February 08 @ 00:00:09 PST |
Re: Union @ Cornell
Date: February 06, 2023 03:07PM
Carlo Ugolini 73, nickname The Magician, was a classic Harkness-era recruit, scrappy if not big (I loved how at 5-foot-7 I could look him straight in the eyes) and, legend has it, he talked his way into a number of second assists and possibly one when he was in the penalty box. (Hey, if Uri Geller could bend a spoon with his nose...) Sixty-nine points senior year, didnt even make the All-America team. Cornell played fewer games then but we beat a lot of early-season opponents by 9-4, 7-1 kinds of scores.ACM
upprdeck
I dont know I never realized that the major becomes a new PP after a goal is scored.. Maybe its because we never seemed to score on them..
did anyone look up fastest 4 goal stretches.. 4 in a min is pretty crazy
Three goals in 22 seconds might be a Cornell record for fastest 3 goals. The Cornell Hockey Record Book (image attached) shows 3 goals in 24 seconds in a game back in 1976, but the Record Book only goes through the 1986-87 season, and I dont have the means (or the desire) to check the intervening 36 years.
Ive never researched fastest 4 goals. I do recall Carlo Ugolini scoring 4 goals in 53 seconds in a freshman game, though.
John Fumio the same era, same size and energy -- think Topher Scott size, but in the era of plaid slacks and leather coats -- who took a shot in the face, lost a tooth, and Im trying to recall if he finished the game. This is the era before dental implants made human teeth more expendable.
Arthur, about Record Book stats from the Harkness-Bertrand-McCutcheon-Reycroft coaching era, there is -- will? -- always be someone with a desire to better capture history. When he was at Cornell, Donald Kagan History wrote *four volumes* on the outbreak of the Peloponnesian War. One of these is more important.
Re: Union @ Cornell
Date: February 06, 2023 03:36PM
a database would make this pretty easy to figure out and compute without having to go look for stuff. I know we have something here is listed even if it doesnt work.
Re: Union @ Cornell
Date: February 06, 2023 03:47PM
* Which makes 2023 a nice outlierupprdeck
I dont know I never realized that the major becomes a new PP after a goal is scored.. Maybe its because we never seemed to score on them.
* There can never be a perfect (eg 4 goals, or 5) five-minute powerplay. If even one tick of the five-minute timer is left, theres a one-second PP opportunity that you dont score on.
* On the plus side, a team that cant score on a 5-minute power play is oh-for-one, not oh-for-2.5.
Edited 1 time(s). Last edit at 02/06/2023 03:49PM by billhoward.
Re: Union @ Cornell
Date: February 06, 2023 03:50PM
But, cf, Union v Colgate, 2/3/2023billhoward
* Which makes 2023 a nice outlierupprdeck
I dont know I never realized that the major becomes a new PP after a goal is scored.. Maybe its because we never seemed to score on them.
* There can never be a perfect (eg 4 goals, or 5) five-minute powerplay. If even one tick of the five-minute timer is left, theres a one-second PP opportunity.
___________________________
quality tweets| ALAB Series podcast| Hostile Witness podcast| Mic Dicta podcast archives| Freelance clip file
Re: Union @ Cornell
Date: February 06, 2023 04:09PM
Carnegie Mellon did have a team - for 3 years - just prior to WW II. When it was known as Carnegie Tech.
Pitt also had a team - first when it was known as Western University of Pennsylvania, even before 1900 ... and then two of the same three years CM had a team.
Done quite a bit of research on all the Pittsburgh area schools. Things were quite active a few years with them, and teams in Ohio.
And Ivies would go to Pittsburgh and play tournaments.
Pitt also had a team - first when it was known as Western University of Pennsylvania, even before 1900 ... and then two of the same three years CM had a team.
[
[www.collegehockeynews.com]
Done quite a bit of research on all the Pittsburgh area schools. Things were quite active a few years with them, and teams in Ohio. And Ivies would go to Pittsburgh and play tournaments.
Re: Union @ Cornell
Date: February 06, 2023 04:12PM
upprdeck
a database would make this pretty easy to figure out and compute without having to go look for stuff. I know we have something here is listed even if it doesnt work.
Even if there was a database of every goal, its time and game -- computing fastest in xx span is not easy. I know because I have this data dating back to 2002 for all of college hockey, and the entire NHL, and I havent gotten around to writing a method for computing it yet, because its challenging. And I do this for a living. Game level spans are hard enough - for any floating x game span ... Ive done it for NHL. Havent done it yet for NCAA.
