Rating Fighter Zero

[Cap]

Curious how all you guys rate your fighters. Do you, for instance have a boxer zero against whom all other boxers are compared? Someone so bad that everyone else uses him for a doormat, never mind a stepping stone? I would assume that, on the other hand, you'd have Ali or Louis as the absolute best of the heavyweights, so everyone else falls somewhere in between.

Is Fighter Zero, if he exists in your alternate universe, a real person? I recall seeing a boxer in an old Ring Record Book who lost every single match he was in, most by knockout. Now, there's dedication to your craft.

Cap

[bear]

Originally Posted by Cap

Curious how all you guys rate your fighters. Do you, for instance have a boxer zero against whom all other boxers are compared? Someone so bad that everyone else uses him for a doormat, never mind a stepping stone? I would assume that, on the other hand, you'd have Ali or Louis as the absolute best of the heavyweights, so everyone else falls somewhere in between.

Is Fighter Zero, if he exists in your alternate universe, a real person? I recall seeing a boxer in an old Ring Record Book who lost every single match he was in, most by knockout. Now, there's dedication to your craft.

Cap

Cap
CONN CHRIS created and posted, somewhere, 18 zero rated templates representing gradations in the rating. In creating these templates he utilized Thunder's hypothesis that each gradation be based on a 60% possibility of beating the next lower grade. The lowest possible fighter that can be created by most gamers is just Chris's -7. To create the lower templates one would have to be a data base editor, which I am not. Back when we Beta tested I was too much of a noobie at creating fighters.
Among the templates there are ten additional levels of bad. For example punches landed can be reduced below 22. The absolute zero has CF's of 0, endurance of 0, aggressiveness 0, HP 0, punching 0, counter punching 0, punches missed 78, and KI at 1. All damage inflictions are maxed. A swinging heavy bag could be more dangerous.

[Cap]

Tapadh leat, Bear. I wasn't aware of those templates. Guess they're only useable with version 2.5, eh? I whipped up a bunch of pork & beaners the other day just to pad guys' records and they're pretty bad, but I'll have to try making 'em worse. I had no idea you could lower ratings by that much.

Slainte!
Cap

[bear]

Originally Posted by Cap

Tapadh leat, Bear. I wasn't aware of those templates. Guess they're only useable with version 2.5, eh? I whipped up a bunch of pork & beaners the other day just to pad guys' records and they're pretty bad, but I'll have to try making 'em worse. I had no idea you could lower ratings by that much.

Slainte!
Cap

Yeah, they only work with 2.5. Those of us who post fighters to the forum don't use the templates all that much. To begin doing so now would necessitate a re-rating of all of the current TC's to adjust their relative rankings . Not a project anyone wants to undertake. I snuck in a couple of guys with lower PL numbers in my TC parade when the templates first showed up. May even be one TC embodying each of the template numbers but I don't remember I created them, or just thought about doing so or who they are. If I created a guy with the worst rating it likely would be the store clerk that Ben Johnson clocked in under a minute in a London street. It was reported in some earlier editions of Dowling's Fistiana and I created a bunch of Corinthians a while back. Since he was a civilian rather than a pugilist he became a candidate for the "fighter zero" status. I think the only way fighter zero wins, is if he's nailed with an egregious foul by someone and gets a DQ win.

[professordp]

Concerning the templates, I've been using (and advocating their use) since Chris released them several years ago.

You certainly reduce HP inflation when doing ratings.

[Thunder]

Originally Posted by bear

Cap
CONN CHRIS created and posted, somewhere, 18 zero rated templates representing gradations in the rating. In creating these templates he utilized Thunder's hypothesis that each gradation be based on a 60% possibility of beating the next lower grade. The lowest possible fighter that can be created by most gamers is just Chris's -7. To create the lower templates one would have to be a data base editor, which I am not. Back when we Beta tested I was too much of a noobie at creating fighters.

For the record, it was 62/38 that represented one OR (Overall Rating) point. I got this by exporting the Database to a spreadsheet and separately totalling all the ratings of all the 15s and dividing by the number of fighters to get an "Average" OR15 . . did the same thing to get an Average OR14, Average OR13, etc. down to the 0s.

Then I fought some large number of fights (1000 maybe) of every matchup possible Avg. OR15 vs. Avg. OR14 . . Avg. OR15 vs. Avg. OR13, etc. down to Avg. OR1 vs. Avg. OR0.

Two things happened. First, OR15 vs. OR14 was the smallest sample size and producing odd results (like the 14s were better, or maybe 50-50), so I tossed the data. Second, OR1 vs. OR0 was winning at over an 80% clip, so I recognized there were multiples levels there and lopped all the data involving the 0s.

