Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*)

Moderators: Clyde Frazier, Doctor MJ, trex_8063, penbeast0, PaulieWal

trex_8063
Forum Mod
Forum Mod
Posts: 12,609
And1: 8,238
Joined: Feb 24, 2013
     

Re: Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*) 

Post#21 » by trex_8063 » Thu Jun 3, 2021 9:59 pm

70sFan wrote:I may look wrong, but it seems that you only updated Lakers game up to the end of 3rd quarter.


Check it again [maybe everything hadn't saved to the cloud or something when you looked??]. It's finished to the final score [99-113, Bulls]; last entry is on line 477 of spreadsheet.
"The fact that a proposition is absurd has never hindered those who wish to believe it." -Edward Rutherfurd
"Those who can make you believe absurdities, can make you commit atrocities." - Voltaire
70sFan
RealGM
Posts: 29,912
And1: 25,248
Joined: Aug 11, 2015
 

Re: Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*) 

Post#22 » by 70sFan » Thu Jun 3, 2021 10:16 pm

trex_8063 wrote:
70sFan wrote:I may look wrong, but it seems that you only updated Lakers game up to the end of 3rd quarter.


Check it again [maybe everything hadn't saved to the cloud or something when you looked??]. It's finished to the final score [99-113, Bulls]; last entry is on line 477 of spreadsheet.

It's available now, thanks! :)
trex_8063
Forum Mod
Forum Mod
Posts: 12,609
And1: 8,238
Joined: Feb 24, 2013
     

Re: Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*) 

Post#23 » by trex_8063 » Mon Jun 7, 2021 2:20 am

Thanks again for your help and generosity on games for me to log, 70sFan. I'll see about logging those oldies you PM'ed me soon. I just start logging Game 5 of the '72 Finals [Knicks @ Lakers], though, so I'll finish that first I think (new spreadsheet has been added to OP, where I'll keep games from '72 and '73).

btw, general question to you [or ANYONE who knows the answer]: In the event of a team rebound where the ball goes out of bounds after a missed shot......do they credit a rebound to the player [on the team that gets possession] who was closest to the ball when it went out?
To this point, I've just been logging them as "Team Rebounds" and leaving it at that. But I'm wondering if, for my completed boxscores, if that's how they typically do those things [commentary on the game I'm presently logging would suggest so]. Does anyone know for sure?
"The fact that a proposition is absurd has never hindered those who wish to believe it." -Edward Rutherfurd
"Those who can make you believe absurdities, can make you commit atrocities." - Voltaire
trex_8063
Forum Mod
Forum Mod
Posts: 12,609
And1: 8,238
Joined: Feb 24, 2013
     

Re: Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*) 

Post#24 » by trex_8063 » Sun Jun 13, 2021 2:37 pm

70sFan wrote:.


Good day sir; thanks again for the video suggestions. Was going to make some comments about the recently completed G5 of the '72 Finals, though will also reply to a few things you'd PM'ed about....

Game 5 of the '72 Finals
Man, Wilt has a BEAST of a game here. The game is only about 85% complete (last 6-7 minutes are absent), but already at that point Wilt has 22 points @ 70.3% TS (which is near-monstrous efficiency by 1972 standards), 20 rebounds, 4 assists [plus 2 FT assists], 5 blocks, and 0 turnovers. Wow.

The other guy for the victorious Lakers who somewhat quietly has a really good game going is Happy Hairston: 13 points @ 53.3% TS (good for '72), 15 rebounds, 3 assists, only 1 turnover.

Jerry West and Gail Goodrich both have sort of sub-par games (certainly by their standards, anyway).

Jim McMillian has a sort of weird "polarized" game.....some really good aspects of his statline, though one really bad aspect too:
He finished [as of where the video cuts the game off] with 18 points @ 59.2% TS, 5 rebounds, [1 FT assist], 5 steals, 1 block, plus 2 offensive fouls drawn and 1 "great D/hustle play" where he forces a turnover.
However, he also had a game-high 6 turnovers.
Can't decide if that's a good game overall; not too bad anyway.


For the Knicks, it's Walt Frazier (and to a lesser degree Dave DeBusschere) who kinda pop visually.
Frazier has a very strong game going statistically: at the point where the video cuts off he had 22 points @ 51.9% TS, 8 rebounds, 9 assists [plus 2 FT assists], 5 steals [plus 1 "great D/hustle play" that results in a turnover], only 1 turnover.

DeBusschere has [as usual] an interesting hodge-podge of a statline going: 13 points @ 45.4% TS, 13 rebounds, 2 assists, 3 steals [plus 1 "great D/hustle play"], 1 turnover.

Phil Jackson has a pretty good game off the bench. No one else is playing particularly well for them; Bill Bradley in particular has a stinker of a game: 10 points @ 31.7% TS, 2 reb, 1 ast [plus 1 FT assist], 4 turnovers. Just an awful night for him.