Re: Union @ Cornell
Date: February 06, 2023 06:24PM
Official box scoreconfirms: Cornells 5 goals were scored on 5 shots with nothing else occurring in between except for the 4 faceoffs and 2 wide Cornell shots between goals 2 and 3.
There were not only no other shots, there were no other stoppages. Holy hell.
Re: Union @ Cornell
Date: February 06, 2023 06:27PM
I actually built a database with every Cornell (and opponent) goal since the program restart.upprdeck
a database would make this pretty easy to figure out and compute without having to go look for stuff. I know we have something here is listed even if it doesnt work.
I could write a query and test this. But I have never actually written a query that walks through the data that way (if I did that I could do *a lot* of analysis). Maybe in retirement.
Edited 2 time(s). Last edit at 02/06/2023 06:28PM by Trotsky.
Re: Union @ Cornell
Date: February 06, 2023 07:00PM
Trotsky
confirms: Cornells 5 goals were scored on 5 shots with nothing else occurring in between except for the 4 faceoffs and 2 wide Cornell shots between goals 2 and 3. There were not only no other shots, there were no other stoppages. Holy hell.
We only had four shots on goal in an entire major? Cmon, get it together team.
Re: Union @ Cornell
Date: February 06, 2023 07:47PM
I feel nerdsniped.
Can I get the schema for the db?
I think it’s something like:
For each game order the CU goals by time and generate a row_number
For each game, compute the shortest period between two goals by self-joining on row_number = row_number + 1 and subtracting the timestamps, then taking the min of the results.
Do the same for 2, 3, 4, etc up to the max number of goals in a game. Use a left join to prevent joins beyond the number of possible goals. It’s a manual query build but it’s ok since there’s what 12-15 max goals in a game?
If you get me a schema I’ll try it….
Can I get the schema for the db?
I think it’s something like:
For each game order the CU goals by time and generate a row_number
For each game, compute the shortest period between two goals by self-joining on row_number = row_number + 1 and subtracting the timestamps, then taking the min of the results.
Do the same for 2, 3, 4, etc up to the max number of goals in a game. Use a left join to prevent joins beyond the number of possible goals. It’s a manual query build but it’s ok since there’s what 12-15 max goals in a game?
If you get me a schema I’ll try it….
Re: Union @ Cornell
Date: February 06, 2023 07:53PM
Ill give you the whole db, its Access. I have been happily giving it away for years and strongly suspect it is why the SIDs records got a lot better (without acknowledgement because TRADITION!).BMac
I feel nerdsniped.
Can I get the schema for the db?
I think it’s something like:
For each game order the CU goals by time and generate a row_number
For each game, compute the shortest period between two goals by self-joining on row_number = row_number + 1 and subtracting the timestamps, then taking the min of the results.
Do the same for 2, 3, 4, etc up to the max number of goals in a game. Use a left join to prevent joins beyond the number of possible goals. It’s a manual query build but it’s ok since there’s what 12-15 max goals in a game?
If you get me a schema I’ll try it….
Re: Union @ Cornell
Date: February 06, 2023 08:00PM
BMac
I feel nerdsniped.
Can I get the schema for the db?
I think it’s something like:
For each game order the CU goals by time and generate a row_number
For each game, compute the shortest period between two goals by self-joining on row_number = row_number + 1 and subtracting the timestamps, then taking the min of the results.
Do the same for 2, 3, 4, etc up to the max number of goals in a game. Use a left join to prevent joins beyond the number of possible goals. It’s a manual query build but it’s ok since there’s what 12-15 max goals in a game?
If you get me a schema I’ll try it….
Theres fancier ways to do this - with WINDOW functions. But if I told you, Id have to kill you.
Re: Union @ Cornell
Date: February 06, 2023 09:18PM
If you can figure out the right way to do this with a window function, good on you. I spent a little time thinking about it and figured it was easier to just build progressive CTEs manually for each goal amount since we don’t need more than 12-15ish….
I’m sure it’s possible though. I’ve even done regressive sql before and felt like a golden god
I’m sure it’s possible though. I’ve even done regressive sql before and felt like a golden god
Re: Union @ Cornell
Date: February 06, 2023 09:38PM
Im in favor of having all stats dating to Beebe Lake days being digitized and query-able. By somebody else. Im sort of in favor of handing them off to Chatbot GPT.