So I combined all the remaining +1 matchup results from OR14 vs. OR13 down through OR2 vs. OR1 and the average winning pct. of the superior fighter in those 13, +1 match-ups was 62/38, so I used that as a constant.

I took my average OR1 chap and designed a fighter he would beat at a 62/38 and that became my average "True Zero." Then I fiddled with another new guy that OR0 would beat 62/38 and he became OR-1 (Overall Rating minus-one) . . and so on until I got to the standard bottom of all the ratings, which was OR-7.

I think those are the first batch of files Chris posted. Then he broke the mold and got all the way down to -18. If you go the other way, what would it take to beat an OR15 at a 62% clip, then go to the next level that would beat that guy at a 62% clip . . I think you can get up to around overall ratings in the 30s.

Technically speaking, Muhammad Ali has an functional overall rating of 22 already since the "real" 0-rated fighter is functionally a -7. I wouldn't mind if 2.6 or 3.0 re-figured what was shown in the overall rating category and if the top rated fighters are 22 . .or 25, 30, whatever, I'd be cool with that.

Also, the shackles need to come off with regard to rating limits. George Foreman has HP: 14, but we can't rate a fighter HP: 14, well, why not dammit? Actually, I dislike that rating and would never make anyone a 14, but that's another kettle of fish.

But how about defense of -12 or +12? Even without a programming glitch, you can rate 25 levels of offensive accuracy (PL 1-22 through PL 1-46), but only 13 levels of defense (+6 to -6). Why?

Let everyone use ratings like Punches Landed 1-10 or some of the other ratings Chris used in his -8 to -18 slobs. As it stands now, the ratings are too compacted to accurately produce realistic results for the kinds of fighters that are being rated.

No matter how much data I can produce that says most of the fighters in the database that are rated OR0 are, in fact, still over-rated or that you really can't accurately rate someone like Jukebox Timebomb, given the constrictions of the system, people aren't going to stop doing it. So at least let us easily make those guys as bad as they really are.

[Mark42661]

Where can these templates be found now?

Thanks

[PWillisTheMan]

Originally Posted by Thunder

For the record, it was 62/38 that represented one OR (Overall Rating) point. I got this by exporting the Database to a spreadsheet and separately totalling all the ratings of all the 15s and dividing by the number of fighters to get an "Average" OR15 . . did the same thing to get an Average OR14, Average OR13, etc. down to the 0s.

Then I fought some large number of fights (1000 maybe) of every matchup possible Avg. OR15 vs. Avg. OR14 . . Avg. OR15 vs. Avg. OR13, etc. down to Avg. OR1 vs. Avg. OR0.

Two things happened. First, OR15 vs. OR14 was the smallest sample size and producing odd results (like the 14s were better, or maybe 50-50), so I tossed the data. Second, OR1 vs. OR0 was winning at over an 80% clip, so I recognized there were multiples levels there and lopped all the data involving the 0s.

So I combined all the remaining +1 matchup results from OR14 vs. OR13 down through OR2 vs. OR1 and the average winning pct. of the superior fighter in those 13, +1 match-ups was 62/38, so I used that as a constant.

I took my average OR1 chap and designed a fighter he would beat at a 62/38 and that became my average "True Zero." Then I fiddled with another new guy that OR0 would beat 62/38 and he became OR-1 (Overall Rating minus-one) . . and so on until I got to the standard bottom of all the ratings, which was OR-7.

I think those are the first batch of files Chris posted. Then he broke the mold and got all the way down to -18. If you go the other way, what would it take to beat an OR15 at a 62% clip, then go to the next level that would beat that guy at a 62% clip . . I think you can get up to around overall ratings in the 30s.

Technically speaking, Muhammad Ali has an functional overall rating of 22 already since the "real" 0-rated fighter is functionally a -7. I wouldn't mind if 2.6 or 3.0 re-figured what was shown in the overall rating category and if the top rated fighters are 22 . .or 25, 30, whatever, I'd be cool with that.

Also, the shackles need to come off with regard to rating limits. George Foreman has HP: 14, but we can't rate a fighter HP: 14, well, why not dammit? Actually, I dislike that rating and would never make anyone a 14, but that's another kettle of fish.

But how about defense of -12 or +12? Even without a programming glitch, you can rate 25 levels of offensive accuracy (PL 1-22 through PL 1-46), but only 13 levels of defense (+6 to -6). Why?

Let everyone use ratings like Punches Landed 1-10 or some of the other ratings Chris used in his -8 to -18 slobs. As it stands now, the ratings are too compacted to accurately produce realistic results for the kinds of fighters that are being rated.

