Comparison Of Talent Distributions From FOF2K4 to FOF2K7

[Ben E Lou]

A few people have commented on the difference in talent in the two games. I just checked the csv.txt files from the two, and they confirm that a significant change has been made, and I assume that this will hold true for game-generated draft files as well.

FOF2004
9 - Rare Superstar - 0.3%
8 - Superstar - 1.4%
7 - Near Superstar - 1.8%
6 - Elite Starter - 1.7%
5 - Excellent Starter - 3.5%
4 - Very Good-Plus Starter - 4.3%
3 - Very Good Starter - 6.0%
2 - Good Starter - 16.2%
1 - Replacement-Level Starter (decent backup) - 27.2%
0 - Roster Filler - 37.6%

FOF2007
9 - Rare Superstar - 0.3% (nc)
8 - Superstar - 0.4% (-1.0%)
7 - Near Superstar - 0.6% (-1.2%)
6 - Elite Starter - 1.1% (-0.6%)
5 - Excellent Starter - 1.5% (-2.0%)
4 - Very Good-Plus Starter - 2.6% (-1.7%)
3 - Very Good Starter - 3.9% (-2.1%)
2 - Good Starter - 12.2% (-4.0%)
1 - Replacement-Level Starter (decent backup) - 23.9% (-3.3%)
0 - Roster Filler - 53.4% (+15.8%)


So, we're talking about approximately 1/8 fewer players who are good or better--less than I would have guessed, but definitely significant. This will no doubt have an impact on how we build rosters and allocate salaries to top-tier guys.

[yabanci]

I had noticed this too, and it's a good thing. The question will be whether the AI understands that a 45/52 rated player is quite valuable and shouldn't always be released when the new season starts.

On sort of a side note, this is where I think a lot of the AI roster building problems came with FOF2004. The AI would release most of the players in the 40-55 range. Then it would sign players in this range to contracts with bonuses. Then the next year would come around and the AI would release those same players, leaving dead cap space. Then the cycle would continue -- release players, dead cap room, sign same quality players, release players, more dead cap room, and so on.

This is obviously one of the most difficult areas to program because there are so many factors and you can't program the AI teams to reason like humans, but hopefully there will be some improvement with the new version.

[Peregrine]

I had noticed this too, and it's a good thing. The question will be whether the AI understands that a 45/52 rated player is quite valuable and shouldn't always be released when the new season starts.

On sort of a side note, this is where I think a lot of the AI roster building problems came with FOF2004. The AI would release most of the players in the 40-55 range. Then it would sign players in this range to contracts with bonuses. Then the next year would come around and the AI would release those same players, leaving dead cap space. Then the cycle would continue -- release players, dead cap room, sign same quality players, release players, more dead cap room, and so on.

This is obviously one of the most difficult areas to program because there are so many factors and you can't program the AI teams to reason like humans, but hopefully there will be some improvement with the new version.



I think what the AI understands is definitely improved. I've seen the AI teams putting franchise tags on a lot of players in the 40-50 range, and when I check their stats, they have been putting up good numbers, so it seems to actually be taking field performance into account here.

[yabanci]

I think what the AI understands is definitely improved. I've seen the AI teams putting franchise tags on a lot of players in the 40-50 range, and when I check their stats, they have been putting up good numbers, so it seems to actually be taking field performance into account here.

Good to hear. I haven't even simmed a full season yet, so I wasn't sure.

[nilodor]

I think the advicated reduction in talent is a good thing, especially in the draft. One thing of note if you look at the players.fdt file that came with FOF2K7 the splits are a little different.

Rating - # - % (Note there are 1753 players in total)
0 - 595 - 33.9
1 - 474 - 27
2 - 314 - 17.9
3 - 154 - 8.8
4 - 70 - 4
5 - 55 - 3.1
6 - 32 - 1.8
7 - 35 - 2
8 - 16 - 0.9
9 - 8 - 0.5