There is a danger that people having to write a Wednesday hockey column or fill dead air while theres yet another video replay will clamp onto a numerical pattern that has been calculated to several decimal places yet has no actual value: Say, Cornells record in one-goal games played after Coach has come back from an illness or game suspension, when Cornell is wearing white jerseys in Friday road games. You hear NFL announcers discussing whether to go for it on fourth down when the team has been 3 of 5 on recent non-punting fourth downs with less than 2 yards to go as if 60% odds guarantee the next one will work.
I did like the magic of the Celtics Bill Russell speaking to the mystery of whether the game clock reading :02 (pre-tenths on the clock) has 2.1 seconds or 2.9 seconds and youre down by one.
That was poetry, not data.
There is a danger that people having to write a Wednesday hockey column or fill dead air while theres yet another video replay will clamp onto a numerical pattern that has been calculated to several decimal places yet has no actual value: Say, Cornells record in one-goal games played after Coach has come back from an illness or game suspension, when Cornell is wearing white jerseys in Friday road games. You hear NFL announcers discussing whether to go for it on fourth down when the team has been 3 of 5 on recent non-punting fourth downs with less than 2 yards to go as if 60% odds guarantee the next one will work.
I did like the magic of the Celtics Bill Russell speaking to the mystery of whether the game clock reading :02 (pre-tenths on the clock) has 2.1 seconds or 2.9 seconds and youre down by one. That was poetry, not data.
Re: Union @ Cornell
Date: February 06, 2023 10:09PM
Im working on it, dammit.billhoward
Im in favor of having all stats dating to Beebe Lake days being digitized and query-able. By somebody else. Im sort of in favor of handing them off to Chatbot GPT.
Re: Union @ Cornell
Date: February 06, 2023 10:15PM
I dont think anyway its designed is gonna be that big a deal.
just ball park it say 75 yrs at 30 games so 2000 games at like at best 4 goals a game so 8000 rows.
any query for fastest 4 goals can ignore all the non 4+ goal games so probably cuts it by 1/2 - 2/3
just store the goals in order and go 1 to 4, 2 to 5, stop with x+3 greater than max goals..
probably could run thru it for any number of goals in a few seconds.
a query with a sub query for the +y.
just ball park it say 75 yrs at 30 games so 2000 games at like at best 4 goals a game so 8000 rows.
any query for fastest 4 goals can ignore all the non 4+ goal games so probably cuts it by 1/2 - 2/3
just store the goals in order and go 1 to 4, 2 to 5, stop with x+3 greater than max goals..
probably could run thru it for any number of goals in a few seconds.
a query with a sub query for the +y.
Re: Union @ Cornell
Date: February 06, 2023 10:22PM
The goals are already ordered. There are 7,103 of them for Cornell.
I had no intention of using database queries to do the work, I was planning on outputting the records to a data file and then writing a simple C++ program to do it. (Yes, I know, Im old, be happy its not PL/1 and get away from me with your heathen post-1983 data structures).
I was assuming I would run through them all, create a table and for each row record count n the # of uninterrupted Cornell goals it was and then dump the time elapsed for the prior 1, 2, .
.. n-1 Cornell goals into a linked list.
Then take that table and do the simple metrics.
I like thorough solutions. With 7k records the run time will be trivial.
I had no intention of using database queries to do the work, I was planning on outputting the records to a data file and then writing a simple C++ program to do it. (Yes, I know, Im old, be happy its not PL/1 and get away from me with your heathen post-1983 data structures).
I was assuming I would run through them all, create a table and for each row record count n the # of uninterrupted Cornell goals it was and then dump the time elapsed for the prior 1, 2, ... n-1 Cornell goals into a linked list.
Then take that table and do the simple metrics.
I like thorough solutions. With 7k records the run time will be trivial.
Edited 11 time(s). Last edit at 02/06/2023 10:34PM by Trotsky.
Re: Union @ Cornell
Date: February 07, 2023 11:14AM
So sure has been discussed before but is there any chance in relatively near future that UPenn and Columbia get D-I hockey programs and Ivy can form a new 8-team conference with an autobid? Then Atlantic and ECAC can maybe merge and improve?
Hello, new poster here 05 grad now in Boston. 2003 was my sophomore year, had season tix.
