Lahman's Database, hockey version

Long_Long_Name · 11-12-2005, 06:38 PM

I plan on creating a database similar to Lahman's database using the NHL's data. The idea came to me when realizing that we could very easily adjust point totals for different eras, and that several similar operations could be done. What's even better is that since the rinks are all more or less identical, there is no such thing as "rink factor" in hockey.

My problem is that I don't know whether to start in Access or Excel. I'm very familiar with Excel and I see that it's very limited and that I couldn't do a very complete database with it. However, I haven't used Access in a very long time and even though I don't quite know it, it doesn't seem very appropriate to make calculations, adjust data and similar things.

Could anyone help me out here?

canadiancreed · 11-12-2005, 06:50 PM

now this would be an AWESOME idea

I totally endorse this for what it's worth.

Myself I'd go with Excel. it's easier to export to whatever the heck would be needed to use the data, access is well.......ya

Long_Long_Name · 11-12-2005, 06:56 PM

I just have no idea how to query information in Excel... for example, if you have every single year of every player on one line, how can you end up with a player's career stats on just one line, you know?

Also, since era adjustment would require a coefficient which would be the average of all years played in the NHL, it would require adjustment every year... and that would better be automated, and I just don't know how to do that.

canadiancreed · 11-12-2005, 06:59 PM

Quote:

Originally Posted by Long_Long_Name

I just have no idea how to query information in Excel... for example, if you have every single year of every player on one line, how can you end up with a player's career stats on just one line, you know?

Also, since era adjustment would require a coefficient which would be the average of all years played in the NHL, it would require adjustment every year... and that would better be automated, and I just don't know how to do that.

it'd probably require macros in excel, which is beyond my knowledge

access woudl let you do the queries of course. The second line remidns me of stored procedures, which I'm almost certain Access won't let you do. I think they hae some VB-lite thing that will let you extend it to do something liek that but again, beyond my scope

Long_Long_Name · 11-12-2005, 07:17 PM

Basically, al I'm trying to do is to find an era-adjusted version of, say, point totals for all the players ever to have played the game.

As for the era coefficient, instead of changing it every year, I suppose it would be easier to have one number adjusted instead of every year. For example, if the average of goals per game in 1967-68 was 6.5, well, we could divide all of the players' point totals that year by 6.5, and never need to adjust them again. However, what would change would be the coefficient that would be multiplied to the point totals in order to make them more accurate. Let's say that in 1919-1920 the goals per game average was 3.25, well, a point in that year would be worth the double of a goal in 1967-68, and the coefficient's role would only to make the statistics look more realistic, because in reality one could only do GPG/points and find out the "absolute" performance of the player - it would only look unrealistic, but the rates would be accurate.

jamus23 · 11-12-2005, 10:15 PM

Quote:

Originally Posted by Long_Long_Name

What's even better is that since the rinks are all more or less identical, there is no such thing as "rink factor" in hockey.

I love the idea, however I would disagree somewhat with this point. Although the size of the rinks is the same, the quality of the ice is not generally considered equal. I know Dallas, for instance, has typically been considered as having very poor ice, while Edmonton is usually said to have very good ice.

I imagine this can have a potentially significant effect on scoring. High quality ice speeds things up, which probably increases scoring, whereas low quality ice slows things down, and probably makes it harder to score. That's just my opinion, though.

IatricSB · 11-13-2005, 12:54 AM

Like the idea but I don't think Excel is the way to go. Even if not familiar with it, at a minimum I'd go with Access with the idea that someday the data will probably be put into SQL by someone (if you're going to have a web site and the whole nine yards).

Long_Long_Name · 11-13-2005, 12:55 AM

CC, just for your personal knowledge, had Gretzky played his entire career with the 2003-2004 settings (82 games, 5.14 GPG), he'd ended up with 2113 career points. Impressive, certainly, but that's much lower than the 2857 career points he ended up with. He'd be much closer to Howe than he is now, he who would have 1979 points.

Give me a player and I'll tell you how he performed.

Go Tribe · 11-13-2005, 06:38 PM

If it would help, I could make a Catobase for Hockey that doesn't work

Long_Long_Name · 11-13-2005, 07:29 PM

I don't think I have the patience to do that alone... even if I do figure out how...

By the way, attached is the GPG average for every year in NHL history. The average GPG for all of the NHL's history is 6.28.

Sandman7103 · 11-13-2005, 07:36 PM

Quote:

Originally Posted by Long_Long_Name

My problem is that I don't know whether to start in Access or Excel.