I now have >7 games sample on Walt Frazier, and he's still white-hot from the 10-16' range (63.5%!). I know that's not long-term sustainable, but it's becoming more and more likely that he was in reality a very elite shooter [in an all-time sense] from that range.
I have a little over 4.5 games on Wilt now, and he's finishing a monstrous 86.5% at the rim. However, suggestions that he was solid in the mid-range are starting to look questionable (he's 33.3% from outside of 3' so far: 8 of 26 (30.8%) from 3-10', and 1 for 1 from outside of 10').



Other notes.....
I answered my own question (from post #22 or 23 above) about Team Rebounds. I looked up the PbP of the last Bucks/Nets game: team rebounds are listed just as Team Rebounds on the play-by-play, and ultimately are NOT included in the boxscore. The Bucks in game 3 had 65 rebounds, including 9 Team Rebounds.....but their boxscore shows just 56 rebounds [obv the Team ones not included].
Although it's clear that in the 60s/70s they WERE including those Team Rebounds in the boxscore totals, and just crediting them to whichever player was closest [or sometimes arbitrarily saying "just give it to Wilt"], I'm going to continue to record them as they are on a modern box [that is: SEPARATE from indivdual rebounds].


To answer your PM, yes, I'd like to do those other games from the early 70s, though for now I'm actually logging one of those games from the early 50s you sent me:


I'll put a link up in OP to the "early 50s" gamelog spreadsheet. There are some possessions missing, but I can work with that. The video quality is fairly outstanding for the time-period.
There's no audio, except for when the author of the video offers his commentary.......which I've decided to mute because he's kinda pissing me off with his ignorance (and implied logical inconsistency). But that's another discussion.
"The fact that a proposition is absurd has never hindered those who wish to believe it." -Edward Rutherfurd
"Those who can make you believe absurdities, can make you commit atrocities." - Voltaire
trex_8063
Forum Mod
Forum Mod
Posts: 12,609
And1: 8,238
Joined: Feb 24, 2013
     

Re: Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*) 

Post#25 » by trex_8063 » Sat Jun 19, 2021 11:05 pm

I finished logging a game from 1950! (linked at bottom)......which was pretty fun. Very interesting to watch [imo], and fantastic video quality for a film that old.
I already did a long rant pertaining to commentary by the author of the video, and general comments on how to watch these old games.

Here I'll just note some general observations and some statistical talking points....

Firstly, it's a really different kind of game to watch, and definitely does look a bit "raw" in many ways.
There is A LOT of travelling on the perimeter (the refs allow a fair bit of shuffling the feet at the start of a drive---I wonder if it has anything to do with the poor quality of the floors---but otherwise will NOT allow that third step). Carl Braun has this weird head-fake [he often doesn't move the ball at all, it's just a toss of his head], and then drive; and he travels a lot when he does it.

At first glance, the quality doesn't look terrific, though on closer scrutiny some other suggestions emerge.....

For example, let's look at Bill Henry.
Bill Henry is visibly the biggest guy in this game, and is listed at 6'9" and 215 lbs according to bbref. Take a look at the move and converted reverse lay-up at 9:42 in the video. Not too shabby.
And check out the defensive play at the 8:18 mark in the video: here Henry recognizes the play pretty quickly, makes a very early rotation to help, contests the shot hard forcing a miss, and grabs the defensive rebound. Really nice defensive play. (SIDENOTE: it's also right around this point in the video where the narrator is making his ridiculous claim that Bill Russell wouldn't even be able to stay in the league by the early 80s, btw).

I bring ^^these plays up because.....Bill Henry was a mere role-player for just two seasons ['50 would be his final season]. To be sure, professional basketball offered highly questionable financial prospects, so he may have bowed out of the league for personal reasons [likely, in fact], rather than washing out.

However, it's still worth noting that a 6'9" guy who was coordinated enough for the play at 9:42, had solid enough defensive instincts for the play at 8:18 [and who shot a respectable 67% at the FT line that year] was NOT a star in this 1950 league [or even close to it].
That's gotta say a little something about the quality of the league, no? It certainly serves as yet another counterpoint argument to what the narrator was spewing.

Just as another quick suggestion that a degree of talent [certainly more than the narrator gives them credit for] was required: Clint Wager was 6'6" and 210 lbs with really well-defined legs (can see at 33:17, as well at a jump-ball that follows shortly).......but he was a scrub in '50, which was his one and only season.


One other guy I wanted to specifically point out was Hall of Famer Dick McGuire (#15 for New York). He scores just 2 pts on 1 of 6 shooting [though the one he made was a doozy], but I credit him with 10 assists [and a FT assist] in this game......and there are a handful of very nice passes, imo:
7:46 - no assist here because Gallatin misses the shot, but a really nice look-off pass
20:15 - really fine no-look pass to Vandeweghe (Kiki's dad, fwiw)
32:55 - another look-off assist to Gallatin

His one awesome made FG is at the 20:40 mark.....it's like an "alley-oop" reverse lay-up.
At 24:42 he does a really nice step-back dribble [perhaps not easy under ball-handling rules of the time]; unfortunately he doesn't have a good jumper or quick release to follow it up with. But I nonetheless thought it was worth pointing out.