Thought that is how things were and would make Frozen Four every few years lol. Was in Buffalo for the national semifinal loss. Been at Lynah East, Brown & MSG games for most years since. Was at Placid last time Red there few years ago, sat next to Vanderlaans dad for that loss. Happy to join you all here on eLynah!
Hello, new poster here 05 grad now in Boston. 2003 was my sophomore year, had season tix. Thought that is how things were and would make Frozen Four every few years lol. Was in Buffalo for the national semifinal loss. Been at Lynah East, Brown & MSG games for most years since. Was at Placid last time Red there few years ago, sat next to Vanderlaans dad for that loss. Happy to join you all here on eLynah!
Edited 2 time(s). Last edit at 02/07/2023 06:16PM by 617BigRed.
Re: Union @ Cornell
Date: February 07, 2023 11:35AM
Welcome.
There has never been any talk of Columbia going above club level.
Penn has a great barn. It used to be a perennial suggestion that they might restore their program, which was still active in the mid-70s, but other than a very dedicated and tiny Penn hockey cult I know of no serious measure.
IMO it is more likely Brown drops D-1 someday than Penn adds.
There has never been any talk of Columbia going above club level.
Penn has a great barn. It used to be a perennial suggestion that they might restore their program, which was still active in the mid-70s, but other than a very dedicated and tiny Penn hockey cult I know of no serious measure.
IMO it is more likely Brown drops D-1 someday than Penn adds.
Re: Union @ Cornell
Date: February 07, 2023 12:18PM
Penn brings back hockey.. does not rejoin the ivy because they want to have scholies. they become decent as an independent because they are located close to so many. This forces the ivies to get out of the dark ages
Re: Union @ Cornell
Date: February 07, 2023 01:48PM
Pretty sure you only need 6 teams in a conference for the auto-bid.617BigRed
So sure has been discussed before but is there any chance in relatively near future that UPenn and Columbia get D-I hockey programs and Ivy can form a new 8-team conference with an autobid?!
I guess that leads to the modified question of why the ECAC doesnt split into two conferences so they get two auto-bids. Teams could still play exactly the same regular season schedule - calling ivy vs. non-ivy games OOC, just hold two post-season tournaments
Re: Union @ Cornell
Date: February 07, 2023 03:14PM
doesnt it take a yr or 2 for this to happen though
Re: Union @ Cornell
Date: February 07, 2023 03:41PM
BMac
If you can figure out the right way to do this with a window function, good on you.
I spent a little time thinking about it and figured it was easier to just build progressive CTEs manually for each goal amount since we don’t need more than 12-15ish….
I’m sure it’s possible though. I’ve even done regressive sql before and felt like a golden god
Ive done it with Window functions. Just not on the event level. Only on a game level (i.e. most goals in any 20-game rolling span). In fact, this is (half)-implemented on College Hockey News right now. I say half, because its finding X-game spans from beginning or end of season -- which is slightly easier than ANY rolling x-game span, though has its own challenges.
Event level would be the same thing, in essence.
Re: Union @ Cornell
Date: February 07, 2023 05:08PM
What a rush. Thank you!marty
[13 minutes and 13 seconds]for your entertainment.
Re: Union @ Cornell
Date: February 07, 2023 06:37PM
DL
What a rush. Thank you!marty
[13 minutes and 13 seconds]for your entertainment.
I didnt see Nick Young put up his dukes to incur the penalty - I was in section L on Saturday. But I had seen someone else belt one of the Cornell players earlier in the period in a similar move. Couldnt feel bad for Union after realizing that at least one of the soon to be formerly known as Dutchman was getting a well deserved penalty.
Re: Union @ Cornell
Date: February 07, 2023 06:51PM
adamw
upprdeck
a database would make this pretty easy to figure out and compute without having to go look for stuff. I know we have something here is listed even if it doesnt work.
Even if there was a database of every goal, its time and game -- computing fastest in xx span is not easy. I know because I have this data dating back to 2002 for all of college hockey, and the entire NHL, and I havent gotten around to writing a method for computing it yet, because its challenging. And I do this for a living. Game level spans are hard enough - for any floating x game span ... Ive done it for NHL. Havent done it yet for NCAA.
Serious suggestion, not trolling.
Ask ChatGPT to write the code (or do the task)
From
[platform.openai.com]
Using source data to write database functions.
