Warrior General Games

Originally posted by Bukowski
Originally posted by ClemUSA

Would it be possible to run some sql queries to get all kind of statistics and compare them to the NFL (and/or NCAA) ?

Then choose a common ratio that would be a goal in all next fix/improvements.

It's probably not good for the game if we get the same stats than NLF. But it's worse for the game if we get stats like NFL for sacks (or whatever stat) and 3 times as much pass yardage than the NFL.


And this ratio shouldn't be 1, as even if people want realism, they want it with inflated stats, because it's a game. But with a common ratio, it's still realistic.




NO. You cannot manipulate the statistics to fit a certain criteria, the builds/tactics have to decide that.

When will people understand this?

Them trying to manipulate the statistics to please the "realism" people, is what got us into this mess in the first place - don't think it hasn't.

Bukowski is right. Stats in the NFL are created by the physical and mental abilities of the athletes working under the constraints of the laws of physics and the rules of the game. So in order to simulate that in GLB you model the laws of physics and the rules of the game in the simulation and then let the builds determine the stats. If a DE can get a sack on every passing sack, then that's a good build against a bad build. If every DE gets a sack on every passing down, then that's bad modeling and you change the sim. Deciding before hand that DEs should get X sacks per season means that the build and the sim become secondary, and if that's the case, then there is no point to play the game.

T

Originally posted by tsherr

Bukowski is right. Stats in the NFL are created by the physical and mental abilities of the athletes working under the constraints of the laws of physics and the rules of the game. So in order to simulate that in GLB you model the laws of physics and the rules of the game in the simulation and then let the builds determine the stats. If a DE can get a sack on every passing sack, then that's a good build against a bad build. If every DE gets a sack on every passing down, then that's bad modeling and you change the sim. Deciding before hand that DEs should get X sacks per season means that the build and the sim become secondary, and if that's the case, then there is no point to play the game.

T

this should be framed and put on Bort's desk.

It doesn't work like that.

Yes you "model the laws of physics and the rules of the game in the simulation" which means a bunch of mathematical formulas.
But "let the builds determine the stats" is BS.
It's a simulation. So you have to base your modeling on something real. Laws of physics and rules of the game equal certain stats in real life. As GLB is a simulation of competitive professional football the NFL is the natural league to simulate.

So you need something like this:

GLB stats = NFL stats * coefficient

* coefficient obviously >1 because it's a video game people want to see inflated stats
* you have to specify GLB stats, maybe by using only the stats in competitive game. You don't want to have stats from joke games having and influence. That's important because for example you shouldn't reduce efficiency of DE to get sack because they are getting to many sacks against gutted team. You should fix the problem of gutted teams instead. (I think that's a major problem right now, with the sim being defined to accommodate for gutted teams when theses teams shouldn't be part of the game)

Then you have to take the build into the equation. Of course good build should have better stats than average build. Yes good build should beat bad build on the field. But they are just dots. They don't determine the game, the math does. A good build is just a build which characteristic numbers are more optimized for the mathematical formula of the sim than the ones of a bad build.

Originally posted by OttawaShane
Originally posted by tsherr


Bukowski is right. Stats in the NFL are created by the physical and mental abilities of the athletes working under the constraints of the laws of physics and the rules of the game. So in order to simulate that in GLB you model the laws of physics and the rules of the game in the simulation and then let the builds determine the stats. If a DE can get a sack on every passing sack, then that's a good build against a bad build. If every DE gets a sack on every passing down, then that's bad modeling and you change the sim. Deciding before hand that DEs should get X sacks per season means that the build and the sim become secondary, and if that's the case, then there is no point to play the game.

T

this should be framed and put on Bort's desk.



Yep.

Originally posted by ClemUSA
It doesn't work like that.

Yes you "model the laws of physics and the rules of the game in the simulation" which means a bunch of mathematical formulas.
But "let the builds determine the stats" is BS.
It's a simulation. So you have to base your modeling on something real. Laws of physics and rules of the game equal certain stats in real life. As GLB is a simulation of competitive professional football the NFL is the natural league to simulate.

So you need something like this:

GLB stats = NFL stats * coefficient

* coefficient obviously >1 because it's a video game people want to see inflated stats
* you have to specify GLB stats, maybe by using only the stats in competitive game. You don't want to have stats from joke games having and influence. That's important because for example you shouldn't reduce efficiency of DE to get sack because they are getting to many sacks against gutted team. You should fix the problem of gutted teams instead. (I think that's a major problem right now, with the sim being defined to accommodate for gutted teams when theses teams shouldn't be part of the game)

Then you have to take the build into the equation. Of course good build should have better stats than average build. Yes good build should beat bad build on the field. But they are just dots. They don't determine the game, the math does. A good build is just a build which characteristic numbers are more optimized for the mathematical formula of the sim than the ones of a bad build.




Not sure I understand what you're saying. If the builds running under the sim don't determine the stats, but instead you create a stat to aim for and then change the sim to reach these stats then you relegate builds to virtually no importance, and that's not going to be much fun.

As for the idea that the builds "are just dots" actually, the dots are just representations of the results of algorithms from the sim+the builds, so I don't get your send paragraph at all.

However, the bottom line is this: The stats must be created by the interaction between the builds and the sim. Fixing stats that are off (and I mean off for all the plays between two equally matched players) can be fixed by fixing the sim, but changing the sim to make the stats fit a pre-determined number will ruin the fun of the game.

T

let us remember this is a football sim in which we make players and teams. the stats should conform to what those players and teams do, not the other way around.

Originally posted by im317
let us remember this is a football sim in which we make players and teams. the stats should conform to what those players and teams do, not the other way around.

That statement is only true if the sim or engine is perfect. It's not, so lack of stats is the best indicator of a problem that we have.

Originally posted by Signalsgt
Originally posted by im317

let us remember this is a football sim in which we make players and teams. the stats should conform to what those players and teams do, not the other way around.

That statement is only true if the sim or engine is perfect. It's not, so lack of stats is the best indicator of a problem that we have.

i would argue that we need to look at how those stats or lack there of are being generated from the player interaction to best determine were the issues in the sim are.

Originally posted by tsherr
Bukowski is right. Stats in the NFL are created by the physical and mental abilities of the athletes working under the constraints of the laws of physics and the rules of the game. So in order to simulate that in GLB you model the laws of physics and the rules of the game in the simulation and then let the builds determine the stats. If a DE can get a sack on every passing sack, then that's a good build against a bad build. If every DE gets a sack on every passing down, then that's bad modeling and you change the sim. Deciding before hand that DEs should get X sacks per season means that the build and the sim become secondary, and if that's the case, then there is no point to play the game.

T

Abso-damn-lutely.

Here's the testing that needs to be done concerning balance

---------------------------|__Speed OT__|__Strength OT__|__Balanced OT__|
Speed Rusher-------|Sack number_|__Sack Number_|__Sack Number_|
Strength Rusher----|Sack number_|__Sack Number_|__Sack Number_|
Balance Rusher-----|Sack number_|__Sack Number_|__Sack Number_|


There needs to be a zeroing out (or damn near) against the same types. with some success against the opposite type, and minor sucess against balanced types. The grid showing sacks a game should look like this

|0-1|2-4|1-2|
|2-4|0-1|1-2|
|1-2|1-2|0-1|