One kinda cool play I wanted to point out from Carl Braun [he does have a couple nice hustle blocks, too, fwiw] is at 27:37: pretty nice cross-over under the rules of the time, followed up with an actual jump-shot (though he doesn't make it).

There are other legitimate jump-shots [from several different players] to be seen here, too:

18:39 (by Charlie Black)
20:10 (by Ernie Vandeweghe)
26:34 (by Bob Harris)
27:16 (by Dick McGuire [not pretty though])
31:37 (turnaround jumper by John Oldham)

So anyway.....they were happening by then.


And lastly I wanted to look at some collective shot-location data, because some of it was surprising.
For instance, I didn't figure that the era [aside from the present one] where I'd find the MOST shots outside of 25'......would be the pre-shotclock [and pre 3pt-line] era. But they take A LOT from the outside.
In this game [which again is only perhaps 85% complete], I counted 24 attempts from 24+ feet [plus one 22-footer from the corner--->may have been a 3pter in today's league, but here I'm counting those in the 16-23' range].

And they made 7 of 24 (29.2%).......not too shabby, especially considering many of them were WAY out there.
Fred Schaus takes one from about 34-35 feet; no kidding, the edge of the half-court circle was barely one short stride behind his heels when he shoots this. He missed that one, but he later does MAKE one from ~28-29'.
Leo Klier actually MAKES one from just outside of 30 feet (and was 2 for 3 overall from deep, despite ALL THREE being from 27+ feet).
Curly Armstrong took one from >30', too.
John Oldham drills one from outside of 27'.

No question, a lot of these guys had serious range. A 3pt-line would have been a game-changer for this era.

They finished 19 of 31 (61.3%) at the rim.
They were 11 of 30 (36.7%) in the 3-10' range. Not awful, all things considered. Some of these guys were decently accurate in this range with hook-shots.

It was the other mid-ranges where they really struggle (lack of reliable jump-shots for many of them being the reason):
4 of 27 (14.8%) from 10-16'
1 of 19 (5.3%) from 16-23'
[so a combined 5 for 46 (10.9%) from 10-23'.....fairly ugly in there, no question]

Anyway, cool stuff. Thanks for pointing me to it, 70sFan. I'll be logging some early 70s stuff next, though I will get back to the other early games later.

"The fact that a proposition is absurd has never hindered those who wish to believe it." -Edward Rutherfurd
"Those who can make you believe absurdities, can make you commit atrocities." - Voltaire
70sFan
RealGM
Posts: 29,912
And1: 25,248
Joined: Aug 11, 2015
 

Re: Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*) 

Post#26 » by 70sFan » Sun Jun 20, 2021 9:08 am

trex_8063 wrote:And lastly I wanted to look at some collective shot-location data, because some of it was surprising.
For instance, I didn't figure that the era [aside from the present one] where I'd find the MOST shots outside of 25'......would be the pre-shotclock [and pre 3pt-line] era. But they take A LOT from the outside.
In this game [which again is only perhaps 85% complete], I counted 24 attempts from 24+ feet [plus one 22-footer from the corner--->may have been a 3pter in today's league, but here I'm counting those in the 16-23' range].

And they made 7 of 24 (29.2%).......not too shabby, especially considering many of them were WAY out there.
Fred Schaus takes one from about 34-35 feet; no kidding, the edge of the half-court circle was barely one short stride behind his heels when he shoots this. He missed that one, but he later does MAKE one from ~28-29'.
Leo Klier actually MAKES one from just outside of 30 feet (and was 2 for 3 overall from deep, despite ALL THREE being from 27+ feet).
Curly Armstrong took one from >30', too.
John Oldham drills one from outside of 27'.

No question, a lot of these guys had serious range. A 3pt-line would have been a game-changer for this era.

I thought from watching some early 1950s games that they shot a lot more shots from three point range than in later eras and your work seems to back it up. We don't know how well they shot from there (sample of size isn't big enough), but I wouldn't be surprised if they shot above 25% which is good considering that they didn't have the three point line.


They finished 19 of 31 (61.3%) at the rim.
They were 11 of 30 (36.7%) in the 3-10' range. Not awful, all things considered. Some of these guys were decently accurate in this range with hook-shots.

This is better efficiency from inside that I'd expect - maybe I underrate the early 1950s players as well?
pdevos
Ballboy
Posts: 48
And1: 28
Joined: Jan 16, 2017

Re: Game Logs [and Shot Location Data] for old games (reboot: *OFFICIAL*) 

Post#27 » by pdevos » Wed Apr 20, 2022 5:15 am

I'm interested in helping this project along.

I'm pretty good with Python and Databases. Also could build some tools to help people catalog games much more quickly, e.g. voice to text and also create an app with shot chart where if you had an iPad, you could just tap the location while watching the game. Again, all of this would do the digital part for you. I have been contemplating building it for awhile now, just not sure how many games there are on video to do something like this.

I've posted this a few times for years on several forums, but no luck.

Here's my latest on Reddit:
https://www.reddit.com/r/VintageNBA/comments/u7dgjz/re_historic_nba_video_an_audacious_goal_to

Return to Player Comparisons