Just as a human programmer would benefit from understanding the database structure and the column names, Codex can use this data to help you write accurate query requests. In this example we insert the schema for a database and tell Codex what to query the database for.
# Table albums, columns = [AlbumId, Title, ArtistId]
# Table artists, columns = [ArtistId, Name]
# Table media_types, columns = [MediaTypeId, Name]
# Table playlists, columns = [PlaylistId, Name]
# Table playlist_track, columns = [PlaylistId, TrackId]
# Table tracks, columns = [TrackId, Name, AlbumId, MediaTypeId, GenreId, Composer, Milliseconds, Bytes, UnitPrice]
# Create a query for all albums by Adele
Re: Union @ Cornell
Date: February 07, 2023 06:58PM
You think that if you retire or scale back to half-time, youll have time.
Except youll do things such as sleep til eleven that suck up your day.Trotsky
Im working on it, dammit.billhoward
Im in favor of having all stats dating to Beebe Lake days being digitized and query-able. By somebody else. Im sort of in favor of handing them off to Chatbot GPT.
My plan was to organize a million photos -- paper, film, slide, old diplomas and yearbooks -- which I started in 2014 while working full-time, I did maybe 10,000 images and in that period added ~50K digital images. Organizing hockey stats or photos takes a stable personality. More correctly, an Augean Stable personality.
Re: Union @ Cornell
Date: February 07, 2023 07:16PM
Hand around that joint. If theres any left.upprdeck
Penn brings back hockey.. does not rejoin the ivy because they want to have scholies. they become decent as an independent because they are located close to so many. This forces the ivies to get out of the dark ages
The Pennsylvania Class of 1923 (mostly them) funded the rink that opened in 1970. The rink Wiki page notes it holds 2,500 (for hockey / indoor football and further notes it is the second largest college rink in the state behind only Penn States Pegula Arena (6,014). By 1978 mens varsity hockey was gone.
Penn made a bunch of spending cuts that year, many were rescinded, not hockey.
Penn has a womens student organization hockey team; that seems to sound like club hockey.
If hockey came back, itd have to be two teams, not one. Its not impossible. The NHL is as far south as Florida, Arizona, California. Other than Arizona State, Princeton sits at 40 degrees North latitude, Lindenwood (I had to look it up) in St. Lous at 39 degrees. So its not crazy that Penn wouldnt at some day come back. Maybe Maryland and Rutgers would give it a good in the Big10. If I was a Class of 23 at Penn, Id be pissed my donation mattered for less than a decade. IIRC, there was a donor-VIP box, glassed in, on stilts (?), at the rink.
Re: Union @ Cornell
Date: February 07, 2023 07:22PM
More stats is sort of good. But theres other historical-more-than-statisical stuff: when Clarkson sweaters went from okay to terrible, the year Union players allegedly got in some sort of racist/religious mischief, the highest temperatures in visitor locker rooms (did Harkness retire that award), the longest road trips by bus, attendance vs. rink capcity, the geo-center of the ECAC (I tried that once drawing lines and the center is pretty close to Albany), etcetera.
Also which rinks have and which viewer feeds get overhead goal cam views.
Re: Union @ Cornell
Date: February 07, 2023 07:35PM
billhoward
Also which rinks have and which viewer feeds get overhead goal cam views.
This just reminded me. RPI used to have 2 over each goal. One official cam and one RPI TV cam.
Maybe Ursaminor can comment, but did the RPI TV free live feed die? Are they allowed to archive the games?
Re: Union @ Cornell
Date: February 07, 2023 07:41PM
Its ready to be split:nshapiro
I guess that leads to the modified question of why the ECAC doesnt split into two conferences so they get two auto-bids.
- 6 in the Ivy Division
- 6 in the Safety Division
If the Bobcats had a sense of humor -- they do or did have the Ed McMahon School of Broadcasting -- fan T-shirts would say The Last Shall Be First.
Re: Union @ Cornell
Date: February 07, 2023 07:43PM
Probably. Fricking students were embarrassing professional video crews.marty
billhoward
Also which rinks have and which viewer feeds get overhead goal cam views.
This just reminded me. RPI used to have 2 over each goal. One official cam and one RPI TV cam.
Maybe Ursaminor can comment, but did the RPI TV free live feed die? Are they allowed to archive the games?
Sorry, only registered users may post in this forum.
|
|
| |
|
|