No matter how much data I can produce that says most of the fighters in the database that are rated OR0 are, in fact, still over-rated or that you really can't accurately rate someone like Jukebox Timebomb, given the constrictions of the system, people aren't going to stop doing it. So at least let us easily make those guys as bad as they really are.

I like the raising and lowering of the defense. A lot of the pure Boxers lose more than they would because they don't inflict enough damage. A -15 for a Whitaker or Mayweather would be pretty accurate compared to what it's like for guys to try and hit other fighters.

[Thunder]

Originally Posted by Mark42661

Where can these templates be found now?

Thanks

Good question. I would have thought in this sub forum, but I'm not finding them. I'll touch base with Chris and hopefully we can find them.
If not, I'm sure one of us will be able to post them again.

[Thunder]

Originally Posted by PWillisTheMan

I like the raising and lowering of the defense. A lot of the pure Boxers lose more than they would because they don't inflict enough damage. A -15 for a Whitaker or Mayweather would be pretty accurate compared to what it's like for guys to try and hit other fighters.

Agreed. I was going to start a new thread on this, but I haven't gotten around to it. Plus Paul recently said he was going to cap ratings where they were for 2.6. I do have what I think is a very good case for expanding defensive ratings though . . . maybe tomorrow . . . and maybe it can be changed for 3.0!

[Mark42661]

Originally Posted by Thunder

For the record, it was 62/38 that represented one OR (Overall Rating) point. I got this by exporting the Database to a spreadsheet and separately totalling all the ratings of all the 15s and dividing by the number of fighters to get an "Average" OR15 . . did the same thing to get an Average OR14, Average OR13, etc. down to the 0s.

Then I fought some large number of fights (1000 maybe) of every matchup possible Avg. OR15 vs. Avg. OR14 . . Avg. OR15 vs. Avg. OR13, etc. down to Avg. OR1 vs. Avg. OR0.

Two things happened. First, OR15 vs. OR14 was the smallest sample size and producing odd results (like the 14s were better, or maybe 50-50), so I tossed the data. Second, OR1 vs. OR0 was winning at over an 80% clip, so I recognized there were multiples levels there and lopped all the data involving the 0s.

So I combined all the remaining +1 matchup results from OR14 vs. OR13 down through OR2 vs. OR1 and the average winning pct. of the superior fighter in those 13, +1 match-ups was 62/38, so I used that as a constant.

I took my average OR1 chap and designed a fighter he would beat at a 62/38 and that became my average "True Zero." Then I fiddled with another new guy that OR0 would beat 62/38 and he became OR-1 (Overall Rating minus-one) . . and so on until I got to the standard bottom of all the ratings, which was OR-7.

I think those are the first batch of files Chris posted. Then he broke the mold and got all the way down to -18. If you go the other way, what would it take to beat an OR15 at a 62% clip, then go to the next level that would beat that guy at a 62% clip . . I think you can get up to around overall ratings in the 30s.

Technically speaking, Muhammad Ali has an functional overall rating of 22 already since the "real" 0-rated fighter is functionally a -7. I wouldn't mind if 2.6 or 3.0 re-figured what was shown in the overall rating category and if the top rated fighters are 22 . .or 25, 30, whatever, I'd be cool with that.

Also, the shackles need to come off with regard to rating limits. George Foreman has HP: 14, but we can't rate a fighter HP: 14, well, why not dammit? Actually, I dislike that rating and would never make anyone a 14, but that's another kettle of fish.

But how about defense of -12 or +12? Even without a programming glitch, you can rate 25 levels of offensive accuracy (PL 1-22 through PL 1-46), but only 13 levels of defense (+6 to -6). Why?

Let everyone use ratings like Punches Landed 1-10 or some of the other ratings Chris used in his -8 to -18 slobs. As it stands now, the ratings are too compacted to accurately produce realistic results for the kinds of fighters that are being rated.

No matter how much data I can produce that says most of the fighters in the database that are rated OR0 are, in fact, still over-rated or that you really can't accurately rate someone like Jukebox Timebomb, given the constrictions of the system, people aren't going to stop doing it. So at least let us easily make those guys as bad as they really are.

This argument is very, very compelling and makes a lot of sense.

[Thunder]

Originally Posted by Mark42661

Where can these templates be found now?

Thanks

Page 125

http://www.ootpdevelopments.com/boar...-cans-125.html

[Mark42661]

Originally Posted by Thunder

Page 125

http://www.ootpdevelopments.com/boar...-cans-125.html

Thank you so much!

[Thunder]

Originally Posted by Mark42661

Thank you so much!

No sweat. I had a lot of fun figuring stuff out about the game engine with those dudes as well as creating them.

[IceTea]

Originally Posted by Mark42661

Where can these templates be found now?

Thanks

I printed them off some time ago.