I'm fairly certain you can switch back and forth if you need to, so go with what you're comfortable with.

11-12-2005, 06:38 PM	#1
Long_Long_Name Hall Of Famer Join Date: Jan 2002 Location: Montréal Posts: 7,065	Lahman's Database, hockey version I plan on creating a database similar to Lahman's database using the NHL's data. The idea came to me when realizing that we could very easily adjust point totals for different eras, and that several similar operations could be done. What's even better is that since the rinks are all more or less identical, there is no such thing as "rink factor" in hockey. My problem is that I don't know whether to start in Access or Excel. I'm very familiar with Excel and I see that it's very limited and that I couldn't do a very complete database with it. However, I haven't used Access in a very long time and even though I don't quite know it, it doesn't seem very appropriate to make calculations, adjust data and similar things. Could anyone help me out here? __________________ Beta Baseball. Join it!

11-12-2005, 06:50 PM	#2
canadiancreed Hall Of Famer Join Date: Aug 2004 Posts: 11,660	now this would be an AWESOME idea I totally endorse this for what it's worth. Myself I'd go with Excel. it's easier to export to whatever the heck would be needed to use the data, access is well.......ya __________________ PT21 PT22

11-12-2005, 06:56 PM	#3
Long_Long_Name Hall Of Famer Join Date: Jan 2002 Location: Montréal Posts: 7,065	I just have no idea how to query information in Excel... for example, if you have every single year of every player on one line, how can you end up with a player's career stats on just one line, you know? Also, since era adjustment would require a coefficient which would be the average of all years played in the NHL, it would require adjustment every year... and that would better be automated, and I just don't know how to do that. __________________ Beta Baseball. Join it!

11-12-2005, 07:17 PM	#5
Long_Long_Name Hall Of Famer Join Date: Jan 2002 Location: Montréal Posts: 7,065	Basically, al I'm trying to do is to find an era-adjusted version of, say, point totals for all the players ever to have played the game. As for the era coefficient, instead of changing it every year, I suppose it would be easier to have one number adjusted instead of every year. For example, if the average of goals per game in 1967-68 was 6.5, well, we could divide all of the players' point totals that year by 6.5, and never need to adjust them again. However, what would change would be the coefficient that would be multiplied to the point totals in order to make them more accurate. Let's say that in 1919-1920 the goals per game average was 3.25, well, a point in that year would be worth the double of a goal in 1967-68, and the coefficient's role would only to make the statistics look more realistic, because in reality one could only do GPG/points and find out the "absolute" performance of the player - it would only look unrealistic, but the rates would be accurate. __________________ Beta Baseball. Join it!

11-13-2005, 12:54 AM	#7
IatricSB Hall Of Famer Join Date: Dec 2001 Location: California Posts: 3,493	Like the idea but I don't think Excel is the way to go. Even if not familiar with it, at a minimum I'd go with Access with the idea that someday the data will probably be put into SQL by someone (if you're going to have a web site and the whole nine yards). __________________ Life is not a journey to the grave with the intention of arriving safely in a pretty and well preserved body; but rather to skid in broadside, thoroughly used up, totally worn out, and loudly proclaiming, "Wow! What a Ride!" Chicago(N) - Boys of Summer Oakland - 20th Century League Bakersfield - Wild Things Brooklyn - QBA Dodge City - NBSL California - ABC Dodger's Senioriest fan on the OOTP Boards

11-13-2005, 12:55 AM	#8
Long_Long_Name Hall Of Famer Join Date: Jan 2002 Location: Montréal Posts: 7,065	CC, just for your personal knowledge, had Gretzky played his entire career with the 2003-2004 settings (82 games, 5.14 GPG), he'd ended up with 2113 career points. Impressive, certainly, but that's much lower than the 2857 career points he ended up with. He'd be much closer to Howe than he is now, he who would have 1979 points. Give me a player and I'll tell you how he performed. __________________ Beta Baseball. Join it!

11-13-2005, 06:38 PM	#9
Go Tribe All Star Starter Join Date: Dec 2001 Posts: 1,534 Infractions: 1/0 (0)	If it would help, I could make a Catobase for Hockey that doesn't work

11-13-2005, 07:29 PM	#10
Long_Long_Name Hall Of Famer Join Date: Jan 2002 Location: Montréal Posts: 7,065	I don't think I have the patience to do that alone... even if I do figure out how... By the way, attached is the GPG average for every year in NHL history. The average GPG for all of the NHL's history is 6.28. Attached Images __________________ Beta Baseball. Join it!