FHM12 Database problem - standard game

no way · 04-10-2026, 12:44 PM

So I'm working on Access to mysql to website project. I did run the standard modern day game save with maximum leagues active to create as large database as possible to stress test the interface and noticed couple problems with internal database and csv exports:
First tournaments (olympic dunno if any other) have wrong teams assigned - case McDavid he certainly did not play for Montreal Canadiens. this gets exported into player_skater_career_stats_po.csv and screws up the data since Montreal is valid TeamId Notice third line: two lines for 2027 one with teamId 27 and one with TeamId 0.
PlayerId Year Team Id League Id GP G A PIM +/- PP G PP A SH G SH A GR GWG SOG FO FO W HIT GvA TkA SB TOI PPTOI SHTOI Fights Fights Won
14785 2028 23 0 10 5 8 0 -1 2 3 0 0 78 2 29 181 101 7 4 2 13 10845 1151 340 0 0
14785 2027 23 0 16 13 15 2 13 2 5 0 0 86 1 81 365 184 19 4 6 19 19088 2169 344 0 0
14785 2027 0 0 5 4 4 0 8 1 1 0 0 82 2 28 91 50 8 2 3 1 5584 572 135 0 0
14785 2026 23 0 4 0 2 0 -8 0 0 0 0 56 0 10 67 34 2 1 0 0 4458 392 0 0 0
14785 2025 23 0 13 5 11 4 -3 1 4 0 0 71 0 49 275 143 18 6 2 7 14588 1979 121 0 0

See first 2 pictures. You can see the marked line have no boxscore data (time on Ice section is calculated from boxscores instead of career stats)
This I can clean out I guess by not importing lines with 0 TOI after boxscore calculation but that would kill all lines for leagues that have data but are not playable. still better to remove them than keep this in.

Next Issue (and this one is pretty much game breaking for this project) is thousand of players have PlayerId = -1 so no chance to clean this at all as opposed to previous one. After 4 years of simming: 2068 players have -1 Id
What this creates is player page for Id-1 is completely insane, and some teams in NCAA have like hundreds if not thousends lines in the game boxscore because of the duplicate player Id which behaves like this if called upon in SQL join querry

I plan to make this available for use once it's done but this is quite game breaking. I can upload the save file if needed.

no way · 04-10-2026, 03:56 PM

Some aditional info regarding the "-1" Id players
2025/26 - none
2026/27 - 755
2027/28 - 1502
2028/29 - 2068
2029/30 - 2068
dunno what happens to behave like this but some of them do legit play for some playable teams.

no way · 04-21-2026, 12:48 PM

Simmed into season 2063/64 now and there is now 14851 players with Id -1
And for example scoring csv have 2432 instances where goal scorer, Assist1 and Assist2 are all -1. That is pretty much broken at this point
I mean tell me how do you sort this out?
This could use some cleaning up.

04-10-2026, 12:44 PM	#1
no way Major Leagues Join Date: Dec 2012 Location: Plzeň, Czech Republic Posts: 301	FHM12 Database problem - standard game So I'm working on Access to mysql to website project. I did run the standard modern day game save with maximum leagues active to create as large database as possible to stress test the interface and noticed couple problems with internal database and csv exports: First tournaments (olympic dunno if any other) have wrong teams assigned - case McDavid he certainly did not play for Montreal Canadiens. this gets exported into player_skater_career_stats_po.csv and screws up the data since Montreal is valid TeamId Notice third line: two lines for 2027 one with teamId 27 and one with TeamId 0. PlayerId Year Team Id League Id GP G A PIM +/- PP G PP A SH G SH A GR GWG SOG FO FO W HIT GvA TkA SB TOI PPTOI SHTOI Fights Fights Won 14785 2028 23 0 10 5 8 0 -1 2 3 0 0 78 2 29 181 101 7 4 2 13 10845 1151 340 0 0 14785 2027 23 0 16 13 15 2 13 2 5 0 0 86 1 81 365 184 19 4 6 19 19088 2169 344 0 0 14785 2027 0 0 5 4 4 0 8 1 1 0 0 82 2 28 91 50 8 2 3 1 5584 572 135 0 0 14785 2026 23 0 4 0 2 0 -8 0 0 0 0 56 0 10 67 34 2 1 0 0 4458 392 0 0 0 14785 2025 23 0 13 5 11 4 -3 1 4 0 0 71 0 49 275 143 18 6 2 7 14588 1979 121 0 0 See first 2 pictures. You can see the marked line have no boxscore data (time on Ice section is calculated from boxscores instead of career stats) This I can clean out I guess by not importing lines with 0 TOI after boxscore calculation but that would kill all lines for leagues that have data but are not playable. still better to remove them than keep this in. Next Issue (and this one is pretty much game breaking for this project) is thousand of players have PlayerId = -1 so no chance to clean this at all as opposed to previous one. After 4 years of simming: 2068 players have -1 Id What this creates is player page for Id-1 is completely insane, and some teams in NCAA have like hundreds if not thousends lines in the game boxscore because of the duplicate player Id which behaves like this if called upon in SQL join querry I plan to make this available for use once it's done but this is quite game breaking. I can upload the save file if needed. Attached Images

04-10-2026, 03:56 PM	#2
no way Major Leagues Join Date: Dec 2012 Location: Plzeň, Czech Republic Posts: 301	Some aditional info regarding the "-1" Id players 2025/26 - none 2026/27 - 755 2027/28 - 1502 2028/29 - 2068 2029/30 - 2068 dunno what happens to behave like this but some of them do legit play for some playable teams. Attached Images

04-21-2026, 12:48 PM	#3
no way Major Leagues Join Date: Dec 2012 Location: Plzeň, Czech Republic Posts: 301	Simmed into season 2063/64 now and there is now 14851 players with Id -1 And for example scoring csv have 2432 instances where goal scorer, Assist1 and Assist2 are all -1. That is pretty much broken at this point I mean tell me how do you sort this out? This could use some cleaning up. Attached Images